EEVblog Electronics Community Forum
Electronics => Beginners => Topic started by: NeverDie on May 11, 2020, 05:47:57 pm
-
Hi,
The advertisement for uCurrent Gold says:
"Measure the standby and sleep current of the latest generation of microcontroller and other digital electronics, energy harvesting devices etc, down to the nanoamp (or even picoamp!) level with superb accuracy."
https://www.kickstarter.com/projects/eevblog/current-gold-precision-multimeter-current-adapter (https://www.kickstarter.com/projects/eevblog/current-gold-precision-multimeter-current-adapter)
Great!
Question #0: What exactly must I do to get the advertised "superbly accurate" picoamp measurements?
I've tried using the uCurrent Gold in conjunction with a 500,000 count DMM, and I'm not getting superbly accurate picoamp measurements.
Background:
I was hoping to use the uCurrent Gold to measure picoamps for some energy harvesting circuitry that's under development: https://forum.mysensors.org/topic/10812/the-harvester-ultimate-power-supply-for-the-raybeacon-dk/190 (https://forum.mysensors.org/topic/10812/the-harvester-ultimate-power-supply-for-the-raybeacon-dk/190)
I recently acquired a 500,000 count DMM (an Amprobe AM-160-A) to complement the uCurrent Gold so as to accomplish picoamp measurements.
However, even after taking anti-static precautions, which seem to have the millivolt readings on my DMM under control, the uCurrent Gold measurements below 1 nanoamp are all over the map. I've grounded my body and have a grounding mat underneath the DUT, the uCurrent Gold, and the DMM and test leads. However, when I move my hands and arms anywhere near it, the picoamp measurements shift around (but stop when I stop moving). The problem is: every time I move, the measurements settle on a different value, so there's no repeatability.
Question #1: Have I diagnosed the problem correctly? Does this phenomena, as I've described it, sound as though static charge is the problem, or is it likely to be something else? Unfortunately, to get the right offset, I have to first measure with uCurrent Gold in shorted mode, and then I have to physically switch the uCurrent Gold to measurement mode, so there's no way I have avoid either touching it or moving near it.
Question #2: Is nanoamp accuracy about as good as I can expect for taking manual measurements using the uCurrent Gold, or would additional anti-static mitigation allow me to get clean, repeatable, manual measurements down to 1 picoamp? If picoamp accuracy is feasible with the uCurrent Gold, what are the minimum prerequisites required to attain it? I need to know so that I can decide whether to spend more on upgraded anti-static measures (if it will work) or else improvise some instrumentation so that I can take the uCurrent Gold picoamp measurements wirelessly, so that I (and my static charge) are nowhere nearby when the picoamp measurements are taken. Would that work, or would even that not be enough? Is there something else I should do/try first?
-
With all the static precautions you're taking, it sounds more like a noisy environment issue (EMI-wise).
Do you have a mobile phone/wi-fi/LED light fittings/SMPS anwhere near the test setup?
-
With all the static precautions you're taking, it sounds more like a noisy environment issue (EMI-wise).
Do you have a mobile phone/wi-fi/LED light fittings/SMPS anwhere near the test setup?
Nothing nearby. No SMPS. I'm using just a battery for power.
-
It seems that this could be RF related.
Therefore: a wild, wild, wild idea:
So how about AC grounding yourself?
The normal grounding straps have a 1 Meg resistor in series for safety purposes. How about you put a capacitor in parallel with said resistor? The capacitor value should be calculated to provide a reactance of no less than 100k at your power line frequency for safety purposes.
Otherwise you’ll perhaps require a small Faraday cage.
-
Is anyone here able to get superbly accurate, repeatable picoamp measurements using their uCurrent Gold? Has anyone here besides me even tried? It would help my troubleshooting to know if the problem is unique to me or whether others here are either having no problems at all or else either are experiencing or have experienced the phenomena I've described.
I'm having no problems getting repeatable measurements down to about 1 nanoamp of current. It's below that current level where it gets problematic.
-
First, measuring at the picoamp level is difficult - even if you have an electrometer which is specifically designed for the job, you have to be careful about what you do - shielding your experiment, using triaxial cables, etc. etc., to get reliable results.
I like the uCurrent Gold and have used it to get ridiculously sensitive and repeatable measurements in the picoamp range, but I used it in an AC bridge to do it - not DC.
By using AC to test, you can filter everything (noise) out of the result except the exact frequency of your test current (for example, 400Hz or 1KHz). All worries like thermal voltages at wire junctions etc. disappear when you use AC. (A bunch of new worries crop up, of course, they always do.)
I might try to see how low I can get with DC some time this week, just for fun, so we have something to compare.
-
Thanks. So, if it's a given that shielding is required, which from your post it sounds like it is, then I guess my plan will be to configure an arduino to autonomously do the measurements as well as log them. Once that's working I'll have to put the whole thing inside a sealed metal can while the measurements are taken. If that doesn't solve the issue, then I don't know what will.
I guess the cleanest way would be to use a latching relay to switch between the uCurrent's Shorted Mode (to get the offset) and the uCurrent's Masurement Mode. That way the arduino can let the relay settle after switching and between doing the measurements without creating possible interference during the measurement itself.
I'll try using a 24 bit ADC to read the uCurrent Gold's millivolt output. The ADC I have on hand is this one: https://www.amazon.com/gp/product/B071P16YCS/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1 (https://www.amazon.com/gp/product/B071P16YCS/ref=ppx_yo_dt_b_search_asin_title?ie=UTF8&psc=1)
Hopefully it will be at least as good (?) for that purpose as my DMM would have been. ::)
I'd run the arduino directly from a battery with no LDO. I'd rather not hack the ADC, but I suppose if it proves necessary I could make it run directly from either a battery or a supercap. However, I'm guessing the ADC probably generates its own reference voltage, and that just is what it is. All I need are two measurements, plus redoing them a few times to see if the measurements are repeatable or whether they vary.
Does this plan sound like it should work? Does anyone foresee any problems with this approach?
-
I had a play with it using DC and found things got very noisy below about 10nA, meaning that I got to about the same limit as you - 1nA just about discernible, but not really a satisfying high quality measurement. We probably have to face the fact that the uCurrent is not designed to be a picoammeter and we are not going to be able to make it 1000x better by shielding or anything else.
I suspect that getting significantly further will require something like an electrometer op amp circuit (there are plenty of threads here on the EEVblog). You might be able to build that inside a shielded box together with your experiment, battery powered, and capture the output with your ADC.
-
I have a sinking feeling that you're probably right.
I found this: https://www.eevblog.com/forum/projects/picoammeter-design/150/ (https://www.eevblog.com/forum/projects/picoammeter-design/150/)
However, just throwing something together in an altoid can... not sure about that. I suspect board layout will be important. For instance, TI says, "a "Guard" trace is recommended when designing sub-nanoamp systems." Even the author of a DIY picoammeter admits that a proper PCB is the way to go: http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf (http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf)
There's a for-profit fully assembled PCB made by a UK company that I suppose is an option. It got a nice review here: http://physicsopenlab.org/2019/09/23/picoammeter/ (http://physicsopenlab.org/2019/09/23/picoammeter/)
It sells for about $150 delivered to the US, but the shipping from the UK takes 3 weeks.
However, I see no way to hook it up to an oscilliscope. If so, that would severely limit its usefulness.
It doesn't make sense for us all to be reinventing the wheel. If there exist gerber files for some open source PCB design which had already been vetted... but I'm not finding it.
-
I just ran across this video, which uses the uCurrent Gold and essentially the same OEM 500,000 count multimeter as what I have:
https://www.youtube.com/watch?v=xY10nkN97vw (https://www.youtube.com/watch?v=xY10nkN97vw)
At time index 9:35 he's waving his hands near the multimeter and uCurrent Gold, and you can see the numbers jump around. In my case they jump around more than his, but apparently it's par for the course when attempting to measure picoamps.
-
Electrometer circuits are actually very simple, with only a few components. For a home lab project they can be made using "dead bug construction" techniques in order to reduce leakage as much as possible, better than pretty much any PCB board. To calibrate the circuit, maybe a suitably large resistor and a voltage divider to drive it... you can probably do some "close enough" work with standard components.
E.g. https://www.eevblog.com/forum/projects/picoammeter-design/msg790045/#msg790045 (https://www.eevblog.com/forum/projects/picoammeter-design/msg790045/#msg790045)
-
Which would be better? His design or TI's LMP7721 eval board (I posted a link to the gerber files here: https://www.eevblog.com/forum/projects/picoammeter-design/msg3065852/#msg3065852 (https://www.eevblog.com/forum/projects/picoammeter-design/msg3065852/#msg3065852) )
-
The LMC662 has very similar typical bias current (around 3fA) as the LMP7721, is available in dip package (LMP is SO8 only) and is much cheaper. The 662 has a wider Vos spec. but for a one-off, it's easy to trim out, especially as it is a dual.
Chris (him)
P.S. The 662 also has significantly lower input referred current noise.
-
Cool.
So, it looks as though a way to minimize external influences is to put the picoammeter part in its own sealed case and then ground that to an old computer case, which, has the DUT and test leads, such as here (with the lid off):
(https://d3i71xaburhd42.cloudfront.net/c97ebfd01a1be99c876b938bf5bd8e102b401e82/1-Figure1-1.png)
https://www.semanticscholar.org/paper/Low-cost-picoammeter-for-dielectrics-Epure-Belea/c97ebfd01a1be99c876b938bf5bd8e102b401e82 (https://www.semanticscholar.org/paper/Low-cost-picoammeter-for-dielectrics-Epure-Belea/c97ebfd01a1be99c876b938bf5bd8e102b401e82)
Is that sufficient, or should the whole shebang also sit on a grounded anti-static mat?
Any particular specs or ratings to look for in an anti-static mat? I have a cheap one, but it doesn't have the conductive rubber on the back side of it that the more expensive ones have.
-
As long as you have the whole setup in a Faraday shielding metal enclosure, it doesn't really matter what you have underneath it. Just making sure that you ground yourself to the case using a proper ESD strap while making adjustments to anything ESD susceptible (soldering iron tip too if relevant).
-
As long as you have the whole setup in a Faraday shielding metal enclosure, it doesn't really matter what you have underneath it. Just making sure that you ground yourself to the case using a proper ESD strap while making adjustments to anything ESD susceptible (soldering iron tip too if relevant).
Looking at the circuit diagram in the other thread, I'm not sure I understand the location of the 1M resistor at the input... is there a "typo", should it be on the other side of the junction with the 1G resistor, to form an inverting op amp circuit with amplification 1,000x ? Or is this just me not getting how it works...
Looking at the pictures, it seems it is me that doesn't get how it works. - I suppose the idea is that with a really small current flowing into the input terminal, the op amp will balance that current by flowing an equal current backwards through the 1G resistor - so it is actually a kind of inverting amplifier, but without an input resistor as you'd expect with a normal voltage driven inverting op amp?
I should really build one too...
-
If you mean the one on my picoammeter, the 1M is purely there as an ESD protection resistor (limiting the current at high over-voltages to within the capability of the LM662's internal bootstrapped protection diodes). The circuit operates as a transimpedance amplifier, using the 1G resistor as the feedback to the input (giving 1mV/pA), and keeping the input at 0V.
Edit: Good idea, every home should have one. :)
-
Which parts do you recommend for these?
The reason I ask is that the guy who wrote http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf (http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf) said:
Be really careful. These
diodes can’t be generic because their
reverse current must remain far lower
than the measuring range
Of course, he was referring to the non-zener diodes in his circuit, not the zener diodes in your circuit, and he wires his diodes differently than you do yours, but... Did you pick zener diodes and wire them the way you did because that way they have lower leakage?
The guy in the video above picked a jfet for that purpose and said he's getting single digit picoamp reverse leakage. Is that better or worse than the leakage from the two zener diodes?
-
For anyone who's interested, the moving hand problem described in the OP and shown in the video is a symptom of "electrostatic coupling," as explained here:
If the interference, say power line noise, increases as you move your hand close to the circuit, and reduces if you touch ground, then you are suffering from electrostatic coupling.
The most common effect of electrostatic coupling is the “proximity” effect, or the ability for a sensitive circuit to “see” your hand moving from centimeters away, or the “antenna” effect, where the circuit can “sense” another signal (power line noise) from a distance.
https://www.edn.com/design-femtoampere-circuits-with-low-leakage-part-2-component-selection/ (https://www.edn.com/design-femtoampere-circuits-with-low-leakage-part-2-component-selection/)
It goes on to say that the solution is to surround the input stage with conductive shielding.
-
Yes shielding is vital for such sensitive measurements.
Your body moving around it can electrostaticaly attract electrons in your test setup and these electrons create a current that gets measured. So measurements like this are typically done in shielded boxes (Large diecast metal enclosures are quite popular).
The shielding doesn't need to be any anything fancy (Like RF tight) Just covering the thing with aluminum foil connected to ground will do since it brings everything around it to a fixed voltage potential. Tho if you do have a lot of RF it might need more, for example that USB microscope could be spewing out some RF, so perhaps poke a small hole in your shield for the microscope to look trough.
-
FWIW, the lowLevelHandbook recommends the attached. For D1 and D2 it suggestions using 1N3595. Unfortunately, it didn't suggest any particular zener diodes.
-
Which parts do you recommend for these?
The reason I ask is that the guy who wrote http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf (http://circuitcellar.com/wp-content/uploads/2014/03/CC_042010_Lacoste_ReprintedwPermission-web.pdf) said:
Be really careful. These
diodes can’t be generic because their
reverse current must remain far lower
than the measuring range
Of course, he was referring to the non-zener diodes in his circuit, not the zener diodes in your circuit, and he wires his diodes differently than you do yours, but... Did you pick zener diodes and wire them the way you did because that way they have lower leakage?
The guy in the video above picked a jfet for that purpose and said he's getting single digit picoamp reverse leakage. Is that better or worse than the leakage from the two zener diodes?
Those two zener diodes on my circuit are on the output terminals. They were an afterthought (probably totally unnecessary) to provide a bit of ESD protection on the output leads.
The reason that they're air wired on the terminals? They were an afterthought. :P
-
:palm: Gestalt error on my part. :-[
Thanks, Gyro! (aka his/him) :-DD
I'll order the parts (and a few 1N3595 "for good measure") and build it.
BTW, I found a picoammeter circuit in the datasheet for the CA3420: https://www.mouser.com/datasheet/2/698/ca3420-1528689.pdf (https://www.mouser.com/datasheet/2/698/ca3420-1528689.pdf)
What do you think about having a range switch? Any advantages to that, or not?
-
BTW, I found a picoammeter circuit in the datasheet for the CA3420: https://www.mouser.com/datasheet/2/698/ca3420-1528689.pdf (https://www.mouser.com/datasheet/2/698/ca3420-1528689.pdf)
What do you think about having a range switch? Any advantages to that, or not?
A 1G resistor gives a maximum range of at least +/-2.5nA at 1mV/pA, assuming reasonable utilisation of the 9V battery discharge life. This was good enough for me, there are other ways of measuring greater than 1nA currents. I have a 1uV resolution bench meter, so I have no need for greater sensitivity.
It really depends how sensitive you want to go.
-
Makes sense. I ordered an aluminum enclosure for it. That way I can turn it on-off with a reed switch on the inside and a magnet on the outside.
-
The LMC662 has very similar typical bias current (around 3fA) as the LMP7721, is available in dip package (LMP is SO8 only) and is much cheaper. The 662 has a wider Vos spec. but for a one-off, it's easy to trim out, especially as it is a dual.
Chris (him)
P.S. The 662 also has significantly lower input referred current noise.
Can you explain how the offset trim works in your circuit, using the second op amp?
-
Can you explain how the offset trim works in your circuit, using the second op amp?
+1
-
Can you explain how the offset trim works in your circuit, using the second op amp?
+1
It looks as if I'd better copy the schematic over from the Picoammeter thread to supply some context here...
(https://www.eevblog.com/forum/projects/picoammeter-design/?action=dlattach;attach=179209;image)
The offset adjustment is really quite simple (once you get it). The resistor network between +B and -B (100k, 15R 15R, 100k) is used to establish the mid rail point for the rail splitter, and for the offset adjustment. The two 15R resistors in the middle provide a voltage drop of a few mV (I'll let you do the sums). The 100k pot across this 30R span allows the main opamp's ground reference (In+) to be shifted slightly relative to mid rail, to compensate it's offset voltage. Because the span is so small, it can use a single turn preset and still get decent zero adjustment. The second opamp (the rail splitter) simply takes it's mid-rail reference from the junction of the two 15R resistors. The two 100n capacitors drop the AC impedance of the midpoint to keep noise down.
Remember that in this case, the rail splitter is driving the case, so it effectively floats the battery to be +/- 4.5V relative to the case. Because there is a reasonable amount of capacitance involved (battery casing to case, board to case, switch wiring etc.) it needs some damping to prevent oscillation - hence the 10n/1M feedback network and the 100R on the output.
P.S. the two 1k resistors in series with the output terminals are there for stability purposes too - to isolate the main opamp and rails splitter from stray capacitance on the DMM and its leads (and a bit of protection).
-
Today I received all the parts in Gyro's picoammeter BOM, so I hope to put the picoammeter together soon. I expect it will look similar to Gyro's but with some small differences. I'll post photos when I finish it.
Meanwhile, I have a related question: since the DUT and picoammeter and all related measurement equipment is supposed to be enclosed together under a conductive shield prior to measurements being taken, how is it that I can use the picoammeter in conjunction with my Rigol 1054Z oscilliscope? I mean, if the Rigol is under the conductive shield too (?), then how do I see what's on the Rigol's screen? Do I need to connect my PC to the Rigol over a fiberoptic cable or something? Or, must I configure the equipment to broadcast measurement data to me via soundwaves through the metal shielding? Or, do I hire a psychic to remote view the oscilliscope? :-DD Or, ....? Seriously, how should it be done?
"In for an inch, in for a mile."
-
Looking forward to seeing another implementation. Remember cleanliness!
You can treat the output connectors of the picoammeter as the external interface of the shielded enclosure (you certainly don't want a noisy Rigol scope inside! :)).
Measuring single digit picoamps isn't really scope territory, well I guess it can be at low timebase speeds, but remember that the picoammeter circuit has a bandwidth of about 0.5Hz (due to the 330pF feedback capacitor). Assuming that you have the conductive shield referenced to mains ground (reasonable), then you can connect the scope probe ground clip to the output +ve terminal (which is effectively connected to the picoammeter case and shielded enclosure). Connect the probe tip to -ve output terminal.
The scope will read increasingly negative voltage with increasing (+ve) current - set the scope channel to 'invert' if this bothers you.
P.S. It's probably about time for you to download a copy of the Tek (formerly Keithley) Low Level Measurements Handbook... https://www.tek.com/document/handbook/low-level-measurements-handbook (https://www.tek.com/document/handbook/low-level-measurements-handbook)
-
Picoamp measurements can take a while to settle down to a steady reading.
Overall, you are probably better off with a DMM for this measurement.
On the scope, you could try its Chart Roll mode (if the Rigol has that?), where it basically acts like a chart recorder, taking maybe a minute to trace across the screen. That might be kind of cool, you'd certainly see when the measurement was settled!
-
Picoamp measurements can take a while to settle down to a steady reading.
Even when the DUT is completely shielded? If so, is there anything that can be done in addition to the shielding to reduce/eliminate the settle-down time without adversely affecting accuracy?
-
Picoamp measurements can take a while to settle down to a steady reading.
Even when the DUT is completely shielded? If so, is there anything that can be done in addition to the shielding to reduce/eliminate the settle-down time without adversely affecting accuracy?
Once the system has settled, I would expect that small changes in the signal would not take long to propagate down the line. But if there is a large step in the signal, it takes time to settle to the new value.
-
I'm nearly done putting it together. Just waiting for a missing part to arrive, hopefully tomorrow.
the picoammeter circuit has a bandwidth of about 0.5Hz (due to the 330pF feedback capacitor).andbook/low-level-measurements-handbook]https://www.tek.com/document/handbook/low-level-measurements-handbook[/url]
Sounds like may I'll need to build something else for the oscilliscope. Are there any schematics you can recommend?
-
How fast is the signal you are looking at?
-
You can reduce the value of the 330pF capacitor. Bandwidth will increase, but so will the noise level. Don't expect miracles though.
-
How fast is the signal you are looking at?
... and what current range?
-
Not the best example,
(https://forum.mysensors.org/assets/uploads/files/1588953153125-cascade.png)
but maybe being able to use the picoammeter to show the AM3 current flow on a oscilliscope? This is a SPICE simulation on TI TINA. At the right it shows a virtual oscilliscope with voltages and currents, and so it provides a target of sorts for the picoammeter (if connected to an oscilliscope) to match up against.
Edit1: If the current picoammeter can do this, then great. But if the frequency were greater....?
-
Remember that a Picoammeter measures current to ground (or at least case for the battery operated one). You might get away with carefully insulating the case and portable DMM from ground using a decent insulating sheet (PTFE or Polystyrene). It doesn't really tie in with the screened enclosure though.
As soon as you attach a scope, you are definitely talking about a to-ground measurement. A Picoammeter doesn't work as a differential current probe.
-
JFYI a cheap DMM with 10M input impedance in mV mode works remarkably well as a micro|nano-ammeter. I read about this advice on this forum, I tried it, and I got much less noisy results than with uCurrent Gold. With 0.01mV resolution it maps into 1pA current resolution. Of course, in reality it won't measure current of a few pA, nor it will be very accurate because input impedance is rarely exactly 10M, but still very capable approach.
-
@exe
You mean like this? https://hackaday.com/2016/02/01/need-a-nano-ammeter-you-already-have-one/
-
@exe
You mean like this? https://hackaday.com/2016/02/01/need-a-nano-ammeter-you-already-have-one/
Yes. I used aneg 8002 or 8008 for my measurements.
-
Not the best example,
(https://forum.mysensors.org/assets/uploads/files/1588953153125-cascade.png)
but maybe being able to use the picoammeter to show the AM3 current flow on a oscilliscope? This is a SPICE simulation on TI TINA. At the right it shows a virtual oscilliscope with voltages and currents, and so it provides a target of sorts for the picoammeter (if connected to an oscilliscope) to match up against.
Edit1: If the current picoammeter can do this, then great. But if the frequency were greater....?
The frequency looks quite low, like about 1Hz? How high would you like to take the frequency?
-
I'd be happy with this frequency, as long as it can show the waveform reasonably well. I guess I'll know soon enough because...
I finished the assembly! Here it is with the lid and back removed so that you can have a good look:
[attach=1]
I air-wired most of it. The white pillars are teflon. I purchased a 3/8" teflon rod 12 inches long and then sawed it into short stand-offs. Loctite makes a superglue plastic adhesion promoter that allows even teflon to be glued, and that's what I used. The battery, the offset-voltage module, and the main module sit on a 1/8" thick FR-4 using nylon standoffs. Aside from the teflon standoff that's supporting the op-amp, the other teflon standoffs are just there to support the wiring so that the legs on the op-amp don't get strained. At the top of the tallest teflon standoff is a reed-switch for powering everything either on or off via a magnet on the outside of the aluminum casing.
[attach=2]
The whole thing is more or less floating and, so far, not grounded to the case:
[attach=3]
If it turns out to be needed, I can replace the nylon stand-offs with teflon stand-offs, since teflon offers a much higher resistance than nylon. It seems the overarching goal is suppressing leakage as much as possible, since in the picoamp realm even small leaks might look big.
-
Another view:
[attach=1]
And here it is all buttoned up:
[attach=2]
I won't be able to try it out until tomorrow, but I'm looking forward to it. :)
One improvement I might make is drilling a hole in the lid above the trim pot so that I can fit in a screwdriver to zero it without having to take off the lid. If I were to do that, though, I'm not sure what kind of removeable conductor I could seal the hole with when I'm not trimming the pot. Thinking... Maybe the best solution would be to tap some threads into the hole and then plug it with a metal bolt when not in use? That way the integrity of the conductive enclosure shield would be maintained. If I were to pursue this, I should maybe (?) also elevate the trim pot so that it's higher up and closer to the lid, so that it would be easier to fit a screwdriver into its screw slot of the trim pot. Well, I'll see how it goes with just the hole first.
-
Looks awesome! 8)
- will be interesting to see the results of your tests!
-
Oh Wow, that's a lot of Teflon, you had some fun there! ;D
I see you've used a reed switch for the power too.
A few points I can see:
- You seem to have the protection resistor directly in series with the input terminal, rather than between the 1G feedback resistor and the opamp input / capacitor junction. This will raise the input resistance of the circuits to 10M ohm, rather than acting as a virtual ground (the input should try to stay at ground potential for maximum accuracy). The protection resistor should be inside the feedback loop for maximum accuracy.
- I can't see where the case is grounded to the Green ground terminal, to form a proper screen.
- Ideally, you would air-wire the junctions between the op-amp input, and the input end of the 1G feedback resistor, the 10M protection resistor and the polystyrene cap through to the input terminal. Teflon (PTFE) is good, air is better.
- The most critical leakage path (weakest link) is on the LMC662 package. This needs to be as clean as possible. I think I see some flux residues, not critical as long as the package itself is clean (IPA is best) between the pins.
The great, and much missed, Bob Pease wrote some interesting stuff on the use of Teflon in high impedance circuits, charge storage etc....
https://www.electronicdesign.com/technologies/test-measurement/article/21773611/whats-all-this-teflon-stuff-anyhow (https://www.electronicdesign.com/technologies/test-measurement/article/21773611/whats-all-this-teflon-stuff-anyhow)
and one of his fun videos...
https://www.youtube.com/watch?v=B4G3YPlO6Wg (https://www.youtube.com/watch?v=B4G3YPlO6Wg)
It will be good to see your results. Note that it will probably take a while to settle - leave it with the lid off in a warm dry place, superglue vapours can settle on surfaces (the fingerprint trick in Beverly Hills Cop 2).
P.S. A small hole for trimming the pot won't make any difference to the screening - you can always put an adhesive label over it to keep dust out.
-
Thanks! I'll make the changes you suggest. I'm not sure how conductive superglue is, but I did superglue the opamp to the teflon mount. Based on what you said just now about the op-amp package being the most critical leakage path, and for it needing to be clean, that may have been a fatal mistake on my part. I have some spares, so maybe I should replace the op-amp with a fresh one just to be sure.
I did end-up drilling a hole above the trim-pot. I also installed a green LED near the trim-pot which lights whenever the reed switch turns on the system. I can see it through the same hole. The LED draws just 10 microamps, and I like having visual confirmation that the system is turned-on. This turned out to be helpful already, because I hadn't anticipated that the magnet, when left alone, would be dragged toward the 9-volt battery (the battery case is steel). It turns out that shift of the magnet's position was enough to turn-off the system. For now I'm taping the magnet in position once I get the green light. Maybe later I'll move the reed switch closer to the battery.
-
Good question, I did some searching on the web but couldn't find a conductivity figure for superglue (Cyanoacrylate). I know that it needs moisture to set, but don't know how that is bound up. To remove superglue or its deposits you need something more aggressive like Acetone, IPA won't touch it.
If you want the lowest leakage then maybe replace the opamp and make it completely air wired (you seem to have enough supported wires around it), but you could just wait and see how it performs anyway.
To correct the circuit location of the 10M resistor, I would simply unsolder the end of the 1G feedback resistor, extend it with a bit of bare wire and solder directly to the input connector. The 10M resistor lead is sharing a pillar with the yellow PVC insulated wire, but it would probably be a bit anal to suggest lifting that off, they should be at the same potential anyway. The rest of the circuit nodes are all low impedance.
A power indicator is a good idea, as long as you make sure it isn't taking more current than the rest of the circuit (battery-life wise).
P.S. Don't forget to ground the case - it may look like a Faraday shield, but only until you make a connection to one of the terminals, then it becomes a big antenna for noise pick-up.
-
Good catch! I see now that I did incorrectly wire the gigaohm resistor. Maybe that's the reason why the picoammeter is totally not working right now.
Uh, you've twice referred to a 10megaohm resistor, but, unless I'm overlooking something, in your schematic it's 1 Megaohm. Based on context, I'm assuming that's what you meant.
The yellow wire isn't actually sharing a standoff with the megaohm resistor. Maybe it looks differently in the picture. However, there is a slight, unintentional glue bridge between the black wire and the blue wire's support offset, so I'll clear that away during the repair.
In case you're interested, the green LED I'm using is a CRE, model 630-HLMP-CM1G-350DD. Costs about 10 cents each in quantity 10 from mouser. I had previously tested it as being dimly visible even at just a 500na current drain, provided you can look at it dead-on (which is when it's narrow optics helps some). Here I'm running it at 10ua because the orientation isn't so great. I had earlier thought of having the LED do double-duty by also illuminating the trim-pot, as a navigation aid for screwdriver insertion, but it turned out the hole is big enough and the trim-pot is high enough that room light alone is enough to see where the screwdriver needs to go.
Lest I lead someone down the wrong path, I should mention that I'm not thrilled with the aluminum enclosure I'm using. I purchased it from amazon, and in pictures it looked like plain aluminum, but in reality it has some kind of non-conductive silvery coating on it. Therefore, as belt and suspenders, I'm going to tap each of the four pieces that constitute the shell of the enclosure and wire them together, similar to what an electrician might do, to guarantee that they really are all electrically connected. The drill-tap for doing that should arrive tomorrow. If I were to do the project over from scratch, I'd probably use a die cast aluminum box instead of this coated semi extruded aluminum box.
-
Damn, yes sorry. I meant 1 Meg. I sized it to limit the current through the LMC662 input protection diodes at reasonable overload.
I selected a 1W Carbon film part for its large body size and 500V rating (hence the 500V warning on my label).
P.S. Yes it's scary how efficient LEDs can be these days, particularly deep green and blue.
-
Which solder flux do you recommend for this situation? I presume a flux that's easy to remove afterward rather than a "no clean" that you simply leave on?
Edit: Or maybe no flux at all? i.e. just solid core solder with no flux?
-
Use ordinary 60/40 or 63/37 with rosin based flux. No clean flux definitely has no place in a high impedance situation. You don't need to remove the rosin flux if it isn't near the package.
If you solder to the tips of the pins, then flux contamination of the package is going to be minimal anyway. It's only really critical for the high impedance input pin.
Don't get IPA on the Polystyrene capacitor (I found that its leakage eventually recovers to normal, but it takes a couple of days).
-
Don't get IPA on the Polystyrene capacitor (I found that its leakage eventually recovers to normal, but it takes a couple of days).
Thanks to your warning I used acetone to clean off the flux in that area and thereby, hopefully, dodged this complication.
I finished all the necessary repairs and even made a few improvements.
Firstly, I realized there was an easier way to electrically internconnect the four pieces of the shell: by grinding off the coating on the end-plates where the screws are used to attach them to the upper and lower pieces of the box. Simple. Here's the before and after on the grinding:
[attach=1]
[attach=2]
In turn, that allowed me to do a quick (temporary) hack to electrically connect the enclosure to input-GND:
[attach=3]
After I received my drill-taps (hopefully today), I'll accomplish essentially the same enclosure to input-GND connection, but in a way that's inconspicuous and with a nicer aesthetic.
Here's the hole in the case I drilled so that I can slide a screwdriver into the trim-pot to zero the circuit prior to measurements:
[attach=4]
If you were to zoom in on the drilled hole shown in the photo, you can see the trim-pot. :)
-
On the inside I
(1) corrected the placement of the gigaohm resistor.
(2) removed the solder flux on the critical input pin of the opamp.
(3) re-positioned the LED so that it would shine up, out of the drilled hole, thereby leveraging the narrow optics of the LED and making it easier to see when the enclosure is sealed up,
(4) Replaced the el-cheapo reed switch with a few of the same el-cheapo's in parallel. I aligned the reed switches randomly, so that there'd be less need for a particular orientation of the magnet when switching on the circuit. What happened instead was a surprise: it appears that when configured in this parallel way, the reed switches act as a latch--they turn-on when excited by a magnet and stay turned on even after the magnet has been removed. I don't understand why. Maybe they're oscillating instead of simply turning on? I don't know. Fortunately, if I wipe the magnet in a particular direction over the reed switches, the latching stops and the system turns off.
[attach=1]
[attach=2]
[attach=3]
-
Seems to be working. I took a 3.3v voltage source, passed it through a 10 gigaohm resistor and then into the picoammeter and then to ground. Hooking up my DMM to the picoammeter and setting it into millivolt mode, the number shown was about 300. i.e. 300 picoamps, which is a plausible value given the 25% tolerance of the 10G resistor. 8)
What's a better test to measure in more detail the accuracy of the particular picoammeter that I built? Is there a particular benchmark test that "everybody" uses?
-
Seems to be working. I took a 3.3v voltage source, passed it through a 10 gigaohm resistor and then into the picoammeter and then to ground. Hooking up my DMM to the picoammeter and setting it into millivolt mode, the number shown was about 300. i.e. 300 picoamps, which is a plausible value given the 25% tolerance of the 10G resistor. 8)
What's a better test to measure in more detail the accuracy of the particular picoammeter that I built? Is there a particular benchmark test that "everybody" uses?
Nice one!
To calibrate the device, maybe make a much lower source voltage (10mV ?) out of a couple of resistors, so you can use smaller and more precise resistors instead of the 10G one?
-
I was thinking along similar lines except using a very accurately calibrated millivolt voltage source and then coupling that with some affordable ultra-precise resistors. I think (?) there may be voltage reference chips/modules that someone with a fancy NIST traceable 8 digit DMM then certifies as having a particular voltage, which he then writes onto the module.
However, your idea has merit as well. Maybe yours would be more cost effective. Not sure: there must be a reason why people prefer having/using a calibration voltage module over only using just their Fluke 87V and some precision resistors to create a voltage divider.
Edit1: I remember now that mjlorton did quite a number of youtube videos on the topic of precision voltage sources. That guy seems genuinely obsessed with accurately measuring voltage. I guess maybe I should go see what he recommends,... unless one of you guys knows of a good reference module to recommend for this picoammeter scenario?
-
Because your device uses so little current, you can get away with a voltage divider as the source. For example, bleeding 100 picoamps off a current of 10mA flowing in a divider is very little error.
The biggest sources of error is going to be the precision of the resistors. - But you can use whatever reasonable resistors you have, and just measure the voltage of the voltage divider you make of them with your multimeter.
Then you just calculate how much current should be flowing into your picoammeter based on the measured voltage and the size of the resistor going to the Picoammeter (including its input resistance), using Ohm's Law.
-
Seems to be working. I took a 3.3v voltage source, passed it through a 10 gigaohm resistor and then into the picoammeter and then to ground. Hooking up my DMM to the picoammeter and setting it into millivolt mode, the number shown was about 300. i.e. 300 picoamps, which is a plausible value given the 25% tolerance of the 10G resistor. 8)
What's a better test to measure in more detail the accuracy of the particular picoammeter that I built? Is there a particular benchmark test that "everybody" uses?
Looking good. :-+
Yes, a resistor and a voltage source is the normal way to check the accuracy. The picoammeter feedback maintains the input at virtual ground, so there shouldn't be any significant voltage burden at low source voltages (as long as you don't go too low). Picoammeters, in the form of SMUs (Source Measurement Units) are often used to measure very high resistances at high voltage - hence the generous overload voltage rating I put on mine.
I'm not sure if I already mentioned it in this thread (it's in the Picoammeter thread), but the way to adjust the zero pot for the opamp offset voltage is to link the input to the negative output (not ground). The mV reading you get after removing the link is then the input bias current of the picoammeter itself.
-
I did some more tests with the uCurrent this morning. Using A/C at 400Hz, I was able to measure 600pA with reasonable accuracy. I was able to "see" 60pA, but the accuracy was out the window.
https://www.eevblog.com/forum/testgear/ucurrent-minimum-measurable-current/msg3084525/#msg3084525 (https://www.eevblog.com/forum/testgear/ucurrent-minimum-measurable-current/msg3084525/#msg3084525)
-
That's interesting.
From just a quick and dirty experiment that I did this morning, I'm getting a great result with Gyro's picoameter.
I put together a simple voltage divider consisting of two resistors: 10ohm and 10kohm. So, a 1000x voltage reduction there.
To make the math easy to check and follow along, I'll express the figures using power of ten exponents. i.e. a (10^-3) voltage reduction.
Then, I put a 1Mohm (10^6) resistor at the divider junction to feed the picoameter.
I fed the voltage divider with 100milliohm (10^-1) DC voltage.
Hence, the voltage at the resistor divider should be (10^-4).
So, the current entering the picoammeter, by ohm's law, should be (10^-4)/(10^6) = (10^-10) = 100 picoammeters.
I turned it on and then walked out of the room to view the voltmeter from the doorway. The last digit bounces around, but with some eyeball averaging it looks to read about 94 millivolts, which corresponds to 94 picoamps.
I'm not at all bothered that it's not on the nose, because this was just a quick and dirty test with 1% tolerance resistors, with long wires everywhere, a number of Chinese alligator clip connections, and no meaningful static shielding external to the picoammeter. And, if anything, I'd expect a slightly lower number anyway from some leakage through the superglue that I mounted the opamp with, so the results are consistent with that possibility as well.
One test is hardly definitive, but I'm impressed! :clap:
I guess the next step would be to put something a little more rigorous together and then stick the whole enchilada inside a shielded enclosure, similar to one of the earlier pictures in this thread.
-
@Neverdie: Once you've screened it etc. if you find that it consistently reads slightly low, you can tweak it by adding a resistor in series with the 1G one.
If the were to read high, then you really wouldn't have any options but to swap the 1G resistor and cross your fingers that you get a higher value one, but with a slightly low reading you have padding opportunity. If it reads 6% low, then something like a 56M in series would probably get you very close.
-
I tried hooking up the picoammeter to my Rigol oscilliscope. Unfortunately, the scope image is too polluted by 60hz artifacts to be useful, so I guess I either need better shielded probes, a differential probe, or a battery powered oscilloscope if I am to get a worthwhile scope image.
Anyone using a scope with their picoammeter and solved this issue already? I'm debating whether to punt on an oscilloscope altogether and just use an arduino's ADC to collect data and play it back after the fact, especially if everything external to the picoammeter needs to sit inside a secondary enclosure anyway.
-
This is how I ended up digging into our ancestors' use of AC for sensitive measurements... it is much easier to filter out noise.
Is there any way the thing you are trying to measure can be made to use AC instead of DC?
-
Not sure, but aren't we up against a hard limit of around 0.5Hz that we can feed into the picoammeter if we expect to get anything recognizable at the output? Or did you mean using an extremely low frequency AC signal and filtering out whatever is above it? I guess it could be done like AM radio, but with just a very low frequency carrier signal?
This has got to be a common problem, so it seems like there should be ready-to-hand solutions for it.
-
The 0.5Hz isn't a hard limit - it's a first order 3dB/Octave roll off (so it should still be a fair way down by 60Hz).
If you're getting lots of 60Hz hum, it suggests that there's either something not grounded to the scope ground, or you have multiple grounds, causing a ground loop. Try the picoammeter itself just connected to the scope and sitting on a piece of card or something if on a conductive surface. Then work your way forward - connect the negative of the current sourcing setup, etc. to see when the hum starts.
-
Not sure, but aren't we up against a hard limit of around 0.5Hz that we can feed into the picoammeter if we expect to get anything recognizable at the output? Or did you mean using an extremely low frequency AC signal and filtering out whatever is above it? I guess it could be done like AM radio, but with just a very low frequency carrier signal?
This has got to be a common problem, so it seems like there should be ready-to-hand solutions for it.
I meant extremely low frequency AC. Your scope, when set to do Averaging, will clean it up nicely and leave you with a measurable signal. It is like a "ghetto lock-in amplifier".
Lock-in amplifiers are sensitive to exactly one frequency, and extremely good at rejecting noise. They are used for many scientific experiments where signals are buried deep in the noise. However, they may struggle go down as low as 0.5Hz, so here the "ghetto" solution with a scope will beat the pros!
-
Or maybe the oscilloscopes you and I are using are obsolete rubbish? I thought it was impressive how picotechnology is able to pull a clear human heartbeat signal out of the noise, including 50/60hz mains noise: https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements (https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements)
Compared to that it's like we're trying to get clear crisp oscilloscope measurements using only stone knives and bearskins.
-
Problem mostly solved: I just now tried this guy's solution, which is both cheap and easy, for the mains noise problem, and for me it eliminated 95% of the problem:
https://www.youtube.com/watch?v=k-vbFcoCZ_k (https://www.youtube.com/watch?v=k-vbFcoCZ_k)
:clap:
I suspect switching to a battery powered oscilloscope would probably improve it even further.
FWIW, from what I've read on one of the EEVBLOG threads, you can actually power the Rigol 1054Z oscilloscope by applying 48VDC directly to the plug from four lead-acid batteries in series. I'm sure there must be a more elegant way, but for quick and dirty it reportedly works. Just hold on to your still good but time-to-replace 12v UPS batteries and the setup cost would be "free."
-
Looks like a winner! Can your experiment be battery powered? That way only the scope would run on A/C...
-
Or maybe the oscilloscopes you and I are using are obsolete rubbish? I thought it was impressive how picotechnology is able to pull a clear human heartbeat signal out of the noise, including 50/60hz mains noise: https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements (https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements)
Compared to that it's like we're trying to get clear crisp oscilloscope measurements using only stone knives and bearskins.
The "secret sauce" for the Picoscope is the use of a differential probe (which you can add to any oscilloscope, but they are typically not cheap!). Basically you take the signal at the two points in the circuit as usual, and subtract them from each other. Most of the noise disappears (it is the same on both) leaving you with pure, beautiful signal!
-
Or maybe the oscilloscopes you and I are using are obsolete rubbish? I thought it was impressive how picotechnology is able to pull a clear human heartbeat signal out of the noise, including 50/60hz mains noise: https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements (https://www.picotech.com/library/oscilloscopes/rejecting-common-mode-noise-in-oscilloscope-measurements)
Compared to that it's like we're trying to get clear crisp oscilloscope measurements using only stone knives and bearskins.
The "secret sauce" for the Picoscope is the use of a differential probe (which you can add to any oscilloscope, but they are typically not cheap!). Basically you take the signal at the two points in the circuit as usual, and subtract them from each other. Most of the noise disappears (it is the same on both) leaving you with pure, beautiful signal!
The commercial ones seem aimed at measuring high voltages, so I'm guessing that's why they're expensive. However, for low voltages, there seem to be a large number of inexpensive DIY designs: https://www.google.com/search?q=differential%20probe%20gerber&tbm=isch&tbs=rimg%3ACWeciPUQwj_1uImCeGH5BJHQbPPW0L_1DhPxAygrLOQpin3YQ0ELk3teP4Iu_1r6j3sD_1SBWt8NQNtM-2F44aBtmkX0diHXK2dPmMIFnrhp25pKxs9HMVYlQDYlx9qMC3CRxEUBA32q5JxZ_1R0qEgmeGH5BJHQbPBFIJvnsVGGJoCoSCfW0L_1DhPxAyEbAPRD3iDVyUKhIJgrLOQpin3YQRAdZXgLy1CSwqEgk0ELk3teP4IhGZhwLTi9hrrioSCe_1r6j3sD_1SBEfM-bHWtpY3LKhIJWt8NQNtM-2ERuPJqwlBQvdMqEgl44aBtmkX0dhGwhV7mT1jK9CoSCSHXK2dPmMIFEbAPRD3iDVyUKhIJnrhp25pKxs8RFxczlJzTtuIqEglHMVYlQDYlxxEq5N5S9qTNbyoSCdqMC3CRxEUBEa7J-cdzD1zUKhIJA32q5JxZ_1R0Rc90D_15XAwGhh1DKYrO9W2L0&rlz=1C1CHBF_enUS872US872&hl=en&ved=0CB0QuIIBahcKEwiY_6mt-t_pAhUAAAAAHQAAAAAQBw&biw=1087&bih=535 (https://www.google.com/search?q=differential%20probe%20gerber&tbm=isch&tbs=rimg%3ACWeciPUQwj_1uImCeGH5BJHQbPPW0L_1DhPxAygrLOQpin3YQ0ELk3teP4Iu_1r6j3sD_1SBWt8NQNtM-2F44aBtmkX0diHXK2dPmMIFnrhp25pKxs9HMVYlQDYlx9qMC3CRxEUBA32q5JxZ_1R0qEgmeGH5BJHQbPBFIJvnsVGGJoCoSCfW0L_1DhPxAyEbAPRD3iDVyUKhIJgrLOQpin3YQRAdZXgLy1CSwqEgk0ELk3teP4IhGZhwLTi9hrrioSCe_1r6j3sD_1SBEfM-bHWtpY3LKhIJWt8NQNtM-2ERuPJqwlBQvdMqEgl44aBtmkX0dhGwhV7mT1jK9CoSCSHXK2dPmMIFEbAPRD3iDVyUKhIJnrhp25pKxs8RFxczlJzTtuIqEglHMVYlQDYlxxEq5N5S9qTNbyoSCdqMC3CRxEUBEa7J-cdzD1zUKhIJA32q5JxZ_1R0Rc90D_15XAwGhh1DKYrO9W2L0&rlz=1C1CHBF_enUS872US872&hl=en&ved=0CB0QuIIBahcKEwiY_6mt-t_pAhUAAAAAHQAAAAAQBw&biw=1087&bih=535)
If we can find an already vetted project that has gerber files, I'd be very interested in putting one together. I'm pretty sure Elektor may have some like that, though it may require becoming a member to access them.
Edit1: I found an open source one that looks fairly complete: https://github.com/nostradomus/LabTools_100MHz-Differential-Probe#printed-circuit-board (https://github.com/nostradomus/LabTools_100MHz-Differential-Probe#printed-circuit-board)
It has gerber files for the PCB, a complete BOM, a schematic, and some articles documenting it.
Is there anything better? If not, then I may order the board and the parts.
Edit2: Because the picoammeter has low bandwidth, maybe it can't really leverage a fast oscilloscope anyway. Maybe something as simple as logging voltages, and perhaps published via either bluetooth or wi-fi, similar to a pockit (https://www.amazon.com/gp/product/B07PSFQ2Z6/ref=ox_sc_act_title_3?smid=A2805HGTJEJ4ZY&th=1 (https://www.amazon.com/gp/product/B07PSFQ2Z6/ref=ox_sc_act_title_3?smid=A2805HGTJEJ4ZY&th=1)), is all that's really needed:
-
Yes, some way of logging the readings is probably the way to go. Maybe your next investment could be a bench DMM with logging? That would also give you averaging and other cool things that might help you with what you are doing.
What's the lowest usable (acceptably stable) reading you can get out of the picoammeter you just built?
Sometimes I use the scope for logging, but obviously you only have about one minute of history at the slowest sweep speed (I don't know, maybe your scope does more?).
-
What's the lowest usable (acceptably stable) reading you can get out of the picoammeter you just built?
Good question. That partly depends on what you mean by stable. For instantaneous readings without a secondary shielding enclosure , the measurements are stable at the 10's digits, while the 1's digits do tend to move around. If one were to do a long-term average, then I'd expect even single digits might be stable. I have a hunch that a secondary shielding enclosure will help, but since I haven't actually tried it yet, your guess is as good as mine.
Sometimes I use the scope for logging, but obviously you only have about one minute of history at the slowest sweep speed (I don't know, maybe your scope does more?).
My scope can show at most 10 minutes of history on one screen.
------------------------------------------
I'm finding that a big culprit for 60hz noise is AC-DC switched mode power supplies. I tested 6 of them, and they were all noisy. Even just plugging one in was enough for a large ripple to show up on the oscilloscope. In contrast, DC-DC switched mode power supplies don't seem to be a problem.
Based on current findings, I'd recommend removing all AC-DC switched mode power supplies from whatever room you're doing your measurements in.
-
Getting a stable 10's of pA measurement is an excellent achievement. A lot of things have to be "just so" for that to happen.
-
Yes, some way of logging the readings is probably the way to go. Maybe your next investment could be a bench DMM with logging?
I think I'll use an arduino to take picoamp measurements at regular intervals and then flash an LED to broadcast the information after each measurement is taken. I'm guessing I'll read the optical telementry by drilling a tiny hole in the secondary shielded enclosure. Once read, I'll immediately plot it and/or enter it into a timebase database of some kind.
So, I'll need to find some easy to use ploting/charting/graphing library. I used to use plot.ly, which I liked a lot, but it has gone corporate and I'm not sure how much of it can be accessed anymore.
-
Don't forget, an Arduino also radiates RF simply because it runs at 16MHz and does a lot of fast switching. It could play havoc if it is near a sensitive measurement. Depending on how sensitive the measurements you are trying to make are, it might be best to keep things analog inside the experiment's shield, and convert to digital outside?
-
It's starting to look as though the noise is getting under much better control. Here are some scope shots I took, at different frequencies, of me sending a sinusoid of between 0 and 500 millivolts to the picoammeter (which, at least a low frequency, should correspond to between 0 and 500 picoamps).
0.01Hz:
[attach=1]
0.1Hz:
[attach=2]
1Hz:
[attach=3]
Obviously the effects of limited bandwidth are becoming apparent.
Adjusting the same 1Hz scope capture to show more detail:
[attach=4]
-
10Hz:
[attach=1]
At 10Hz most of the detail is almost gone.
-
Don't forget, an Arduino also radiates RF simply because it runs at 16MHz and does a lot of fast switching. It could play havoc if it is near a sensitive measurement. Depending on how sensitive the measurements you are trying to make are, it might be best to keep things analog inside the experiment's shield, and convert to digital outside?
I just now did the experiment. I plugged in an Arduino Uno right next to it and even powered it with a buck converter, hoping to create a worst case. I put it right next to the picoammeter, and it didn't perturb the measurements at all. Same thing with an Arduino Mega2560: no effect. It seems that high frequencies don't bother the picoammeter. Or, maybe it's their low power. Either way, all of that is really good news! :-+
Apparently it's the opposite with lower frequencies though. That 60Hz AC frequencyis quite visible if I turn the oscilloscope up to 10mv/div. I need to do further isolation and removal of other SMPS's from the area though before I can be sure the 60hz noise is coming from the Rigol and not something else. I would think that a Rigol would be pretty well designed to avoid giving off that kind of noise, so I suspect it may not be to blame. With the bad SMPS's that I've already removed, they could affect the picoammeter measurements even at several feet away.
--------------------------
In the scope shots I posted just above, it's clear that the swing of the output voltage on the picoammeter is getting reduced as frequency increases. This is not unexpected, given what Gyro said about the bandwidth of this design. Question: Is the picoammeter still faithfully measuring and reporting the amount of picoamps passing through it on the way to ground while the bandwidth limitations are especially noticeable, or does the picoammeter lose accuracy under those conditions? What I mean is, there seem to be three possible cases: (1) are the picoamps being throttled but still accurately reported, or (2) are the picoamps actually the same (unchanged) and simply not accurately reported, or (3) are the picoamps both throttled and not accurately reported? i.e. which of the 3 cases is the reality? I can't answer the question experimentally without an accurate higher bandwidth picoammeter, which I don't have.
-
Getting a stable 10's of pA measurement is an excellent achievement. A lot of things have to be "just so" for that to happen.
The credit goes to Gyro for coming up with a good design, and the credit goes to you for figuring out that Gyro had a worthwhile design and then suggesting it! 8)
-
Cool scope shots! Do you always work this fast? 8)
Hard to say if the picoamps are being affected as well as just the reading. It sounds plausible that it could, since we are depending on a virtual ground created by the op amp inside the box.
What is the max number of picoamps that the Gyro meter can display? - maybe there is an overlap with the low range of the uCurrent? (I got it down to about 500pA on AC, might have got lower if I had an amplifier to boost the signal coming out of it).
Lastly, consider that even if the picoammeter may be immune to RF, that may not be true of the thing you are measuring inside the experiment's enclosure?
-
Getting a stable 10's of pA measurement is an excellent achievement. A lot of things have to be "just so" for that to happen.
The credit goes to Gyro for coming up with a good design, and the credit goes to you for figuring out that Gyro had a worthwhile design and then suggesting it! 8)
Thank you @NeverDie. Whilst the 'Gryo Picoammeter' has a nice ring to it, I can't claim any credit for transimpedance amplifiers, bias networks, air wiring, or even relying on suitably protected internal protection diodes (maybe noticing that they are bootstrapped on the 662 though). Putting them all together in a screened box basically comes down a bit of attention to detail and conservative engineering, which is always good to strive for. :)
P.S. Agreed, really nice scope captures! I wish I'd done some work on characterising the AC performance myself. Although you don't want to mess with unnecessarily soldering Polystyrene capacitors, there's probably a sweet spot for being able to externally resolve the wanted signal, while achieving maximum frequency response. It probably varies from case to case though.
-
Getting a stable 10's of pA measurement is an excellent achievement. A lot of things have to be "just so" for that to happen.
The credit goes to Gyro for coming up with a good design, and the credit goes to you for figuring out that Gyro had a worthwhile design and then suggesting it! 8)
Thank you @NeverDie. Whilst the 'Gryo Picoammeter' has a nice ring to it, I can't claim any credit for transimpedance amplifiers, bias networks, air wiring, or even relying on suitably protected internal protection diodes (maybe noticing that they are bootstrapped on the 662 though). Putting them all together in a screened box basically comes down a bit of attention to detail and conservative engineering, which is always good to strive for. :)
P.S. Agreed, really nice scope captures! I wish I'd done some work on characterising the AC performance myself. Although you don't want to mess with unnecessarily soldering Polystyrene capacitors, there's probably a sweet spot for being able to externally resolve the wanted signal, while achieving maximum frequency response. It probably varies from case to case though.
Just wanted to say impressive results and very nice teamwork. Hope at some point when you guys think the time is right or as you go you might add some summarizing thoughts on this in way that includes a blend of high level fundamental principles for people just getting started with very low current measurements and your key learnings for more advanced measurers. fwiw, this endeavor reminds me a little of other “time nut, volt nut, and other nut” threads and endeavors in which people are pushing the boundaries of what can be cost-effectively measured with respect to pretty small and low level signals. It’s very much a testament to collaborative ingenuity. :-+
-
What is the max number of picoamps that the Gyro meter can display?
+1
Maybe @Gyro can comment?
As for noise, what I'm currently experiencing is comparable to the sort of stuff Dave Jones comments about toward the end of:
https://www.youtube.com/watch?v=OYOYI_IPKGY&t=475s (https://www.youtube.com/watch?v=OYOYI_IPKGY&t=475s)
where he blames his noise on things like long test probe wires and environmental noise. He relegates it to a topic for another video and does nothing about it. i.e. it really is good enough for Australia! :-DD
So, with that as the acceptance criteria, I deem this build of the picoammeter as finished.
Getting rid of the remaining noise will be the topic of a new build. And I think maybe including a BNC connector might be a step in that direction. What do you guys think? The tinyCurrent has one as one of its main selling points:
(https://www.n-fuse.co/assets/products/tinycurrent/MVIMG_20181006_182333_336.jpg)
And their github has some captured images showing a reduction in noise by using it rather than the banana plugs.
-
What is the max number of picoamps that the Gyro meter can display?
+1
Maybe @Gyro can comment?
It can do at least +/- 2.5nA (+/- 2.5V output) based on the 9V battery discharge curve (at the battery replacement point). With a fresh battery it will do at least +/- 4nA.
-
Just wanted to say impressive results and very nice teamwork. Hope at some point when you guys think the time is right or as you go you might add some summarizing thoughts on this in way that includes a blend of high level fundamental principles for people just getting started with very low current measurements and your key learnings for more advanced measurers. fwiw, this endeavor reminds me a little of other “time nut, volt nut, and other nut” threads and endeavors in which people are pushing the boundaries of what can be cost-effectively measured with respect to pretty small and low level signals. It’s very much a testament to collaborative ingenuity. :-+
Agreed, these guys (NeverDie here and SilverSolder in his other thread) are achieving some very good results, given the very basic tools for the job. :-+
You could also have a look at the original Picoammeter thread, which has some useful insights from the like of Alex Nikitin... https://www.eevblog.com/forum/projects/picoammeter-design/ (https://www.eevblog.com/forum/projects/picoammeter-design/)
... but it is most definitely worth downloading 'the bible' - the Tektronix / Keithley Low Level Measurement Handbook, which is conveniently free and is a very easy read, with lots of diagrams:
https://www.tek.com/document/handbook/low-level-measurements-handbook (https://www.tek.com/document/handbook/low-level-measurements-handbook)
Edit: The thread title could do with a bit of tweaking by now though.
-
You're right. I just now renamed the thread.
-
So, I think maybe using an isolated BNC jack for the output:
(https://www.mouser.com/images/amphenol/images/031-10-RFX_1_SPL.jpg)
and a non-isolated BNC jack for the input (so that it immediately grounds to the enclosure) might be the way to go for build #2.
This way, using coax cables, hopefully the noise, especially on the input side, will be reduced.
If it turns out you want banana plugs in some situation, then there are always BNC-to-banana-plug adapters that you could plug in.
I'm hoping this, together with using oscilloscope probes, will maybe overcome some of the "long wire" noise problems. Plus, it would leverage oscilliscope probes and wiring, which you would already have.
Or, is there something even better that could be used instead? e.g. don't some oscilliscope probes have even extra shielding of some kind?
-
You can probably dispense with the 1k resistor on the +Ve output and use a non-isolated BNC for the output too. I put the resistor there as a precaution but it shouldn't affect stability as long as the one on the -Ve terminal (centre pin) is still present.
-
These banana-to-BNC adaptors are super handy - no home is complete without them!
[attachimg=1]
-
Are 50-ohm connectors the right choice? From what I can tell, it seems the most common....
-
You can probably dispense with the 1k resistor on the +Ve output and use a non-isolated BNC for the output too. I put the resistor there as a precaution but it shouldn't affect stability as long as the one on the -Ve terminal (centre pin) is still present.
Should I use a shorting cap on the input BNC when doing the zero calibration on the picoammeter?
(https://www.mouser.com/images/fluke/lrg/5085.jpg)
Or should I use a 50 ohm BNC terminator instead?
(https://www.mouser.com/images/bel/images/inhouse_vbt50_t.jpg)
-
IIRC I didn't have serious noise problems when using a normal scope probe. The circuit was mounted in a conductive box with a small hole for the probe cable, the box was grounded to the probe's alligator clip in some random place.
BNC should work too. You don't need to care about wave impedance, it has nothing to do with connector resistance (which is close to zero) or anything like that. Just take 50Ω.
You also want shielding around the circuit whose current you are measuring, not sure if you have that. I simply put everything except the scope in one box.
In the scope shots I posted just above, it's clear that the swing of the output voltage on the picoammeter is getting reduced as frequency increases. This is not unexpected, given what Gyro said about the bandwidth of this design. Question: Is the picoammeter still faithfully measuring and reporting the amount of picoamps passing through it on the way to ground while the bandwidth limitations are especially noticeable, or does the picoammeter lose accuracy under those conditions? What I mean is, there seem to be three possible cases: (1) are the picoamps being throttled but still accurately reported, or (2) are the picoamps actually the same (unchanged) and simply not accurately reported, or (3) are the picoamps both throttled and not accurately reported? i.e. which of the 3 cases is the reality? I can't answer the question experimentally without an accurate higher bandwidth picoammeter, which I don't have.
(2) or (3), depending on the speed of the opamp.
The thing about the capacitor is that it has low impedance at high frequencies, so tiny swings of the opamp's output produce the desired input AC current and are almost invisible on the output. So you don't see it, but the current flows, as long as the opamp is fast enough to produce it and maintain effective virtual ground on the input node.
-
So, if it really is #2, then it seems like there should exist a way for a computer to reconstruct and plot what the true current flow is. Is there a commonly used computer program for doing that?
--------------------------------
Edit1: I ordered some BNC connectors, and they may arrive as early as tomorrow. I'm debating whether to upgrade the current picoammeter or to make a new one. If I make a new one, then I can compare results, and it will be easier to tell how much improvement (or not) the BNC connectors make, as well as any other upgrades anyone wants to suggest.
Anyhow, I ordered enough parts so that I can both make a new picoammeter and upgrade the "old" one if the upgrades do yield improvement. For example, although it's overkill for my current needs, it might be fun to see if we can accurately measure single digit picoamps.
Therefore, I hereby solicit suggestions for further improvements, whether relating to noise or anything else.
Edit2: For the next build I plan to use a tilt switch instead of a reed switch to turn the picoammeter on-off:
https://smile.amazon.com/gp/product/B07MR88MQL/ref=ppx_yo_dt_b_asin_title_o00_s00?ie=UTF8&psc=1
I'll simply turn the picoammeter upside down when not in use. No magnet required!
-
It's easier not to have the bandwidth limit in the first place. The LMC662 datasheet discusses how much capacitance is required for the amplifier to work correctly and remain stable; IIRC it is much less than people typically used in the picoammeter thread. The reason for the bandwidth limit was to limit noise and make it easier to measure average DC current with high resolution.
-
You can probably dispense with the 1k resistor on the +Ve output and use a non-isolated BNC for the output too. I put the resistor there as a precaution but it shouldn't affect stability as long as the one on the -Ve terminal (centre pin) is still present.
Should I use a shorting cap on the input BNC when doing the zero calibration on the picoammeter?
(https://www.mouser.com/images/fluke/lrg/5085.jpg)
Or should I use a 50 ohm BNC terminator instead?
(https://www.mouser.com/images/bel/images/inhouse_vbt50_t.jpg)
You're right, 50R BNCs are most common and a sensible choice.
No, never use a shorted input or terminator to zero the picoammeter - it won't do any damage, but the output will go to one rail and stay there (it basically becomes a voltage comparator circuit, biased by the opamp offset voltage). The rule is to zero voltmeters with a shorted input and ammeters with an open input.
Connect the input to the -Ve output to zero the 662 offset voltage. When you remove the link, it will then just read the input bias current (hopefully single digits fA). If you want to shield the input to make this measurement as stable as possible, then either put something across the end of the BNC - copper tape, coin, BNC plug with centre pin removed.
-
Haven't read the thread yet but thought it might be worth mentioning Dave did a video on his Keithley 480 Picoammeter and shows the stability just on an open bench at 23mins in. I brought one a while back as well that I got round to repairing a few years ago.
https://www.youtube.com/watch?v=zDcl0-t7ceY (https://www.youtube.com/watch?v=zDcl0-t7ceY)
-
Can desktop ESD matting be used on the floor as well? I certainly hope so, as I just now purchased a 32 feet long x 36" wide 2-layer roll of it that was on discount, so I hope to use it for both surfaces. The conductive side is made out nitrile butadiene rubber (NBR) and the dissipative side is some kind of semi-gloss material. Hakko brand, so I presume it's quality.
I would have gone for natural rubber, but it was only available in a hideous green color, whereas this stuff is neutral gray. I just hope it can withstand me rolling my chair around on top of it. If it's capable of that, then this will be a major upgrade toward getting rid of the static influences on the picoammeter measurements.
Speaking of Hakko, I notice they also sell what they describe as a static eliminating fan (?!): http://www.hakko.com/english/products/hakko_fe510_feature.html#productNav (http://www.hakko.com/english/products/hakko_fe510_feature.html#productNav)
Anyone here tried or using such a thing, and does it work? It looks as though, at present, it's only being sold in Japan and Asia.
-
[...]
Connect the input to the -Ve output to zero the 662 offset voltage. When you remove the link, it will then just read the input bias current (hopefully single digits fA).
[...]
So how do you use that result - do you make a note of the number, and take it as the real zero pA reading?
-
You zero it using the trim pot.
-
Yes, it's better to zero the LM662 offset voltage and then take account of any residual bias current and its polarity when making very low level measurements.
If you try to 'fudge' the overall (offset voltage + bias current) reading to zero using the offset voltage adjustment, you are likely to run into problems where the source voltage is low.
With good insulation and a nice clean LM662, the bias current should be right down in the low fA region anyway, with noise dominating (mine reads zero with open circuit (but screened) input to the 10s of microvolt level after offset voltage trimming).
EDIT: I've just been and checked mine. Its output has shifted by about 10uV from zero at least 2 years after zeroing the offset voltage, that is most likely down to the Vos changing with battery voltage rather than any change in bias current. Again, that's with open circuit but screened input. Noise is about 10uV pk-pk.
-
Yes, it's better to zero the LM662 offset voltage and then take account of any residual bias current and its polarity when making very low level measurements.
If you try to 'fudge' the overall (offset voltage + bias current) reading to zero using the offset voltage adjustment, you are likely to run into problems where the source voltage is low.
With good insulation and a nice clean LM662, the bias current should be right down in the low fA region anyway, with noise dominating (mine reads zero with open circuit (but screened) input to the 10s of microvolt level after offset voltage trimming).
EDIT: I've just been and checked mine. Its output has shifted by about 10uV from zero at least 2 years after zeroing the offset voltage, that is most likely down to the Vos changing with battery voltage rather than any change in bias current. Again, that's with open circuit but screened input. Noise is about 10uV pk-pk.
Aha, I guess by your definition of fudging, I had been fudging it then. I don't see any other way though. How is it that you are supposed to zero just the offset voltage but account for the bias current (and its polarity) separately? I've been treating this as a black box, so maybe I now need to delve under the hood....
-
Due to the very low bias current of the LM662, you should hopefully find that you reach very close to the same result (zero setting), its offset voltage should dominate the adjustment :) The difference at the output should be in the low uV range.
Luckily, with the input shorted to the -ve output, you don't need to worry about screening - you are effectively creating a low impedance voltage follower which follows the offset adjustment pot wiper, and then zeroing the offset voltage relative to ground (the +ve output).
-
How is it that you are supposed to zero just the offset voltage but account for the bias current (and its polarity) separately? I've been treating this as a black box, so maybe I now need to delve under the hood....
By bypassing the feedback resistor and shorting IN- to OUT. This reduces feedback resistance and therefore detection sensitivity a billion or trillion times, so the output no longer reacts to input current and only offset voltage remains.
-
You can probably dispense with the 1k resistor on the +Ve output and use a non-isolated BNC for the output too. I put the resistor there as a precaution but it shouldn't affect stability as long as the one on the -Ve terminal (centre pin) is still present.
Should I use a shorting cap on the input BNC when doing the zero calibration on the picoammeter?
(https://www.mouser.com/images/fluke/lrg/5085.jpg)
Or should I use a 50 ohm BNC terminator instead?
(https://www.mouser.com/images/bel/images/inhouse_vbt50_t.jpg)
You're right, 50R BNCs are most common and a sensible choice.
No, never use a shorted input or terminator to zero the picoammeter - it won't do any damage, but the output will go to one rail and stay there (it basically becomes a voltage comparator circuit, biased by the opamp offset voltage). The rule is to zero voltmeters with a shorted input and ammeters with an open input.
Connect the input to the -Ve output to zero the 662 offset voltage. When you remove the link, it will then just read the input bias current (hopefully single digits fA). If you want to shield the input to make this measurement as stable as possible, then either put something across the end of the BNC - copper tape, coin, BNC plug with centre pin removed.
OK, after re-reading this, I think I get it now.
Could I use a tilt-switch, perhaps in series with a reed switch to make a double trigger, on the inside of the enclosure to do the shorting? By shorting inside the enclosure, I'm thinking that it would accomplish the goal of "shield the input to make this measurement as stable as possible." On the other hand, maybe having that wiring "flapping in the breeze" when the double trigger isn't engaged might cause different problems. Not sure.
-
Luckily, with the input shorted to the -ve output, you don't need to worry about screening - you are effectively creating a low impedance voltage follower which follows the offset adjustment pot wiper, and then zeroing the offset voltage relative to ground (the +ve output).
By "you don't need to worry about screening," do you mean that shielding during the zeroing isn't important after all? See post directly above.
-
Due to the very low bias current of the LM662, you should hopefully find that you reach very close to the same result (zero setting), its offset voltage should dominate the adjustment :) The difference at the output should be in the low uV range.
I think to date that's why I haven't noticed anything other than 0 volts output after completing the zeroing: I wasn't looking for anything in the low uV range. What would that low microvoltage represent though? Does each microvolt represent 1 pa of bias current, which, theoretically, I would need to add to the final picoamp measurement (assuming single digit pa measurements are possible), measured in the millivolt range?
-
Ok, I modified the picoammeter to use BNC connectors:
[attach=1]
I blocked off the unused hole with a piece of single sided copper PCB, held in place with scotch tape. Meh, not ideal, but hopefully it will do for the short-term.
-
[attach=1]
-
[attach=1]
-
[attach=1]
Unfortunately, I won't be able to test this new build until Friday, which is when I receive the BNC-to-BNC patch cable for hooking the output up to the oscilloscope:
(https://images-na.ssl-images-amazon.com/images/I/813v8diJl5L._AC_SL1500_.jpg)
https://www.amazon.com/gp/product/B079GSYT4M/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1 (https://www.amazon.com/gp/product/B079GSYT4M/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1)
In fact, I'm not even sure if those are the proper patch cables for doing that (?). I'd also use it for zeroing the picoammeter.
For hooking up the picoammeter to a DMM, I ordered these:
(https://images-na.ssl-images-amazon.com/images/I/518JNnix8qL.jpg)
https://www.amazon.com/gp/product/B07SQPVMLD/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1 (https://www.amazon.com/gp/product/B07SQPVMLD/ref=ppx_yo_dt_b_asin_title_o03_s00?ie=UTF8&psc=1)
though I'm not sure whether they would be better or worse than just using a patch connecting cable together with one of the BNC-to-Banana-Plug adapters, like the kind that SilverSolder showed in his post above.
If this together with everything else (shielding the device under test within another conductive enclosure and possibly power the Rigol 1054Z oscilloscope from DC battery voltage instead of mains AC) doesn't get rid of the of the 60hz common mode noise, then the only other thing I can think of to try would be to use tightly twisted pair cables and have them feed a unity gain differential op-amp, whose output would then feed into the existing picoammeter's input.
-
Could I use a tilt-switch, perhaps in series with a reed switch to make a double trigger, on the inside of the enclosure to do the shorting? By shorting inside the enclosure, I'm thinking that it would accomplish the goal of "shield the input to make this measurement as stable as possible."
Yes, but its off resistance will appear in parallel with the feedback resistor so you need a good, low leakage switch.
You don't need shielding for zeroing because the whole circuit is low impedance. An external cable will do.
-
Agreed, don't risk a reed switch, it will probably ruin your accuracy. You'll probably find that the offset only needs zeroing every few months at worst (it drifts only slightly as the battery voltage goes down, but you won't see it if your meter resolution is at the mV level).
Due to the very low bias current of the LM662, you should hopefully find that you reach very close to the same result (zero setting), its offset voltage should dominate the adjustment :) The difference at the output should be in the low uV range.
I think to date that's why I haven't noticed anything other than 0 volts output after completing the zeroing: I wasn't looking for anything in the low uV range. What would that low microvoltage represent though? Does each microvolt represent 1 pa of bias current, which, theoretically, I would need to add to the final picoamp measurement (assuming single digit pa measurements are possible), measured in the millivolt range?
No, you probably won't notice a difference - the external zeroing link just makes it much easier to get the offset voltage spot-on zero, without having to worry about noisy readings or the shielding.
Microvolt outputs represent FemtoAmps (pA/1000) 8) The typical spec for the LM662 bias current is 3fA. With the way I restricted the adjustment range of the offset pot, it's possible to zero the opamp to 0uV (assuming that you have a meter with that resolution). The reading is always 1pA/mV.
Luckily, with the input shorted to the -ve output, you don't need to worry about screening - you are effectively creating a low impedance voltage follower which follows the offset adjustment pot wiper, and then zeroing the offset voltage relative to ground (the +ve output).
By "you don't need to worry about screening," do you mean that shielding during the zeroing isn't important after all? See post directly above.
As above, if you have the external link in place, you're bypassing the 1G resistor with a short, so it doesn't care about shielding any more - it's not seeing fA or pA in that state.
Ok, I modified the picoammeter to use BNC connectors:
I blocked off the unused hole with a piece of single sided copper PCB, held in place with scotch tape. Meh, not ideal, but hopefully it will do for the short-term.
Looking good. :-+
-
Here's the present configuration for zeroing the picoammeter:
[attach=1]
In retrospect, maybe it would have been better if I had positioned the BNC female connectors so that I could have used a more elegant connection, such as:
[attach=2]
for zeroing the picoammeter using a connected DMM. I made this adapter by hooking together two separate adapters. I tried looking for a single adapter with this type of functionality, but I couldn't find any that connected the input from two male BNC connectors while also feeding a female BNC (for connecting to a DMM or an oscilloscope).
As for the cabling itself, I just now became aware that there exists something called "Low Triboelectric" cable, which claims to be lower noise than regular cabling. Pomona claims that it is "ideal for low power, low frequency measurements." https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf (https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf)
Would that be the best kind of cabling for connecting the DUT to the picoammeter, or the picoammeter to an oscilliscope? Or does there exist even better cabling for that purpose? Or, does it make no difference?
More testing to do, but the initial tests are demonstrating that the amount of measurement noise showing on the DMM is greatly reduced. :-+
-
BTW, probably the least expensive project box that leaves plenty of space for this picoammeter project is a simple tin box, such as:
(https://images-na.ssl-images-amazon.com/images/I/71ZrTuqB7BL._AC_SL1500_.jpg)
https://www.amazon.com/gp/product/B07DQFP7XM/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1 (https://www.amazon.com/gp/product/B07DQFP7XM/ref=ppx_yo_dt_b_asin_title_o01_s00?ie=UTF8&psc=1)
which, IIUC, is made from tin plated steel. If you buy a few the cost per box is less than $2.
-
For the zero adjustment (with the input connected to the output) the cable / input connector is not critical. The OP is working as a times 1 buffer, thus no more pA anymore. The stability of the offset is more like in the 10 - 100 µV order of magnitude anyway. So the BNC connectors and normal cables are OK.
The critical part is when actually measuring small currents - than the cable type can make a difference and ideally there would be no extra cable. One could consider directly a 2 nd box on top for the parts to test.
The is an alternative way to do the zero adjustment: connect a higher value resistors (e.g. 1 M - exact value does not matter) to connect the input to ground. The OP than works as amplifier for the offset. This way may be a little more sensitive to stray currents and shielding however, but it can work without a sensitive voltmeter at the output.
-
One could consider directly a 2 nd box on top for the parts to test.
How would the picoammeter input be wired to the "2nd box on top for the parts to test"?
-
Would connecting the picoammeter to a box with the DUT using this:
(https://static.bhphoto.com/images/images500x500/pearstone_abnc_a1_bnc_male_to_bnc_1576092181_924499.jpg)
be better than using a cable?
-
For the test box I would consider some connector / feed through and maybe a clip to hold the parts to test.
So the 2nd box directly connected and only one connector, e.g. the BNC output inside the 2nd box - so no additional cable or adapter. If a cable would be used for some external current, leave the 2nd box open.
To get test signals in (e.g. the an external DC voltage to test leakage) have some connectors at the outside and a cable with clip in the inside. So the 2nd box could be closed and the test signal (usually the other side of the part to test) applied externally.
-
Here's the present configuration for zeroing the picoammeter:
(Attachment Link)
In retrospect, maybe it would have been better if I had positioned the BNC female connectors so that I could have used a more elegant connection, such as:
(Attachment Link)
for zeroing the picoammeter using a connected DMM. I made this adapter by hooking together two separate adapters. I tried looking for a single adapter with this type of functionality, but I couldn't find any that connected the input from two male BNC connectors while also feeding a female BNC (for connecting to a DMM or an oscilloscope).
As for the cabling itself, I just now became aware that there exists something called "Low Triboelectric" cable, which claims to be lower noise than regular cabling. Pomona claims that it is "ideal for low power, low frequency measurements." https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf (https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf)
Would that be the best kind of cabling for connecting the DUT to the picoammeter, or the picoammeter to an oscilliscope? Or does there exist even better cabling for that purpose? Or, does it make no difference?
More testing to do, but the initial tests are demonstrating that the amount of measurement noise showing on the DMM is greatly reduced. :-+
The meter (and the scope) can be considered "half deaf" compared to the picoammeter, so they won't notice the quality of the cables unless they are truly terrible. The connection between the DUT and the picoammeter needs to be good, though. Short, and good.
-
Here's the present configuration for zeroing the picoammeter:
(Attachment Link)
In retrospect, maybe it would have been better if I had positioned the BNC female connectors so that I could have used a more elegant connection, such as:
(Attachment Link)
for zeroing the picoammeter using a connected DMM. I made this adapter by hooking together two separate adapters. I tried looking for a single adapter with this type of functionality, but I couldn't find any that connected the input from two male BNC connectors while also feeding a female BNC (for connecting to a DMM or an oscilloscope).
As for the cabling itself, I just now became aware that there exists something called "Low Triboelectric" cable, which claims to be lower noise than regular cabling. Pomona claims that it is "ideal for low power, low frequency measurements." https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf (https://www.mouser.com/datasheet/2/159/d4964_1_02-1514307.pdf)
Would that be the best kind of cabling for connecting the DUT to the picoammeter, or the picoammeter to an oscilliscope? Or does there exist even better cabling for that purpose? Or, does it make no difference?
More testing to do, but the initial tests are demonstrating that the amount of measurement noise showing on the DMM is greatly reduced. :-+
The meter (and the scope) can be considered "half deaf" compared to the picoammeter, so they won't notice the quality of the cables unless they are truly terrible. The connection between the DUT and the picoammeter needs to be good, though. Short, and good.
That makes sense.
What does he mean by "feed through"? Something like this?
(https://media.digikey.com/photos/Pomona%20Photos/MFG_5178.jpg)
-
What do you mean by "feed through"?
-
I just now received this:
(https://images-na.ssl-images-amazon.com/images/I/61w14Z5W-pL._AC_SL1000_.jpg)
My plan is to connect it to the picoammeter using:
(https://static.bhphoto.com/images/images500x500/pearstone_abnc_a1_bnc_male_to_bnc_1576092181_924499.jpg)
and to put the DUT inside it.
-
Sounds like a plan!
-
For doing detailed millivolt measurements on an oscilloscope (and maybe a DMM as well), I'm now convinced that the type of cabling used is absolutely critical. I've reached that conclusion from just now trying a number of different cables and noticing the differences.
WRT to common mode noise, the difference is just night and day when using BNC cables vs using regular banana-plug type cables. So much so that I'm surprised digital multi meters don't have BNC jacks on them as standard equipment. In fact, I just now did a search to see if any of them do, and I couldn't find a single handheld portable DMM's with built-in BNC connectors. At best all I could find were BNC to banana plug adapters. Only some of the high end lab bench multimeters seem to have BNC connectors. The closest I could find in hand-held were combo multimeter-oscilloscopes. I guess for DMM's common mode noise may simply get averaged out?
If you all already know what I mean, then great. If not, and you'd like me to put together some show and tell pictures for compare and contrast, let me know and I'll post some to illustrate just what I mean.
-
For doing detailed millivolt measurements on an oscilloscope (and maybe a DMM as well), I'm now convinced that the type of cabling used is absolutely critical. I've reached that conclusion from just now trying a number of different cables and noticing the differences.
WRT to common mode noise, the difference is just night and day when using BNC cables vs using regular banana-plug type cables. So much so that I'm surprised digital multi meters don't have BNC jacks on them as standard equipment. In fact, I just now did a search to see if any of them do, and I couldn't find a single handheld portable DMM's with built-in BNC connectors. At best all I could find were BNC to banana plug adapters. Only some of the high end lab bench multimeters seem to have BNC connectors. The closest I could find in hand-held were combo multimeter-oscilloscopes. I guess for DMM's common mode noise may simply get averaged out?
If you all already know what I mean, then great. If not, and you'd like me to put together some show and tell pictures for compare and contrast, let me know and I'll post some to illustrate just what I mean.
DMMs should be more or less immune to common mode noise. One of the big issues when making sensitive measurements is to avoid ground loops at all costs.
It would be a great idea to draw a schematic how the measurements are made, including the test instruments and power supplies in the diagram, and paying attention to which of them are connected to Earth.
-
DMMs should be more or less immune to common mode noise.
At least under present circumstances, what I'm seeing doesn't seem to be in complete agreement with that. My 500,000 count DMM does tend to settle if given enough time, but any movement on my part throws it off again.
Let's compare/contrast measurements using BNC cabling versus banana plug cabling:
If I turn off averaging and BW limiting, so as to give the noisest picture on my oscope, here is what my oscope shows both with nothing plugged into it and with a quality 6 inch BNC cable plugged into it (open circuit, so nothing on the other end):
[attach=1]
Now, here it is with the same settings, but with DMM probes plugged in instead:
[attach=2]
Increasing the voltage/reticule shows that it's more than 100 times worse:
[attach=3]
and from the wavelength it becomes clear that it's mains common mode noise that's a strong component of it.
I do think it maybe does make a difference if trying to get single digit picoamp resolution out of the picoammeter. Admittedly, the testing environment can probably be improved yet further, but for now it just is what it is. Less dramatic compare/contrast was done by the tinyCurrent guy. Here is the tinyCurrent noise when using the BNC connector:
(https://github.com/nfhw/tinycurrent/blob/master/Scope_Shots/output_noise_Coax.png?raw=true)
and here it is using leads:
(https://github.com/nfhw/tinycurrent/blob/master/Scope_Shots/output_noise_Leads.png?raw=true)
Well, just to be certain, after I get the lunchbox testing enclosure set up and connected to the picoammeter, I'll try doing the picoammeter measurements using a DMM both with regular DMM cabling and also with using coax connected to the DMM via a BNC adapter. I'm pretty sure I'll see some kind of difference, but I'll wait until then before I reach final conclusions.
The very first thing I'm planning to test with the picoammeter is this Vishay load switch:
https://www.vishay.com/docs/66597/sip32431.pdf (https://www.vishay.com/docs/66597/sip32431.pdf)
which claims to have a leakage current of just 10pa. My conjecture is that, given the environmental noise, it would be very challenging to accurately measure in the single picoamp digits with just a DMM and DMM probes without the benefit of the new test enclosure and upgraded cabling.
-
Anyhow, switching topics, I plan to use this as the passthru into the lunchbox test enclosure:
(https://images-na.ssl-images-amazon.com/images/I/41LaYPzuNuL._AC_.jpg)
I like that this type of passthru can be anchored to the wall of the enclosure with four bolts. A problem I've been having with the non-bolted screw-on type BNC connectors is that they tend to come loose after connecting to them a few times, and then the resulting twisting can put the solder connections inside in jeopardy.
Then for anything I want to test inside the test enclosure, I'll wire it up with its own personal one of these:
(https://images-na.ssl-images-amazon.com/images/I/61nxYBx%2BlaL._AC_SL1000_.jpg)
so that it can be easily connected to the passthru when its ready to be tested and then disconnected and removed from the enclosure when its testing is done, thereby making room for whatever comes next in the testing queue.
These parts won't be arriving until Wednesday at the earliest, though, so in the meantime I may wire up some less elegant temporary connections so that I can get started.
-
"Common mode" refers to the same signal being present on both leads, exactly the same. A DMM or scope is pretty good (but not perfect) in eliminating that.
What you are seeing appears to be plain old "differential mode" noise, caused by magnetic and electric fields in the room.
If you think about it, the leads from your DMM, if you touch the probes together, make up a big loop - essentially, like an inductor with one winding. Any magnetic field that passes inside the area of the loop will be "captured" as a signal in the leads.
A coaxial cable avoids most of that problem by creating the smallest possible area between the two leads... but it is not perfect, obviously.
It can be really hard to get a quiet low level signal in a typical home environment. Just yesterday, I was trying to measure a low level signal and couldn't figure out why the spectrum analyzer noise level was so high... until I unplugged the laptop charging on the bench, and the noise dropped significantly!
-
open circuit, so nothing on the other end
That's your problem right here.
Short the other end or connect it to an opamp configured to output 0V (or anything else) and you will see that shielding makes little difference.
-
A coaxial cable avoids most of that problem by creating the smallest possible area between the two leads... but it is not perfect, obviously.
I have a different model of it than that: think of the coax as joining two faraday cages. Then the internal (non-grounded) wire effectively remains on the inside of the cage and is protected from noise that way. At least that's how it should work when I connect the picoammeter to the lunchbox test enclosure using coax cable. The metal shell is ground in both enclosures, which is connected to the coax sheathing. Then both enclosures and the coax sheathing become one and the same ground, and effectively, one and the same enclosure, with a narrow tunnel (the coax cable) joining the two enclosures. When tied together this way it's like a single dogbone shaped enclosure.
-
open circuit, so nothing on the other end
That's your problem right here.
Short the other end or connect it to an opamp configured to output 0V (or anything else) and you will see that shielding makes little difference.
When I short the two leads together I get:
[attach=1]
So, the 60hz mains noise did disappear, but it appears there remains a lot of other noise. This seems consistent with silversolder's big loop theory.
-
open circuit, so nothing on the other end
That's your problem right here.
Short the other end or connect it to an opamp configured to output 0V (or anything else) and you will see that shielding makes little difference.
When I short the two leads together I get:
(Attachment Link)
So, the 60hz mains noise did disappear, but it appears there remains a lot of other noise. This seems consistent with silversolder's big loop theory.
Nice, but I can't take credit for Ampère's law (part of Maxwell's theory of electromagnetism)! ;D
According to Maxwell, the amount of voltage induced in a loop by a magnetic field that goes through it is dependent on the area covered by the loop and the strength of the field.
It works the other way round too: the bigger the loop area, the more magnetic field a given current will send into the surroundings!
This is a really neat trick to have in mind when you design circuit boards or cable layouts of any kind: keep it tight! Always keep the return current next to the outbound current. That way your project becomes less prone to interference from the outside, as well as less prone to sending noise into other things nearby.
This is also the reason twisted pairs of wire work so well: all the little loops end up canceling out.
-
This is also the reason twisted pairs of wire work so well: all the little loops end up canceling out.
I thought the theory behind twisted pair was that it helps only if you look at the voltage difference between the pair. At least, that's how ethernet does it. Does it really confer any advantage if not doing that?
-
Here's the stopgap version of my lunchbox test chamber:
[attach=1]
and here is what should be a fully functional test setup, with the picoammeter directly connected to the test chamber (as previously described above):
[attach=2]
Now all I need to do is put a DUT inside the lunchbox, solder its output connections to the BNC connection wires, close the lid, and take measurements! With this setup, I expect the measurements will be completely shielded from outside noise and interference. 8)
-
This is also the reason twisted pairs of wire work so well: all the little loops end up canceling out.
I thought the theory behind twisted pair was that it helps only if you look at the voltage difference between the pair. At least, that's how ethernet does it. Does it really confer any advantage if not doing that?
Yes, it does. As the wires twist, you can think of it as a whole bunch of little loops connected in series, but with opposite polarity. So when a large external magnetic field passes through the loops, each one cancels out its neighbor. It totally works, you can test it the same way you did your probes (i.e. short the end of a twisted pair, see what you get compared to an open loop).
It is worth using twisted pairs everywhere inside a sensitive experiment, to add that extra little bit of silence. After all, an aluminum box doesn't stop magnetic fields...
Looking at the voltage difference between a pair of wires is a separate technique that also reduces noise, but it is always better to avoid noise in the first place!
-
I've decided to make an easier first test. I'll test the S-5470, which is a current detecting switch that should turn-on at 700pa nominal (possible range is 520pa to 880pa): https://www.ablic.com/en/doc/datasheet/photo_ic/S5470_E.pdf (https://www.ablic.com/en/doc/datasheet/photo_ic/S5470_E.pdf) I'll have it light an LED when it detects the threshold current, and then measure with the picoammeter to see what the current actually was. That should be an easy measurement target to hit.
Then, its "typical" current consumption is around 10 to 20pa, which makes for a smaller measurement target, but it still should be easy for the picoammeter to measure.
So, we'll see if it the picoammeter's measurements are in agreement with both measurement targets.
-
After all, an aluminum box doesn't stop magnetic fields...
Should I maybe switch to a ferrous metal box for the picoammeter? The lunch box already is. The tradeoff is that aluminum is more conductive than iron, so I'm not sure which is the more important parameter to optimize.
-
In practice it's not going to make any difference. It's more down to practicalities - for instance, your lunch box is probably solderable, whereas Aluminium wouldn't be.
It looks as if you have good metal to metal contact on the lid too.
-
I should have said that aluminum does block magnetic fields for higher frequencies - but it won't block "DC" magnetic fields, or low frequency ones (like mains). That's where using twisted wires and keeping everything tight on your PCB board (minimize loop area, again) helps you a lot, even inside a box, and doing the two things together is normally enough to beat this problem into the ground.
-
Another test that might be fun to try, is to see if the uCurrent and the picoammeter agree (put the two in series and see what they say... for sure, the same current flows through both of them if they are in series, right?).
I would guess that at around 1,000pA, the two should be getting close, but that is just a wild guess...
-
Another test that might be fun to try, is to see if the uCurrent and the picoammeter agree (put the two in series and see what they say... for sure, the same current flows through both of them if they are in series, right?).
I would guess that at around 1,000pA, the two should be getting close, but that is just a wild guess...
Good idea. If I put an additional BNC connector on the lunchbox (making a total of two BNC connectors on the lunchbox instead of just the one that's there now), then I can put the uCurrent Gold inside the lunchbox and export its measurements using the additional BNC connector. That way the uCurrent Gold, which has an unshielded enclosure, can become better shielded. Maybe in this way the uCurrent Gold can even deliver the "superb accuracy" that's advertised for picoamp currents and that I quoted in the very first post of this thread. Also, it will then be possible to directly compare the uCurrent Gold's measurements with the picoammeter measurements, since I'll have access to both readings outside of the lunchbox test chamber at the same time.
Of course, this approach assumes that the uCurrent Gold doesn't throw off noise or interference that might screwup the picoammeter measurements. I don't know whether it will or not, but if we try, then we'll find out.
-
So, the 60hz mains noise did disappear, but it appears there remains a lot of other noise.
Well, that's the second problem: capturing a lot more bandwidth than you need ;) That LMC cirucit probably won't output more than a few kHz at best due to parasitic capacitances in the opamp.
I'm not saying not to shield. It's harmless, it may even help. If there is something I would be worried about, it's the output cable picking up mains electrostatic field and passing it into the shielded enclosure, if the output cable isn't driven with very low impedance but has some protection resistor in series or whatever. So yeah, coax is better or no worse.
Twisted pair does absolutely nothing for electrostatic fields, by the way.
I don't think you need to be worried much about magnetic fields. All the shielding I used was a grounded juice box (and it wasn't even closed perfectly tight) ;D
-
Another test that might be fun to try, is to see if the uCurrent and the picoammeter agree (put the two in series and see what they say... for sure, the same current flows through both of them if they are in series, right?).
I would guess that at around 1,000pA, the two should be getting close, but that is just a wild guess...
Good idea. If I put an additional BNC connector on the lunchbox (making a total of two BNC connectors on the lunchbox instead of just the one that's there now), then I can put the uCurrent Gold inside the lunchbox and export its measurements using the additional BNC connector. That way the uCurrent Gold, which has an unshielded enclosure, can become better shielded. Maybe in this way the uCurrent Gold can even deliver the "superb accuracy" that's advertised for picoamp currents and that I quoted in the very first post of this thread. Also, it will then be possible to directly compare the uCurrent Gold's measurements with the picoammeter measurements, since I'll have access to both readings outside of the lunchbox test chamber at the same time.
Of course, this approach assumes that the uCurrent Gold doesn't throw off noise or interference that might screwup the picoammeter measurements. I don't know whether it will or not, but if we try, then we'll find out.
Good idea (to compare the two).
Remember that the uCurrent Gold has its input negative and output negative terminals connected together, so you need to look carefully at how you connect them in series in the context of the screened enclosure.
The uCurrent does use autozero opamps, whose switching does generate some input noise, but this will be be high enough in frequency not to show up.
-
So, the 60hz mains noise did disappear, but it appears there remains a lot of other noise.
Well, that's the second problem: capturing a lot more bandwidth than you need ;) That LMC cirucit probably won't output more than a few kHz at best due to parasitic capacitances in the opamp.
I'm not saying not to shield. It's harmless, it may even help. If there is something I would be worried about, it's the output cable picking up mains electrostatic field and passing it into the shielded enclosure, if the output cable isn't driven with very low impedance but has some protection resistor in series or whatever. So yeah, coax is better or no worse.
Twisted pair does absolutely nothing for electrostatic fields, by the way.
I don't think you need to be worried much about magnetic fields. All the shielding I used was a grounded juice box (and it wasn't even closed perfectly tight) ;D
At least it is easy to see if you are getting any magnetic interference: Short the leads at the far end... what you see then, is going to be magnetic in origin (assuming the CMMR of your instrument is reasonable! --edit: and assuming there are no ground loops!).
I constantly get surprised by how much magnetic noise there is nowadays, when we are surrounded by switch mode power supplies.
-
It looks as if you have good metal to metal contact on the lid too.
Yup, I confirmed by probing with a DMM that the metal lunch box lid has continuity to the rest of the metal lunch box.
-
If there is something I would be worried about, it's the output cable picking up mains electrostatic field and passing it into the shielded enclosure, if the output cable isn't driven with very low impedance but has some protection resistor in series or whatever.
Yes, good point. That does seem like a possible weakness. To mitigate I'll use coax to connect the uCurrent Gold's output to to the external multimeter. I'll also check to see whether the picoammeter measurements become more noisy after I connect the multimeter to the uCurrent Gold's output by doing before-and-after measurements with the picoammeter (i.e. before connecting to the uCurrent Gold's output and after connecting).
I'm very open to alternative approaches though. If someone here can think of a better way, then I'm happy to do that instead.
-
I received the passthru's today, so, as discussed above, I upgraded the testing enclosure to use two of them:
[attach=1]
[attach=2]
And I have a termination cap to cover one of the passthru's when it's not in use:
[attach=3]
-
Looks good!
-
Here's the test circuit that will go inside the test enclosure.
[attach=1]
I'm laying it out here because it may be easier to see this way than when it's folded up inside the test enclosure.
The BNC connectors will connect to the passthru connectors. As configured, both the uCurrent Gold output and the picoammeter will share the same ground. Hopefully that won't be a problem. If it turns out that it is a problem, then I can install an isolated BNC passthru on the lunchbox to keep their grounds separate.
Test Plan: Initial voltage will be set to 1v, which should translate into 1 nanoamp. Therefore, both the uCurrent Gold and the picoammeter should be able to measure it. From there I'll reduce the voltage in steps and record both measurements with each step down.
I'm aware that the buck converter is a potential source of noise, but in previous tests it didn't seem to bother the picoammeter, possibly because of its high switching frequency. Nonetheless, I will take the minimum precaution of resting it on some 1/8" thick unclad FR4 so that it doesn't make direct electrical contact with the lunchbox through its metal standoffs.
One nice thing about the buck converter is that it has current limiting, which I set to 10ma. Therefore, if a short were to occur, maybe that would help prevent a catastrophic failure. Ideally I would set CC even lower, but I'm not sure how well the chinese design and components would behave if operated near its limits, so I gave it some extra headroom just to ensure that it won't trip over itself.
-
The uCurrent will respond to the noise, up to hundreds of KHz in my testing...
[attachimg=1]
-
Good to know.
What exactly is the chart illustrating? I'm lacking the full context.
-
Good to know.
What exactly is the chart illustrating? I'm lacking the full context.
It is called a Bode plot.
The frequency is along the bottom axis, goes from 500Hz to 500KHz (logarithmic)
The top trace, "Amplitude", shows the loss through the uCurrent in dB (compared to DC). So, you can see, as the frequency increases, at some point the output of the uCurrent starts to drop off. At -6dB, the output has fallen to half (not shown).
The bottom trace, "Phase", shows how many degrees behind the sine wave coming out of the uCurrent is compared to what went in.
To make Bode plots like this is super useful to understand all kinds of electronic components. There are all kinds of cool tools available to do that these days, many scopes have it built in, or you could use something like an Analog Discovery 2.
-
The immediate problem turns out not to be the buck converter but instead the picoammeter sharing the output ground with the uCurrent Gold. With the picoammeter powered on but both with or without power to everything else, what appears to be a positive feedback loop of some kind raises the voltage on the output of the picoammeter until it's about 4.5v (hmmm.. half of 9 volts powering the picoammeter?). It happens whenever I connect the uCurrent Gold BNC connector to the second BNC passthrough. It happens even if I remove both the battery and the buck converter from the enclosure.
So, I guess this means I may need to use an isolation BNC conector to keep the grounds separated. However, if I do that, there will be nothing to stop outside noise/interference entering into the enclosure from the DMM intended to measure the uCurrent Gold, so I'm guessing that won't be a good idea.
I think maybe maybe putting an arduino (or a logging DMM) inside the enclosure to log the voltage output of the uCurrent gold when the lid is closed might be the next step. That would allow the uCurrent Gold's output voltage to float, which seems to be the way it was designed to operate.
Anyone have any other suggestions?
-
Well, OK, probably the easiest way around this will be to take the picoammeter and uCurrent Gold measurements separately rather than at the same time. Maybe not ideal, but worth a try. I'll give it a go.
-
I did it, and the results are interesting:
1. The uCurrentGold is probably good enough for most picoamp measurements.
2. The DMM is the weak link in this test setup: because it is outside of the enclosure, it is still influenced by electrostatic fields. There's just no denying it: I can see it with my own eyes.
Therefore, to get repeatable single digit picoamp measurements, either a lot of external ESD protection needs to be in place (?) or else the DMM needs to take its measurements inside the test chamber enclosure. Maybe both.
What follows are the details.
This is what I had planned to use as the test setup before encountering the grounding problem described above:
[attach=1]
Therefore, to take the uCurrentGold measurements, I disconnected the picoammeter and shorted the uCurrent Gold's negative input lead to ground using an oscilloscope alligate clip, as shown next:
[attach=2]
Here are the measurements. The uCurrentGold was switched to the nanoamp setting. Input voltage is just what was reported on the face of the cheap buck converter (for this first set of measurements, I maybe should have checked the voltage with a voltmeter, but for expediency's sake, I just used the voltage that was displayed on the LEDs of the buck converter):
0.1v input, 0.23mv output
0.2v, 0.33mv
0.3, 0.43
0.4, 0.53
0.5, 0.63
0.6, 0.73
0.7,0.83
0.8, 0.93
0.9, 1.03
1.0, 1.13
So, to the nearest 10 picoamps, the results were remarkably consistent. From eyeballing the numbers, it looks as though there was a 0.13v offset that showed up in all the measurements.
-
Then, for the picoammeter readings to approximate the same test conditions, I wired the uCurrent Gold in series with the picoammeter and let it run, even though I disconnected the uCurrent Gold's output for the picoammeter measurements:
[attach=1]
The picoammeter measurements (see the following) were quite consistent with those of the uCurrent Gold above:
0.1v, 115mv
0.2v, 215mv
0.3v,315mv,
0.40, 412
0.5, 516
0.6, 613
0.7, 0.715
0.8, 0.813
0.9, 0.918
1.00, 1.015
This was just a quick first pass, but the results seem quite linear. This time the measurement offset appears to be only about 15mv.
In conclusion, I'd say either the uCurrent Gold or the picoammeter is probably good enough for measuring to the nearest 10 picoamps. If I were to repeat the measurements and strive for greater accuracy, I'd want either 1. the DMM to do its measuring inside the test enclosure to see whether it cuts down the noise any further or 2. for the DMM to be properly shielded (just as the picoammeter already is) if the DMM is to remain external to the enclosure. I'd be interested to hear ideas on how #2 might be accomplished while still keeping a visible display. Otherwise, option #1 becomes the choice by default.
With those improvements, I'm fairly confident that more precise measurements of one form or another should be possible. However, even without those improvements, the existing setup is probably already good enough for meeting my near-term measurement needs.
8)
I hope posting the above measurements enriches the discussion.
Thank you everyone for your helpful comments and suggestions.
-
That is super interesting - it seems that your theory that the uCurrent suffers from not being in a shielded box, might have something to it!
-
Nice test. :-+
A couple of things that come to mind:
1. The uCurrent gold uses a shunt resistor + voltage amplifier, rather than the transimpedance amplifier Picoammeter. The Picoammeter will attempt to keep its input at 0V, wheras the uCurrent will show a voltage drop (voltage burden) across its shunt resistor. The uCurrent is designed to keep its burden voltage as low as possible by using lots of voltage gain, but it is still there. Your method of generating the test current, using a low output voltage divider and series resistor, will tend to exaggerate the effect of the burden voltage (it would be different if you were using a higher supply voltage and much higher series resistor value).
2. Those little SMPS modules are pretty noisy. That will probably affect the uCurrent more (yes it would benefit from a shielded box due to its high voltage gain - and probably smaller terminals too :P).
P.S. You should be able to calculate the effect of the voltage burden on the uCurrent by including its shunt resistor value (10k on nA range?) as part of your resistor network.
-
I'm willing to remove the smps from the equation. What do you recommend I use in it's place? A 9v battery with potentiometer to dial-in the voltages? Or something else?
One alternate I can think of would be to use a large supercap with a low ESR to create a voltage source by just bleeding it down to different test voltage levels. It would be a bit more finicky to setup up though because of supercap tendencies to "regenerate" after voltage drops: I'd have to let it sit for quite a while for it to stabilize after adding or subtracting a large chunk of charge to/from it.
What do you recommend for an adjustable lowest noise voltage source?
-
A battery or supercap would both work - as long as you know what its voltage is, you can work out mathematically what you should be reading from the test, and compare with that... no need to adjust the voltage or anything.
The main thing is that the voltage has to be reliably stable (which, over a period of a few minutes of testing at low current, a 9V battery will provide, and probably a supercap too).
-
Some of the adjustable LDO's are billed as "ultra low noise." Any favorites for that category?
On the other hand, Dave Jones did a whole video on how linear regulators aren't actually all that different from switched mode power supplies. Or, at least, not as different as you might think.
-
I don't recall the Dave video, but was he saying that a linear regulator cannot quite clean up a switch mode power supply? - if you use it to regulate a battery or supercap, you'll be fine!
-
For the test signal I would use a battery and pot to set a value. At the current test level the relative resolution is limited anyway, so a battery is stable enough if not loaded very much. For low voltage / low power a reference chip could make a good low noise regulator.
An LDO or similar regulator can help a little with the ripple from an SMPs, but the effect is limited. No need for high efficiency for a test current in the nA range.
Higher frequency EMI could effect both the µCurrent and the TIA pA meter. So it is usually not such a good idea.
-
LT3042 was the best ultra-low noise adjustable LDO that I could find: https://www.mouser.com/datasheet/2/609/3042fb-1271662.pdf (https://www.mouser.com/datasheet/2/609/3042fb-1271662.pdf)
It's adjustable all the way down to 0 volts. If it has no advantages though, I'm happy to use just a 9v battery and trim pots, since that is easier to build.
-
@Kleinstein is right, just go with a pot and trim it in. Simple and good and good and simple!
-
If you want to get exact, the resistor values aren't nice clean numbers because you have to account for the parallel resistances:
[attach=1]
R2=9980, and R4=9891. Assuming the 9v battery is exactly 9v, then R1=7271. R1 would need to be adjusted, depending on what the true voltage on the battery actually is.
This way, with the potentiometer set to 1000 ohms, it generates 1na = 1000pa of current at the output (ammeter A5 in the schematic). i.e. 1 picoamp for every ohm on the potentiometer, making it's easy to dial-in an exact pa current if it's a 10-turn 1K potentiometer with a dial-indicator on it.
A bit of a hassle to construct, but it passes SPICE simulation.
If I'm not mistaken, attaching a current mirror to the output would effectively create a picoamp current source that you could also use in other applications. 8)
In my case, though, instead of 9v, I'm going to use an REF102CP precision 10-volt reference, just for convenience. REF102CP outputs exactly 10 volts to within plus or minus 0.0025 volts, so plenty good enough for this endeavor. It means not having to worry about or recheck the source voltage level, as it will always be 10v pretty much exactly. REF102CP has practically no noise: just 5uVpp at the chip, and that number will get divided by 10 million after it goes through the voltage dividers. https://www.ti.com/lit/ds/symlink/ref102.pdf (https://www.ti.com/lit/ds/symlink/ref102.pdf)
Edit1: The other advantage of making use of a 10v reference voltage source is that with it, each stage, if equiped with a trim potentiometer, can be precisely set by applying a precise 10v to it and then, measuring with a dmm, tweaking the potentiometer until the correct voltage division is achieved. This, then, gives an exact result to the accuracy of the DMM, which is more accurate measuring voltages than resistances anyway. No need to calculate the effect of parallel resistances, because those will be accounted for implicitly when tweaking the voltage dividers. In this way, only one precision resistor is needed, namely, the final 1K ohm resistor at the very final stage. All this is good, because with this approach I won't need a 4-wire resistance meter for precisely measuring resistances.
Of course, none of this would account for temperature drift on the resistances. Other than doing recalibrations, I'm not sure how I would adjust for that. Hopefully there won't be any meaningful temperature drift over the short duration of doing the measurements.
Edit2: To mitigate against temperature drift, I'll use high resistance in the voltage dividers, so that very little power will get dissipated into the resistors.
-
The attached circuit should do the trick.
[attach=1]
The first voltage divider takes the 10.00 reference voltage and creates 1 volt across a 10-turn 1K ohm potentiometer. That voltage is then divided by 1000 two times by the following two voltage dividers. Finally, a precision 1K resistor (0.01% tolerance) converts that voltage into picoamps. The trim pots are 25 turns each. The trim pots and test points, together with a quality DMM, are used to accurately set the voltage division at each stage by automatically accounting for the parallel resistances attached to it.
As stated above, this should allow me to precisely dial-in anywhere from one picoamp to 1000 picoamps, in single picoamp increments, with each "click" of the 10-turn 1K ohm potentiometer corresponding to one picoamp. Of course, this assumes that the potentiometer is perfectly linear, which it may not be. Before installing the 10-turn 1K potentiometer, I'll measure it with my DMM to see just how linear it is or isn't. If it turns out not to be good enough, then I'll either order a proper high precision 1K ohm potentiometer to replace it with (0.1% linearity is readily available). If worse comes to worst I'll simply replace the 1K potentiometer with a four decade resistor box that's accurate enough for the purpose.
My 10-turn 1K ohm potentiometer arrives tomorrow, so I'll build it then. :)
This should serve the current purpose of creating accurate very-low-noise test currents for the picoammeter and the uCurrent gold. Thus, it is intended to replace both the buck regulator and the existing less accurate voltage divider in the lunchbox test enclosure.
It may turn out that even if the uCurrent Gold can measure picoamps its construction allows too much leakage of picoamps for it to measure picoamps accurately in the single digits. We'll see soon enough.
-
To test the pA meter / TIA it would take a test current in the 1 nA range - the value depends on the resistor used in the feedback.
Ideally one would use a large resistor to set the current. Something like 1 M sounds reasonable, as one can get reasonable stable and accurate resistors of 1 M. Also measuring is still possible at 1 M with the usual DMMs. If available a good 10 M resistor would also be an option.
There is no need to get exactly 1 nA, so there is no need to have a trimmer or even more to set exactly 1 mV. If at all it would be about a few test point's (e.g. 1, 2, 0.5 mV), not to rely on only one case.
So it could be just a divider like 150K and 100 Ohms with an 1.5 V battery to generate some 1 mV, that are measured with a DMM. A 1 M resistor than sets the current. One may have to take into account the meter resistance (e.g. 10 K shunt with the µCurrent, or isolation resistance at the TIA input, e.g. some 10-100K).
I would do the test with both polarities and zero voltage, to account for a possible offset voltage (e.g. at the DUT).
Depending on the available DMM, one could use accurate resistors for the divider and measure the 1.5 V level.
For pA level currents, one usually does no care so much about the absolute accuracy down to the 0.1% level. Leakage current often change quite a lot with temperature and maybe humidity, so even +-5 % is usually OK.
Using much smaller resistance makes the DUT resistance more important and also is sensitive to offset voltages from the measuring device. It also gets more noisy.
-
Thanks. You raise a good point about needing to take into account the shunt resistance of the uCurrent Gold.
So, I guess my original test circuit was closer to what's desirable. If I use that together with a 1.5v AA battery that's divided to 1v, together with a way to divide that 1v by 1000, so as to have any value from 0.001v up to 1v, then that would be good enough.
I'll drop one stage from the circuit diagram so that the final resistance drop is 1M 1% tolerance instead of 1K 0.01% tolerance, and I'll replace the 10v precision voltage with a 1.5v AA battery. If I'm understanding you correctly, that should put us on the same page. I'll post the revision today.
-
How about this?
[attach=1]
The first stage simply creates a 1v source, divisible into 1000 lesser voltages by a 10 turn 1k potentiometer. The second stage divides that voltage by 1000 and from that voltage emits a current from a 1 megaohm resistor as the output.
The benefits over the circuit I previously used in the testing are: 1. using a 1.5v AA instead of a potentially noisy buck converter, and 2. a bit more precision on the voltage dividing, and 3. ability to dial-in anywhere from 1pa to 1na in 1pa steps using the 10-turn 1k potentiometer in the first stage.
Or, shall I ditch all the trim pots as well and replace them with just a few 1% tolerance fixed resistors? Now that we're relying on the 1% tolerance of the 1 Megaohm resistor, maybe 1% everywhere else will be good enough as well.
-
There is no real need for all the trimmers. If at all it would be 1 trimmer to get the 1 V point right.
The divider could be a single string of some thing like 50 K (possibly adjustable) - 100 K and 100 Ohms.
If possible one could consider more than 1 mV, so maybe 10 mV and 10 M or 5 mV and 5 M.
In addition to the divider I would include a switch or two, so one could change the polarity (e.g. at the battery) and also have an off state.
-
I'm still not understanding as to why having the option to change polarity is a good idea. I'm sure you're right, but I just don't understand why. Under what circumstances would it be needed?
Edit: it would be easy enough to do 10mv increments with a 10M resistor by using the 10v voltage source if that's your preference. I guess you're concerned about the 10k shunt resistor on the uCurrent Gold?
-
The polarity change is a good idea as this would an offset at the current sink. The TIA type input can have an offset voltage, so that even when connecting a resistor from ground to the input one would read a significant current. This is one downside of the TIA type current measurement: zero is with a open input not with a short or resistor to ground.
A similar, but usually smaller offset can be there with the µcurrent: the OPs offset (is small) would give a non zero reading even with no current flowing.
With both polarities one can see the offset. Using the different from + 1nA to -1 nA would eliminate the offset.
So it would be 2 single pole 2 trough switches at the supplies to change the polarity with an off state (also isolates the battery).
The possible offset is a reason for favoring a large resistance. The OPs offset can be in the 1 mV range, possibly even larger if there is no adjustment of the offset. Even than the offset would be stable only to a few 10 µV. So with a 1 mV source voltage one can get avoidable errors.
-
OK, that makes sense. It also explains why the uCurrent Gold (as compared to the picoammeter) seemed to have such a large offset in the earlier measurements.
-
Seems as though what's needed isn't a polarity reversal but instead a zeroing offset, perhaps driving the GND lower, such as what the picoammeter does in its circuitry. Am I wrong? I mean that way the test can continue as planned, whereas if the polarity is reversed, wouldn't you need some other calibration current to still test the accuracy of either the picoammeter or the uCurrent Gold?
Edit: For instance, this may be crude, but it would seem to do just that:
[attach=1]
In this case, two AA's (the 3v source in the schematic) would zero the uCurrent Gold or the picoammeter by means of the 25-turn 100K trim pot, and then the regular testing could proceed as planned.
-
The pA meter with the TIA should have some offset adjustment. Depending on the OP use this could be the adjustment pins at the OP, or just some offset added to the non inverting input.
Independent from that the polarity reversal still is useful for both the µCurrent and pA meter test. The low current meters should all read fine in both directions, so no other additions needed. The reversed polarity allows to test the scale factor even if there is some residual offset. So it can make the test of the scale factor more accurate. It is quite normal to test DMMs or similar with both polarities.
-
Here is the revision with the on-off switch and the polarity-change switch. Also, the second stage is reduced to 100x intead of 1000x voltage division so that the output resistor can be 10M.
[attach=1]
-
A 10M ohm resistor should arrive in the mail today. Once I build the device shown in the latest schematic (^^^above), what kind of reverse polarity measurements should I try to make? Also, I'm still hazy as to what to do with reverse polarity data once it's collected.
Edit: Here's the fancy 10-turn 1K potentiometer that arrived yesterday:
[attach=1]
I thought that it would "click" with each tick of the dial, but, alas, it doesn't. I'll have to make due with visual sighting only. At least it does have a turn-counter for keeping track of where it is in the 10-turns. On the back it says it is a BOURNS 3590S-2-102L and on the bottom it says 1907M. It says it has 5% tolerance and plus or minus 0.25% linearity, so it's probably the weak link in this setup. Ah, well, that's all amazon had to offer in this category. For anything better I'd have to look elsewhere and wait longer. The next step up would be 3% tolerance and 0.1% linearity for a 3549S-1AA-102B. I don't see anything better than 3% tolerance in the mouser catalog, so for anything better I'd probably have to go the resistor box route, which maybe isn't a bad idea anyway if the aim is to get single digit picoamp accuracy. OK, then, I'll be switching over to that with something that will look approximately like:
(https://images-na.ssl-images-amazon.com/images/I/611OaM%2BtcvL._SL1001_.jpg)
-
I see a limited need to fine trim the actual current. The point is more like measuring the voltage, either at the 10 mV level or the 1 V / 100 mV level (and assume the divider is accurate). Just to check the scale factor the 3 point of -1 nA , 0 and +1 nA are good enough. The 0 point is not even needed, more like a check for linearity / offset.
-
The "trick" with these kinds of tests is, you don't bother with super precise resistors... you just measure the voltage and adjust the pot while looking at your DMM, until it is "just right".
Of course the resistors have to be "good enough" that they don't drift ridiculously, but that's unlikely to be an issue here.
-
Yeah, that's what I realized after my first schematic, where I tried to come up with precise resistance values. It might have worked well enough, except that the parallel resistance threw in a wrench and created oddball equivalent resistance values. With the latest schematic, though, which uses an abundance of 25 turn-trim pots and divides voltages by using fairly high resistance values, I expect the voltage division can be dialed in much more precisely than the original approach ever could have, and with less expense too. Ironic, isn't it? I'm actually looking forward to dialing in those resistor ratios and seeing just how close I can make it. Kleinstein is probably right that I don't need this many trim pots. However, I figure having one trim-pot for every digit will make the dial-in as accurate as I reasonably can without going to too extreme. I'll know one way or the other after I attempt the dial-in.
-
If you like tweaking and tuning, knock yourself out! - it is an enjoyable art...
-
And when things don't go perfectly, it's an irritating art: when set to zero, the resistor box has 340 milliohm resistance that will creep into every measurement.
[attach=1]
Not really that big a deal, but I see no way to subtract it, except by post processing after measurement. Maybe a negative offset voltage would do it? Bah, it's not worth the bother, as we don't yet know for sure whether the noise can be suppressed enough that accurate single digit picoamp measurements are even possible. I'll just run with it.
Edit: Aha! I'll diminish that significance of the 340 milliohm by making the first stage voltage divider use higher resistances. That oughta fix it. Well, at least a lot of it.
-
And when things don't go perfectly, it's an irritating art: when set to zero, the resistor box has 340 milliohm resistance that will creep into every measurement.
(Attachment Link)
Not really that big a deal, but I see no way to subtract it, except by post processing after measurement. Maybe a negative offset voltage would do it? Bah, it's not worth the bother, as we don't yet know for sure whether the noise can be suppressed enough that accurate single digit picoamp measurements are even possible. I'll just run with it.
Edit: Aha! I'll diminish that significance of the 340 milliohm by making the first stage voltage divider use higher resistances. That oughta fix it. Well, at least a lot of it.
Don't forget, you actually just care about the voltage coming out (feeding into the big resistor), you don't care what the nominal resistance is...
If you put 5V across your resistor divider, the output voltage at the "tap" will probably be reasonably accurate.
-
And when things don't go perfectly, it's an irritating art: when set to zero, the resistor box has 340 milliohm resistance that will creep into every measurement.
(Attachment Link)
Not really that big a deal, but I see no way to subtract it, except by post processing after measurement. Maybe a negative offset voltage would do it? Bah, it's not worth the bother, as we don't yet know for sure whether the noise can be suppressed enough that accurate single digit picoamp measurements are even possible. I'll just run with it.
Edit: Aha! I'll diminish that significance of the 340 milliohm by making the first stage voltage divider use higher resistances. That oughta fix it. Well, at least a lot of it.
Don't forget, you actually just care about the voltage coming out (feeding into the big resistor), you don't care what the nominal resistance is...
If you put 5V across your resistor divider, the output voltage at the "tap" will probably be reasonably accurate.
Nah, I've already setup for just a single AA battery and drilled the holes for all the other modules accordingly, so for now ixnay on the 5v. I'lll just stick to the build plan. I can always circle back and change it later if we collectively decide it would be worthwhile to do so, after the next data collection. Anticipating that such things might happen, I'm making the build modular.
-
I like your idea though!
-
1.5V will work just as well - it was just a figure I plucked out of thin air! :D
It will be interesting to see what turns up in the wash.
-
Doh! :palm: These 7 decade resistor boards aren't designed to work like potentiometers. They have no "wiper." It would require two of them to do the job. There do exist decade potentiometers, but in blinding hindsight this little resistor board is not one of those. Such a board would be a handy thing to have, though. I wonder if there are published projects which do exactly that, complete with downloadable gerber files.....
So.... back to the 10-turn potentiometer, which I expect will be "good enough" to discover whether we have a shot at single-digit pa measurements or not. i.e. if it turns out there's too much noise even with the lunchbox test enclosure, then it won't have mattered anyway. On the other hand, if it turns out we can either suppress, eliminate, or compensate for noise enough to discriminate single digit picoamps, then I'll circle back to improve the accuracy. Until we know for sure one way or the other, though, pursuing it further now might be just wasted effort. Hopefully this next set of measurements will reveal whether we've already reached the practical limit or whether this is just the start of the next level.
-
The 10 turn pot is great. You can always put a resistor in series with it (on "top" of it) if the voltage going in is too high to divide down accurately.
I believe Conrad Hoffmann of this parish published a Kelvin-Varley divider that looked like the board you got (same construction principles with the jumpers) in his Mini Metrology Lab project.
-
The 10 turn pot is great. You can always put a resistor in series with it (on "top" of it) if the voltage going in is too high to divide down accurately.
I believe Conrad Hoffmann of this parish published a Kelvin-Varley divider that looked like the board you got (same construction principles with the jumpers) in his Mini Metrology Lab project.
Great find! You're right: on its face, it looks like a Kelvin-Varley circuit might be the missing puzzle piece for the voltage divider that we're building. The way Dave Jones describes it at time index 26:54:
https://www.youtube.com/watch?v=onqsjDJq4I0&t=1553s (https://www.youtube.com/watch?v=onqsjDJq4I0&t=1553s)
the Kelvin-Varley circuit is notionally similar to the decade resistor board, but it also has virtual "wipers" that function nearly equivalent to the wiper of a potentiometer.
Bobby Dazzler!
-
You may want to take a look at this article as well, very interesting stuff for DIY accurate voltage division:
http://conradhoffman.com/HamonResistor.html (http://conradhoffman.com/HamonResistor.html)
-
You may want to take a look at this article as well, very interesting stuff for DIY accurate voltage division:
http://conradhoffman.com/HamonResistor.html (http://conradhoffman.com/HamonResistor.html)
Here's a thread on one I built... https://www.eevblog.com/forum/metrology/anyone-else-built-a-hamon-divider/ (https://www.eevblog.com/forum/metrology/anyone-else-built-a-hamon-divider/)
Dr Frank has built a very nice one (linked in the thread). There are several other hits if you do a forum search for Hamon divider too.
-
What do you typically use it for (I mean aside from the obvious of using it to divide voltages). i.e. For what applications do you need a standalone precision divider like that?
-
If you are aiming for super high precision, adding more pots and resistors to improve resolution doesn't really help - things will never get more accurate than the DMM you are using to measure the output.
With a Hamon divider, you can make something that is better than your DMM (almost no matter how good it is!)
Typically you might use a high precision divider to calibrate and verify the operation of other instruments. Or, in this case, to accurately divide down a higher voltage (that your DMM can measure reasonably accurately) to a very low voltage (that is difficult for a DMM to measure accurately).
-
What do you typically use it for (I mean aside from the obvious of using it to divide voltages). i.e. For what applications do you need a standalone precision divider like that?
I use it for meter calibration. The key thing about a Hamon divider (unlike other dividers) is that it is effectively self-calibrating. When you set the jumpers for divide by 2, you simply compare the voltages between Input Hi - Output Hi and Output Hi - Output Lo and make sure they are identical (which is easy, to whatever resolution instrument you have).
When you switch to Divide by 10 mode, you have confidence that it is an accurate divide ratio and can then use it with a voltage reference to calibrate two decades of your voltmeter. If you only have an accurate 10V reference for example (as my situation) you can calibrate the 10V range and then use the divider to calibrate the 1V range, based on the 10V reading. With a 2 decade divider, you can get even better certainty and range, by being able to calibrate the 100V range while cross checking it with your calibrated 10V range and freshly calibrated 1V range. In this way, you can leapfrog ranges and verify / calibrate them from a single known reference.
The limitation is the input resistance of the meter, it cannot significantly load the output of the divider. In my case, my Datron meters have 10G input resistance up to 20V so there is no significant loading of the 900R effective output resistance of my 10k divider. I can calibrate the 100V range with reference to the 10V one - because I can use an ordinary 100V PSU to drive both divider and meter - the meter's 10M input resistance on the 100V range has no consequence. With the 2 decade divider that I have, I can verify the 100V range and the 1V range against the 10V reference... or the 1V range and the 100mV range. Once I have confidence in the 1V range and the 100mV ranges, I can then step down again with an arbitrary PSU (eg 1.5V battery) to calibrate the 10mV range against the 1V and 100mV ranges.
There was recently an in-depth discussion on the effect of the divide by 2 / divide by 10 switch contact resistance on the certainty of the divide ratio (talking in very low ppm terms here). I used a terminal strip and hard screw-down links to minimise contact resistance on my 10k decade (see photo in the thread I linked).
Here's the contact resistance discussion: https://www.eevblog.com/forum/metrology/absolute-divider-concept/msg2924250/ (https://www.eevblog.com/forum/metrology/absolute-divider-concept/msg2924250/)
Edit: Err, sorry, no this one: https://www.eevblog.com/forum/metrology/influence-of-switch-resistance-in-hamon-dividers/ (https://www.eevblog.com/forum/metrology/influence-of-switch-resistance-in-hamon-dividers/)
-
The limitation is the input resistance of the meter, it cannot significantly load the output of the divider.
IIRC there is no precision penalty besides the obvious "input current of the meter times output impedance of the divider". Both are easy to calculate and the effect can be accounted for.
But I also built mine with 1k5 resistors to avoid hassle ;)
-
The limitation is the input resistance of the meter, it cannot significantly load the output of the divider.
IIRC there is no precision penalty besides the obvious "input current of the meter times output impedance of the divider". Both are easy to calculate and the effect can be accounted for.
But I also built mine with 1k5 resistors to avoid hassle ;)
Doesn't that just move the hassle somewhere else? (I.e. now your reference has to supply a higher current and may lose precision instead?)
-
Doesn't that just move the hassle somewhere else? (I.e. now your reference has to supply a higher current and may lose precision instead?)
Fair point. More importantly, it caused me some unexpected hassle with the switches when I calculated how little resistance they need to have. For single digit ppm it could be hell, my target was more modest.
As for output impedance of vrefs, meh, most of them seem to be single digit mΩ. Easily the cables and connectors may dominate.
-
Build-plan executed. Assembly completed:
[attach=1]
Next step: dial-in trim pots.
Afterward: measurements!
Edit: The 10 megaohm resistor (bottom module) is supported by teflon standoffs so as not to lose current from leakage.
-
I added a BNC connector to the tester current output. Also shown is a shorting cap made by Pomona so that the output can be shorted to ground for calibration purposes:
[attach=1]
There are two pushbuttons shown in this picture. One is power-on/off, and the other reverse polarity, as per Kleinstein's earlier suggestion further above.
The test points I received from amazon were so flimsy that I didn't install them. Instead, I made some much beefier DIY test points using a loop of 22AWG wire for each test point and mounting it to the PCB.
In addition, I added a green jumper connector so that the first stage module can be isolated from the last stage module during calibration. I'm not sure that I'll ever need it, but it's there now.
This side view is better at showing the various stand-off heights:
[attach=2]
Aside from the teflon standoffs, all the other standoffs on the test setup are nylon. The supportive board that all the different modules are sitting-on/attached-to is 1/8" thick uncladd FR=4 board.
The red output current wire is UV-glued to the last teflon offset so as to provide some strain relief for the wire. I ended up using solid core 22 AWG gauge wire for the current output wire, but maybe I should have chosen stranded wire, which is more resilient, instead. Anyhow, I did order some thicker gauge wire to replace the 22 AWG wire if/when it ever breaks.
-
I made an attempt at the calibration. I was able to dial the first stage in to exactly 1v:
[attach=1]
-
So far, so good. Then I attempt to calibrate the second stage, but the trim pots have no effect and it seems stuck at 0.00950v:
[attach=1]
The target is 0.01v (for divide by 100 stage). Rather than troubleshoot and fix it, though, I decide to proceed with picoammeter measurements since it's only off by 0.5%.
-
For the sake of clarity, as I continue to describe the readings, we're going to pretend that stage 2 really did read 0.01000v and not 0.00950v.
With that in mind, the device was configured to send 1000pa to the picoammeter, which should read it as 1.0v exactly. Instead, it reads a negative 0.9611:
[attach=1]
which, except for the change in sign, is consistent with the actual (non-pretend) stage2 voltage reading.
[I would show a photo here, but the forum just now started to block my photo uploads. I guess they're too big for its liking. So, no more photos.]
I'm not sure why it reads negative, but I do have a switch-polarity button, so I push that and continue with the measurements.
Anyway, long story short, I then went on to take the following measurements:
What should be 500pa reads as 0.47930 volts (again, consistent with the actual stage 2 division).
10pa, 8.749mv
4pa, 2.5mv
2pa, 0.8mv
0pa reads as 0.6mv, so it seems something is wrong with that measurement.
Now, importantly, even though this was supposed to be the bulletproof enclosure test setup, I'm *still* able to influence the DMM millivolt readings by moving my hand toward the test enclosure, even when it is closed up. So, go figure. :-// Nonetheless, I seem able to get decent DMM measurements down to about 4pa, which means I can now read most of the single digit picoamps. At this point, to go further, I see no choice but to put the DMM inside the enclosure to do its measurements. Either that, or just be happy that it's doing as well as it seems to be. ;)
-
Is the test enclosure grounded? That may make things less sensitive.=
-
The trans-impedance amplifier is inverting. So a negative output with current from a positive source is normal.
Getting still some 0.6 mV at the output suggests that the OP in the TIA has some offset. Assuming 1 GOhms at the TIA and 10 M at the current source this would be a voltage gain of 100. So the 0.6 mV at the output correspond to some 6 µV of offset for the OP. Of cause there could also be some bias current of leakage current.
Another point could be an EMI effect, if the case is not grounded (relative to the TIA circuit). It can also be necessary to have some series resistance / inductance at the TIA input. Otherwise the TIA can be unstable with to much capacitance at the input (e.g. cable).
-
It sounds as if the picoammeter isn't completely nulled. Have you repeated the Null adjustment procedure (-ve out linked to input)? It may be that it has 'settled' a little since you built it.
-
It's not earth grounded, if that's what you mean. There is no enclosure around the DMM, so I'd wager that's the weak link. ESD protection would undoubtedly help mitigate. I just recently received 32 feet (65 pounds) of Hakko ESD matting that I haven't yet installed....
However, short of an all-encompassing enclosure, I think it's likely there's always going to be some electrostatic influence, not nothing. I mean, even a perfectionist like Marco Reps is abused by it. For a good laugh, consider time index 5:44 of:
https://www.youtube.com/watch?v=gGYXcRv3Y2w (https://www.youtube.com/watch?v=gGYXcRv3Y2w)
I'm confident we'll find a way to make this thing accurate all the way down to a single picoammeter. This set of measurements was already an improvement over the last set. I'll start by troubleshooting the second stage voltage divider, and then try again.
-
It's not earth grounded, if that's what you mean. There is no enclosure around the DMM, so I'd wager that's the weak link. ESD protection would undoubtedly help mitigate. I just recently received 32 feet (65 pounds) of Hakko ESD matting that I haven't yet installed....
However, short of an all-encompassing enclosure, I think it's likely there's always going to be some electrostatic influence, not nothing. I mean, even a perfectionist like Marco Reps is abused by it. For a good laugh, consider time index 5:44 of:
https://www.youtube.com/watch?v=gGYXcRv3Y2w (https://www.youtube.com/watch?v=gGYXcRv3Y2w)
Earth grounding the metal boxes can help (but I have also seen it increase noise... there is an element of black art to this) - suck it and see!
An electrometer is much less sensitive to the effects seen in the video when the DUT is inside a metal box, grounded by the triaxial cable that connects to the instrument. Everything has to be shielded, as you said... the DUT, the cable, and the instrument!
When you see how difficult it actually is to measure picoamps, you begin to see the achievement in measuring atto-amps! :D
https://www.eevblog.com/forum/blog/eevblog-1017-enter-the-world-of-atto-amps/ (https://www.eevblog.com/forum/blog/eevblog-1017-enter-the-world-of-atto-amps/)
-
Maybe the following would work as a cheap hack: find a multimeter with a "hold" button on it. Hook it up to the picoammeter but wrap it in aluminum foil, making a solid connection between the aluminum foil and the shared ground. After the system settles, press the "hold" button through the aluminum foil. Then unwrap the foil and read the measurement.
Anyway, I fixed the second stage. It was just a wiring mistake on my part. I'll try taking another set of measurements tomorrow and, @gyro, this next time I will be sure to re-calibrate the picoammeter too.
-
I don't know the details if the circuit. It could help against an effect from the DMM at the output, if there is some resistance (e.g. some 100 Ohms - 1K) at the TIA output, so there is less effect of EMI and cable capacity. Normally the Meter at the output should not effect the reading.
-
When it gets to the single digits on the next set of measurements, I'm going to use the Fluke 87V and turn on averaging and then step away. :-DMM I was reading in the 87V manual that the 87V can DC average over a period as long as 36 hours! :o i.e. a lot more than long enough. ;D Perhaps averaging will be a workable solution to noise at the single digit picoamp level.
I'll read the digits on the DMM from a distance using a telescope.... :-DD
-
Mission Accomplished!
If I take into account the zero current offset, use my lunchbox enclosure, quality BNC cables and BNC connectors, and turn on DMM averaging, then Gyro's picoammeter gives accurate measurements--accurate to much less than a picoamp at single digit picoamp currents! :-+
Congratulations to everyone on this thread who helped with this. :clap:
And this time I didn't even need to step away from the table. :-DD The averaging handled the noise beautifully.
What follows are the details.
First of all, I did all checks and calibrations with the equipment in their final positions.
Answering gyro's earlier question: this time I double checked the picoammeter's calibration, and the picoammeter was still zeroed from the previous time that I calibrated it, so no adjustment needed there.
Here is the current source calibration photo:
[attach=1]
It shows calibration of the second stage of the current source. Exactly 10mv are being produced, which means that 1000pa are being sent to the picoammeter, which registers that as 1.004v (equals 1004pa). You can see the 10mv is positive, but the picoammeter is showing a negative 1.004v.
I did meticulous measurements, which follow:
1000pa, -1.004v
900pa, -0.905v
800pa, -0.805v
700pa, -0.704v
600pa, -0.605v
500pa, -0.503v
400pa, -0.403v
300pa, -0.302v
200pa, -0.201v
100pa, -0.101v and -100.4mv (averaging now turned on)
90pa, -90.3mv
80pa, -80.2mv
70pa, -70.2mv
60pa, -59.7mv
50pa, -49.8mv
40pa, -37.0mv
30pa, -29.3mv
20pa, -19.7mv
10pa, -9.3mv
8pa, -7.0mv
6pa, -5.4mv
4pa, -3.3mv
2pa, -1.5mv
1pa, -0.3mv
0pa, 0.5mv
:clap: :clap: :clap:
-
Nice result, I admire your perseverance (and I suspect that you learned a lot in the process). Well done. :-+
P.S. Try uploading the image again, it looks as if a few folks have found it glitchy today. It would be a shame to lose it due to a broken link in the future.
-
Mission Accomplished!
If I take into account the zero current offset, use my lunchbox enclosure, quality BNC cables and BNC connectors, and turn on DMM averaging, then Gyro's picoammeter gives accurate measurements--accurate to much less than a picoamp at single digit picoamp currents! :-+
Congratulations to everyone on this thread who helped with this. :clap:
And this time I didn't even need to step away from the table. :-DD The averaging handled the noise beautifully.
What follows are the details.
First of all, I did all checks and calibrations with the equipment in their final positions.
Answering gyro's earlier question: this time I double checked the picoammeter's calibration, and the picoammeter was still zeroed from the previous time that I calibrated it, so no adjustment needed there.
Here is the current source calibration photo:
[attachimg=1]
It shows calibration of the second stage of the current source. Exactly 10mv are being produced, which means that 1000pa are being sent to the picoammeter, which registers that as 1.004v (equals 1004pa). You can see the 10mv is positive, but the picoammeter is showing a negative 1.004v.
I did meticulous measurements, which follow:
1000pa, -1.004v
900pa, -0.905v
800pa, -0.805v
700pa, -0.704v
600pa, -0.605v
500pa, -0.503v
400pa, -0.403v
300pa, -0.302v
200pa, -0.201v
100pa, -101v and -100.4mv (averaging now turned on)
90pa, -90.3mv
80pa, -80.2mv
70pa, -70.2mv
60pa, -59.7mv
50pa, -49.8mv
40pa, -37.0mv
30pa, -29.3mv
20pa, -19.7mv
10pa, -9.3mv
8pa, -7.0mv
6pa, -5.4mv
4pa, -3.3mv
2pa, -1.5mv
1pa, -0.3mv
0pa, 0.5mv
:clap: :clap: :clap:
Nice work! That is definitely awesome, getting to this level of measurement is not easy (as you have discovered, LOL!)
I managed to attach the photo.
-
Nice result, I admire your perseverance (and I suspect that you learned a lot in the process). Well done. :-+
I'm very happy with the results. i feel like we just broke the 4 minute mile. When I started this I thought we'd lose at least a few picoamps just to leakage alone or that the noise might be too loud to get both resolution and accuracy. I'm just impressed that for so little money we were able to measure such tiny quantities so accurately, and without even a second all-encompassing enclosure!
P.S. Try uploading the image again, it looks as if a few folks have found it glitchy today. It would be a shame to lose it due to a broken link in the future.
Done. Now it's good for all eternity.
-
Nice work! That is definitely awesome, getting to this level of measurement is not easy (as you have discovered, LOL!)
Well, that's the great irony of it all. It actually is easy once you know what to do. Now that it has been done, as documented in this thread, it will be easy for anyone else who wants to do it. The hard part was getting to the point where it was finally easy.
-
Yes, setting up the experiment is 99% of the way to a good result.
-
Follow-up question:
I'm measuring a timer circuit that supposedly consumes picoamps worth of current while it slowly charges a capacitor, but when the capacitor voltage hits threshhold the circuit turns on more and very suddenly consumes a lot more (not sure yet if its nanoamps, microamps, or maybe even milliamps) for about 40ms before the whole cycle starts over again.
To protect against cases like this one, and potentially far worse ones, would it be possible to build a protection circuit into the picoammeter that could react fast enough and disconnect fast enough to protect the picoammeter from damaging overcurrent? A reference voltage and a comparator could certainly detect the problem, but could it disconnect fast enough? On the other hand putting in an inductor might "buy time" while the disconnect is happening, but it would obviously distort time varying signals as well if the picoammeter output is being monitored by an oscilloscope. What about a line delay? Any different? Is there a solution at all? My multimeter manages to autorange without blowing up during sudden shifts, so there are clearly solutions that are "good enough" for multimeters.
How to best approach this?
-
The usual pA meter will have protection (resistor in front of the OP input). So if the circuit tries to draw more current than the circuit capacity, the resistance see will go up (e.g. from some 100 Ohms to some 100K). In most cases this current will thus not go up to much, but the circuit will see a dropping voltage. This protection can be good up to a few 100 V, how much depends on the details of the circuit.
The multimeter with shunts and for a larger current range usually have a different type of protection. There are relatively large diodes (often 2 or 3 in series each) in parallel to the shunts. So if the drop at the shunt goes over some 1 V the current would flow through the diodes in stead of the shunts. For gross over-current there is a fuse. This type of protection could also be be used for a TIA. However it would add to the error and noise. So not that good for the very low currents, but maybe OK for 10 or 100 pA resolution.
-
The usual pA meter will have protection (resistor in front of the OP input). So if the circuit tries to draw more current than the circuit capacity, the resistance see will go up (e.g. from some 100 Ohms to some 100K). In most cases this current will thus not go up to much, but the circuit will see a dropping voltage. This protection can be good up to a few 100 V, how much depends on the details of the circuit.
Well, I won't be going to hundreds of volts. So, basically, I don't need to worry about it then?
-
One should still have the resistor for protection (directly in series with the inverting input) in the circuit. Diodes to divert the residual current are usually inside the CMOS OPs.
-
Nice work! That is definitely awesome, getting to this level of measurement is not easy (as you have discovered, LOL!)
Well, that's the great irony of it all. It actually is easy once you know what to do. Now that it has been done, as documented in this thread, it will be easy for anyone else who wants to do it. The hard part was getting to the point where it was finally easy.
”It actually is easy once you know what to do.”
Heroic: tremendous vision, collaboration, skill, perseverance, and very humble too. Amazing accomplishment. Really, Really Outstanding :-+ :-+
-
The usual pA meter will have protection (resistor in front of the OP input). So if the circuit tries to draw more current than the circuit capacity, the resistance see will go up (e.g. from some 100 Ohms to some 100K). In most cases this current will thus not go up to much, but the circuit will see a dropping voltage. This protection can be good up to a few 100 V, how much depends on the details of the circuit.
Well, I won't be going to hundreds of volts. So, basically, I don't need to worry about it then?
You've already got the protection resistor - it's the 1M one on the input of the opamp. As I mentioned previously, I used a 1W carbon film type for body size, it has a voltage rating of 500V (higher on withstand voltage). At 500V input it is only putting 0.5mA through the LMC662s bootstrapped input protection diodes, which are rated at 5mA max. So basically overload capability is just based on the voltage rating of this resistor. As the resistor is inside the feedback loop, it has no effect on accuracy.
@Kleinstein: The schematic is shown in reply #27: https://www.eevblog.com/forum/beginners/static-control-requirements-for-picoamp-measurements-using-ucurrent-gold/msg3068708/#msg3068708 (https://www.eevblog.com/forum/beginners/static-control-requirements-for-picoamp-measurements-using-ucurrent-gold/msg3068708/#msg3068708) It's the same as the one in the Picoammeter thread in the Metrology section.
-
Nice work! That is definitely awesome, getting to this level of measurement is not easy (as you have discovered, LOL!)
Well, that's the great irony of it all. It actually is easy once you know what to do. Now that it has been done, as documented in this thread, it will be easy for anyone else who wants to do it. The hard part was getting to the point where it was finally easy.
”It actually is easy once you know what to do.”
Heroic: tremendous vision, collaboration, skill, perseverance, and very humble too. Amazing accomplishment. Really, Really Outstanding :-+ :-+
If someone happens to know Dave Jones, they might want to suggest Gyro's picoammeter as a video topic. It's an elegant circuit and deserves to reach a wider audience, and in that sense it fits with both the theme and the appeal of Dave Jones's other videos.
-
Nah, the circuit itself isn't that special and cerainly it isn't the first such implementation. :D He could maybe do one on TIAs though - if he hasn't done one already.
-
I think you guys are onto something very good and deserve a lot credit for figuring stuff out at an impressive level in terms of small signal measurement both in theory and practice, not to mention pretty cost-effectively. Seems like it's subject worthy of Dave doing a video - he could reference your work, add his own thoughts, and discuss it from an educational standpoint. In any event, you have covered a lot of very good ground. Congrats on all the good work.
-
Nah, the circuit itself isn't that special and cerainly it isn't the first such implementation. :D He could maybe do one on TIAs though - if he hasn't done one already.
Either way.
A big difference now exists between your circuit though and most other picoammeter circuits on the internet, and that difference is that yours is now independently vetted and shown to be accurate--by this very thread. There's value in knowing that if you build a particular circuit, it will perform well, as compared to other circuits which may be similar but whose tested performance is unknown.
-
Thanks. :)
He could probably do something on the lessons in effective screening - and also the actual measurement and pA test sources (probably the biggest areas of work you have done in this thread).
-
I found this very useful:
http://www.hep.fsu.edu/~wahl/phy3802/expinfo/labmanual/intlabdoc/instruments/keithley/Low_Cur_Meas_AN.pdf (http://www.hep.fsu.edu/~wahl/phy3802/expinfo/labmanual/intlabdoc/instruments/keithley/Low_Cur_Meas_AN.pdf)
https://download.tek.com/document/LowLevelHandbook_7Ed.pdf (https://download.tek.com/document/LowLevelHandbook_7Ed.pdf)
-
I think the people here are still active on the forum or atleast i hope so. I am building this circuit to test the picoampere meter in an exhaust gas analyzer. I am still very much a beginner and have some questions on how to put the input on the analyzer.
The situation: there is an electrode that creates a small voltage that is going to the analyzers pcb with an bnc cable. In my understanding the input on the pa meter flows the currunt to ground/negative of the circuit under testing. So i cant put it in series with the positive terminal of the voltage source since this would cause a short circuit since there isnt a load/consumption present. the pa meter of gyro needs to be behind the load, correct?
now you could put it in series on the negative coming from the pcb going to the voltage source. so in series of the bnc shielding. my concern is that static discharge gets picked up on the shield and causes an inaccurate measurement. the shielding is of course to protect the inner cable.
so my idea was to create another shielding around the shielding of a regular bnc cable to shield the negative of the circuit from static noise. my question am i thinking the correct way and would this work. if not has somebody other ideas. the pcb needs to be connected since the idea is to compare the pa readings of gyros circuit to the industrial grade one in the analyzer. I dont have room for mistakes since the electrode is already several thousand euros and the analyzer way more. thats why i wanted to get my theory checked by people who know the inner workings of gyros circuit.
-
The pA meters and likely also the input at the analyzer are usually as a TIA and this one end at the power supply of meter.
For high currents it is easy to just have 2 ampmeters in series. This does not work that well with the TIA type meters. One at least has the shielding issue.
At the pA level it is more a test source and 1 meter and than exchange the sources and meters. For comparing 2 meters directly one would have a large resistor between the meters inputs (center wire) and a voltage source between the grounds (shield). Here the voltage source and resistor make up a special source that can serve the 2 meters.
-
thanks for the swifty reply. I just fineshed the meter and it works great. okay so the safest option would be just to calculate the resistance load that the analyzer has since the voltage and pico amps is known from the analyzer. then feed the analyzers voltage source trough a resistor to gyros meter. this compares the two. to trick the analyzer i can just feed the pcb the known voltage from a different source.
thanks for the help i greatly appreciate it.