Author Topic: How to increase an analog Galvanometer sensitivity ?  (Read 6427 times)

0 Members and 1 Guest are viewing this topic.

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
How to increase an analog Galvanometer sensitivity ?
« on: February 20, 2018, 02:02:31 pm »
Recently I bought a NOS galvanometer (Photo).

Measured thru DMM the coil winding resistance (Rg) is 120 Ohm, and at the max deflection from the needle resting middle position to max right is 500 uA, and vice versa to left as well with reversed polarity.

Any help would be appreciated on how to increase the sensitivity, say down to 5 uA or even better 50 nA or any other scale that is reasonable ? I mean through an extra front end circuit.

TIA

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #1 on: February 20, 2018, 02:45:09 pm »
Electronically?  Or, Mechanically?

For electronically, just use an op-amp circuit & choose the appropriate gain resistors for a gain of 10, or 100, or even more.

 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #2 on: February 21, 2018, 05:53:08 am »
Ok, thank you.  :-+

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #3 on: February 21, 2018, 10:13:52 am »
I suspect that the meter you have there isn't actually intended as a galvanometer (other than any moving coil meter being a galvanometer). I think the 'g' marking actually refers to acceleration, the movement probably comes from an acceleration or vibration meter. Galvanometers are usually in the small uA sensitivity range, eg. 25-0-25uA (the center zero version of a 50uA meter). Yours appears to be a more robust center zero version of a 1mA meter.

Regardless, it is a center zero meter, so with some opamp gain added, as BrianHG suggests, probably configured as a transimpedance amplifier, you can make a nice center zero current meter that goes down to the pA range if you want. The multiple scales will be useful for range selection.
« Last Edit: February 21, 2018, 11:02:28 am by Gyro »
Best Regards, Chris
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #4 on: February 21, 2018, 11:53:51 am »
I suspect that the meter you have there isn't actually intended as a galvanometer (other than any moving coil meter being a galvanometer). I think the 'g' marking actually refers to acceleration, the movement probably comes from an acceleration or vibration meter. Galvanometers are usually in the small uA sensitivity range, eg. 25-0-25uA (the center zero version of a 50uA meter). Yours appears to be a more robust center zero version of a 1mA meter.

Wow, thank you.  :-+

I didn't realize that the "g" has other meaning rather than for "G"alvanometer.

But maybe you're right, as the needle is quite stiff compared to other moving coils meters I have like Simpson, Yokogawa and other good quality ones (not cheap China meter). As its doesn't move a lot (not more than one tick mark left or right) when I move or even swing the meter physically, while others like moved loosely up to like half deflection level, subjective I know.

Is this kind of behavior of a robustly designed moving coil meter ? I'm a total noob when it comes to moving coil measurement devices, it just I've been exposed to these things just recently.  :-//


Regardless, it is a center zero meter, so with some opamp gain added, as BrianHG suggests, probably configured as a transimpedance amplifier, you can make a nice center zero current meter that goes down to the pA range if you want. The multiple scales will be useful for range selection.

Pardon for my noobness, I can google for "transimpedance amplifier" circuits, its just say I want down to pA level, is there any insight, tips or tricks on building such circuit ? As this will be one off, I don't mind a bit niche type of project build.  :P


Offline ZomBiE80

  • Contributor
  • Posts: 39
  • Country: fi
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #5 on: February 21, 2018, 12:01:20 pm »
Electronically?  Or, Mechanically?

For electronically, just use an op-amp circuit & choose the appropriate gain resistors for a gain of 10, or 100, or even more.

Do you understand principles of galvanometer?
 

Offline ZomBiE80

  • Contributor
  • Posts: 39
  • Country: fi
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #6 on: February 21, 2018, 12:03:30 pm »
Use something like brake cleaner to rinse away any greasy-like residues too.
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #7 on: February 21, 2018, 12:12:48 pm »
The OP has a meter with +/- direction with a letter 'g' printed on the face-plate.  He want's to use it as a sensitive current meter for extremely low currents.  He can easily use an op-amp circuit to do so.

Here is an example circuit:
Note to change R2 to 1K, or 10K, or 100K to increase the sensitivity/gain of the circuit.
The input impedance is 10 ohm within the range of the circuit.
There are better op-amps than TLO61, just use a fet input and low voltage offset opamp.
I chose +/-3v as the power supply, basically 3 lithium coin cells, but, you can use 2 9v batteries to get +/-9v.

For below Picoamp and Fentoamp, this circuit will not suffice.

Actually going for higher voltage supply, you will need to change the r3 to 1k and the trimpot to 10k.
« Last Edit: February 21, 2018, 12:38:44 pm by BrianHG »
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #8 on: February 21, 2018, 12:27:06 pm »
Pardon for my noobness, I can google for "transimpedance amplifier" circuits, its just say I want down to pA level, is there any insight, tips or tricks on building such circuit ? As this will be one off, I don't mind a bit niche type of project build.  :P

Knock yourself out  :)   https://www.eevblog.com/forum/projects/picoammeter-design/  There's no reason why you can't use a moving coil meter instead of a DVM, switching in lower value feedback resistors would reduce the sensitivity (into the nA range)... https://www.eevblog.com/forum/projects/picoammeter-design/

Forum search will doubtless bring up other circuits. Here are a bunch of variations on the theme too: https://duckduckgo.com/?q=picoammeter+circuit&iar=images&iax=images&ia=images

Quote
But maybe you're right, as the needle is quite stiff compared to other moving coils meters I have like Simpson, Yokogawa and other good quality ones (not cheap China meter). As its doesn't move a lot (not more than one tick mark left or right) when I move or even swing the meter physically, while others like moved loosely up to like half deflection level, subjective I know.

Is this kind of behavior of a robustly designed moving coil meter ? I'm a total noob when it comes to moving coil measurement devices, it just I've been exposed to these things just recently.  :-//

Yes, lower sensitivity meters can afford stiffer hairsprings and generally more robust construction than, say a 50uA meter (50uA is about the practical limit of what can be achieved in mass production panel meters with bearing friction etc).
« Last Edit: February 21, 2018, 12:35:11 pm by Gyro »
Best Regards, Chris
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #9 on: February 21, 2018, 12:34:30 pm »
Electronically?  Or, Mechanically?

For electronically, just use an op-amp circuit & choose the appropriate gain resistors for a gain of 10, or 100, or even more.

Do you understand principles of galvanometer?
Yes.
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #10 on: February 21, 2018, 12:36:33 pm »
Use something like brake cleaner to rinse away any greasy-like residues too.
???? No... What?

If you were commenting on my mechanical fix, I was talking about changing the spring, or re-winding the coil for a new current sensitivity level.  This has been done before.
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Best Regards, Chris
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #12 on: February 21, 2018, 04:23:00 pm »
The OP has a meter with +/- direction with a letter 'g' printed on the face-plate.  He want's to use it as a sensitive current meter for extremely low currents.  He can easily use an op-amp circuit to do so.

Brian, I'm sorry to say, well intentioned as you are, that you are very much on the wrong track.

Quote
Here is an example circuit:

Note to change R2 to 1K, or 10K, or 100K to increase the sensitivity/gain of the circuit.
The input impedance is 10 ohm within the range of the circuit.
There are better op-amps than TLO61, just use a fet input and low voltage offset opamp.
I chose +/-3v as the power supply, basically 3 lithium coin cells, but, you can use 2 9v batteries to get +/-9v.

For below Picoamp and Fentoamp, this circuit will not suffice.

Actually going for higher voltage supply, you will need to change the r3 to 1k and the trimpot to 10k.

Yes, he wants to measure lower currents than the meter natively does (+/-500uA) but you are offering a solution with a 10 ohm input impedance. He needs to increase the input impedance, not reduce it. Furthermore, you've just given him a design for a voltage input circuit, not a current input circuit.

Here's a very simple design for a current input amplifier that amplifies by 10 (for the current meter's 120 ohm coil resistance). It's produced in LTSpice, the forum's favourite circuit simulator, and the LTSpice .asc file is also attached.



Rsense is simply set in ratio to the meter winding resistance, in the amplification ratio you want. In this case the amplification ratio is 10, the meter winding is 120 \$\Omega\$ so Rsense is simply 1200 \$\Omega\$. This results in a complete system with a full scale swing of +/-50 uA. If you want a current gain of 100 then Rsense becomes 12k and so on.

The op amp is only being asked to operate at unity gain, whatever the current gain you call for; so it will have a non-critical gain bandwidth. Op amp input impedance (i.e. input bias current) does matter, so it is necessary to pick a JFET/CMOS input amplifier but a cheap one will do. A TL071 or similar would fit the job in hand fine.

This is just a sketch, not a complete schematic and obvious consideration needs to be given to power supplies, input protection and overall stability. This is probably perfectly stable into the meter impedance given but you never really can tell until you try it with real electrons.

As the maximum input voltage is set by the full scale voltage of the meter (+/-500uA * 120 \$\Omega\$) = +/- 60 mV, input protection is easy, a pair of parallel diodes to limit the input voltage across Rsense to 1 diode drop will do the job admirably. Diode reverse leakage currents may be an issue, if so, substituting a pair of diode wired small signal bipolar transistors for the diodes will keep leakage to the level of a few picoamps.

Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #13 on: February 21, 2018, 04:44:47 pm »
Furthermore, you've just given him a design for a voltage input circuit, not a current input circuit.
My circuit will still function as a current input since the - input of the opamp will always appear as a short to GND. (Unless you exceed the op-amp's output feedback capability)  So the system still appears to be measuring current.  Though my circuit has a 10ohm input impedance, you can use different resistor values like 120 ohm for R1, and 1.2k for R2.  In this setup, I get some gain on sensitivity, but keep a relative 120 ohm input impedance equivalent of the original meter, though, I do introduce some gain error and some noise, which wouldn't be seen on a mechanical meter.  Also, adding a cap in parallel on R2 in my circuit will also make a nice low pass filter.

Yes, your circuit is simpler and has a strict input impedance, but no protective isolation from the op-amp's inputs and even a direct potential path to potentially blow the opamp and maybe the meter.  (the protection diodes will take care of that)  But, yes, you are right, if you don't want an equally lower input impedance, like the meter's original 120ohm, your circuit is the better, much easier/simpler way to go.  Especially if you are going down to the picoamp range.
« Last Edit: February 21, 2018, 04:56:52 pm by BrianHG »
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #14 on: February 21, 2018, 05:17:48 pm »
Furthermore, you've just given him a design for a voltage input circuit, not a current input circuit.
My circuit will still function as a current input since the - input of the opamp will always appear as a short to GND.

I'm sorry, but it just doesn't behave how you think it does. Remember, the OP wants to increase the sensitivity of the meter.

Here's the results of simulating your design. You don't specify how to set the variable resistor so I've simulated for minimum, mid-scale and full-scale: almost-zero, 500 \$\Omega\$ and 1k \$\Omega\$. In every case, with a current source, your circuit acts as an attenuator - the meter gets less current than you put into the input.

Edited to add:

Yes, your circuit is simpler and has a strict input impedance, but no protective isolation from the op-amp's inputs and even a direct potential path to potentially blow the opamp and maybe the meter.

I did say it was a sketch and I didn't put in any protection.



X is current input, Y is current through the meter. Each slope is for each simulated variable resistor value.

« Last Edit: February 21, 2018, 05:25:33 pm by Cerebus »
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #15 on: February 21, 2018, 05:44:52 pm »
I'm sorry, but it just doesn't behave how you think it does. Remember, the OP wants to increase the sensitivity of the meter.
Ok, perhaps, you are looking at my circuit and values the wrong way, so, make this change in the simulation:
Make R1 = 120 Ohm.
Make R2 = 120 Ohm.
Make R3 and RP1 = 0 Ohm.

The current you feed in will now match the current going into the 120 Ohm meter, correct?

Next, make 1 change, make R2 1.2k and cut the current going into the input by 1/10th.  The meter should now see 10x the source current.  Correct?

Now, change R2 to 12k and cut the current going into the input by 1/100th.  The meter should now see 100x the source current.  Correct?

Yet, all this time, the source impedance will always be 120 ohm.
« Last Edit: February 21, 2018, 05:47:00 pm by BrianHG »
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #16 on: February 21, 2018, 06:42:32 pm »
I'm sorry, but it just doesn't behave how you think it does. Remember, the OP wants to increase the sensitivity of the meter.
Ok, perhaps, you are looking at my circuit and values the wrong way, so, make this change in the simulation:
Make R1 = 120 Ohm.
Make R2 = 120 Ohm.
Make R3 and RP1 = 0 Ohm.

The current you feed in will now match the current going into the 120 Ohm meter, correct?

Next, make 1 change, make R2 1.2k and cut the current going into the input by 1/10th.  The meter should now see 10x the source current.  Correct?


I'm not trying to make a fool out of you here, just trying to be helpful to both you and the OP but I'm not very happy with the way this is going. Saying "perhaps, you are looking at my circuit and values the wrong way" and then selecting completely new values looks very much like trying to make me look like a fool. I hope that is not the case. I suspect that you've got some fundamental misunderstanding of what is going on here - I'll try to put you back on the road

Let's do as you ask and change all the component values. I'll cut out the middle man and not post simulations for this part but, yes with R1=120 and R2=120 you'll get 1x current gain, with R1=120 and R2=1200 you'll get the desired effect of 10x current gain into the meter. But you've still got a voltage amplifier there, not a current amplifier. It works now, but not by your choice of R1, or its ratio to R2 but by your choice of R2. And I can prove it:

Here's a new variation on this circuit:



Here are it's input (X) and output currents (Y):



Yes, your input resistor has gone and it's still producing the desired result. That's a transresistance amplifier, and it's the best circuit for this task - provided that you have an op amp that can output at least as much current as your current input is plus the current in your load (in this case the 120 ohms of the meter). It produces a voltage output that causes a negative current feedback to the input summing node. The effective current gain, from the meter's point of view, is R2/Rmeter.

Here's the full-blown version with some protection added. Where it falls down is that the protection circuitry offers lots of protection from overvoltage, but fails the second that I(R2) isn't enough to balance the input current causing the voltage on the summing point to go out of bounds. That's why I went with the other option - it's simpler, easier to understand, with protection it's much less fussy of op amp and the power available to provide protection current.



Here's the simple amp with protection circuitry and a 1,000,000% overload. You can see it behaves pretty benignly. The same overload on the transresistance setup above would require the op amp to output 500mA to remain protected - only the very beefiest power op amps can manage that.



Sorry, some frantic editing with the images there as they'd got muddled.
« Last Edit: February 21, 2018, 06:49:30 pm by Cerebus »
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 
The following users thanked this post: BravoV, BrianHG

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #17 on: February 22, 2018, 03:35:35 am »
Thanks guys, really appreciate the replies. Also Cerebus, thanks for the trouble drawing the circuit.  :-+

Give me some time to chew & digest all of these.

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #18 on: February 22, 2018, 04:57:28 am »
Thanks Cerebus, I was not saying that anyone here took me for a fool.  My circuit was a cut and paste from another project and designed to be played with on a bread board.

On the other side, having a 0 ohm going into the - opamp input delivering a theoretical near 0 ohm impedance if the load is within the op-amp's drive capability, I am so accustomed to having that resistor there at the negative input that I never guessed you would ever use an opamp like that.  It seems obvious now.  Something to play with and see if I can tame the noise in such a circuit with either a slow op-amp or a small few pf cap here and there.  Thanks...
 

Offline Cerebus

  • Super Contributor
  • ***
  • Posts: 10576
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #19 on: February 22, 2018, 09:27:36 am »
Thanks Cerebus, I was not saying that anyone here took me for a fool.  My circuit was a cut and paste from another project and designed to be played with on a bread board.

On the other side, having a 0 ohm going into the - opamp input delivering a theoretical near 0 ohm impedance if the load is within the op-amp's drive capability, I am so accustomed to having that resistor there at the negative input that I never guessed you would ever use an opamp like that.  It seems obvious now.  Something to play with and see if I can tame the noise in such a circuit with either a slow op-amp or a small few pf cap here and there.  Thanks...

In a transresistance amplifier, contrary to one's usual expectation, that a high feedback resistance will lead to high noise because of Johnson (thermal) noise, in fact high resistance leads to lower noise because of lower current noise into the summing node. Searching for "noise in photodiode amplifiers" or "noise in transresistance amplifiers" or "noise in electrometers" ought to turn up plenty of helpful stuff.
Anybody got a syringe I can use to squeeze the magic smoke back into this?
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #20 on: February 23, 2018, 06:23:00 am »


Cerebus, thanks again for the circuit, do you have any suggestion on the op-amp ? Type ? Accuracy and etc ? Understand this moving coil has limited accuracy, but since this is one off project, as I mentioned, I don't mind its a bit expensive for the supporting components.  :P

« Last Edit: February 23, 2018, 07:33:29 am by BravoV »
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #21 on: February 23, 2018, 07:42:11 am »
When selecting an op-amp for this circuit, your #1 concern is the lowest possible input current.

My other priorities would number here:
#2 Should be a combination between offset voltage and noise.  If you want to get fancy here, get a chopper stabilized op-amp.
#3 should be power supply voltage & power consumption.

The least important thing about your op-amp selection is op-amp bandwidth.


The input current of these 2 is less than 0.003pA (yes, less than 3 fento Amps)

Very cheap, but requires +/- 2.25v, or 5v minimum to operate: https://www.digikey.com/product-detail/en/texas-instruments/LMC6041IMX-NOPB/LMC6041IMX-NOPBCT-ND/3527044
For this guy, I would power it with 2x lithium coin cells, or +/-3v.  (Use the center of the 2 batteries as 0v)  Read the full datasheet, they give advice on how to achieve the low 2fA capability of this opamp.

A bit more pricey, but begins to operate at +/-0.9v or 1.8v : https://www.digikey.com/product-detail/en/texas-instruments/LMP7721MA-NOPB/LMP7721MA-NOPB-ND/1835472  WARNING: Non standard pinout on this guy!!!
For this guy, I would power it with 2x AA, or AAA batteries, or +/- 1.5v.  (Use the center of the 2 batteries as 0v)

« Last Edit: February 23, 2018, 08:00:44 am by BrianHG »
 

Offline BravoVTopic starter

  • Super Contributor
  • ***
  • Posts: 7547
  • Country: 00
  • +++ ATH1
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #22 on: February 23, 2018, 08:14:18 am »
BrianHG, thanks.

As your priorities list, CMIIW, the LMP7721 nor LMC6041 are not chopper stabilized, does it matter ? Impressive input bias current indeed.

Also coincidentally I found few unused LMC662 in my drawer with 4 pA input with 6 mV offset, is this decent ? Its a dual op-amps configuration.

How about the resistors involved here ?

About the power, thanks for the button cells tips, but as I said, for the dual lanes + and - rails, I wouldn't mind even to use 2 x 9 Volt batteries if needed. Heck, even two 18650 Li-Ion cells, as I have plenty of these.

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #23 on: February 23, 2018, 08:16:54 am »
Now this guy is a chopper-stabilized opamp.  It has a much higher input current of 1pA, but, it's offset voltage is +/-0.5uV instead of the above listed amplifiers whose offset voltage is +/-1000uV for the first one, +/-50uV for the lower voltage one LMP7721.

https://www.digikey.com/product-detail/en/linear-technology-analog-devices/LTC2055CMS8-PBF/LTC2055CMS8-PBF-ND/755205

Your best pick between both worlds is the one with the weird pinout, LMP7721, though it is a 5$ opamp.

Your 6mV offset opamp, or, +/-6000uV is much higher.  The slight offsets will shift the center of your meter which you might need to mechanically adjust with that screw.  Except for the chopper-stabilized, this also drifts with op-amp temperature.  So, with your 6000uV op-amp, the LMP7721 will be 100x more centered and 100x less effected by temperature drift while the LTC2055 will be 10000x better centered...  However, you asked for 50na sensitivity and your meter peaks in each direction just below 100mV.  Your 6mV error on your opamp will potentially shift the reading as much as 5% in either direction.  (Not exact math and you are using the worst case scenario spec for your opamp..)  For the typical error on the LMP7721, this error drops to 0.05% in either direction.  Well, the chopper, ok, you get the idea.  Next comes the current at the input pin.  For 50nA full range, 5nA for 1 digit on your meter, maybe 1nA would have a visible shift on the meter at this scale, a 4pA load should do very little.

Use your LMC662 opamp, you will have no problems.  Your only gain would be to use a chopper-stabilized opamp if you seriously don't want your center 0v to ever drift.
« Last Edit: February 23, 2018, 08:43:42 am by BrianHG »
 

Offline Gyro

  • Super Contributor
  • ***
  • Posts: 9410
  • Country: gb
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #24 on: February 23, 2018, 09:43:18 am »
Yes, go with the LMC662 - its typical input current is also in the 3fA region. It's cheap (a bargain for the input current) and easy to wire - In DIP pkg you can lift the input pin and air wire to avoid board leakage effects. Being a dual, you can use the other opamp to provide a center bias for single supply operation and also null its offset. You can achieve very sub-pA performance with it and trim to a stable zero offset.  https://www.eevblog.com/forum/projects/picoammeter-design/msg790045/#msg790045

The problem with choppers is noise - not the output noise but the noise that its input chopping action injects back into the circuit that you are monitoring.
« Last Edit: February 23, 2018, 09:57:13 am by Gyro »
Best Regards, Chris
 

Offline BrianHG

  • Super Contributor
  • ***
  • Posts: 7661
  • Country: ca
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #25 on: February 23, 2018, 12:37:23 pm »
The problem with choppers is noise - not the output noise but the noise that its input chopping action injects back into the circuit that you are monitoring.
Thanks to the series input resistance in the circuits, you can cap filter that with a 1-10nf.  I would also decrease the 1m in the last circuit, the one using the negative input, to 100k.
 

Offline Tj138waterboy

  • Regular Contributor
  • *
  • Posts: 103
  • Country: us
Re: How to increase an analog Galvanometer sensitivity ?
« Reply #26 on: April 15, 2020, 11:33:26 am »
Sorry to beat a dead horse but this topic has made me question the layout difference in voltage sense and current sense depending on application. I Have read through most of the Ucurrent and picoammeter posts and now my brain is mush. I will do my best to explain my goal which is what I would assume is needing a dual purpuse whetstone bridge null as well as voltage reference/hamon divider null.
My current setup is using an ad620a INA due to the 1 resistor gain gontrol. I know this isn't best specs for either purpose but will be easy enough to replace.
Where im having issues is that I purchased a defunct Fluke 931b rms differential voltmeter that I lost interest in restoring so am going to repurpose but want salvage what I can to make a new possibly improved board.
The center 0 panel meter measures 3.16k across terminals
With on screen print showing volts-% deviation. Im unable to find any free online schematics but assume that the rotary switch decks are similar to the 8xx series fluke.
If suppose I gut the whole meter leaving only the panel meter and only using to null whetstone bridge "externally", and comparing 10v references or hamon dividers, which circuit would suit both needsor would I need both. I already jumped the gun by tying 2 10v ref outputs to the + - of the opamp just to test the gain amd had bad connection which pegged the needle.  I would like to know if its better practice to have maybe 4.6k resistors in series with unbuffered vref outputs to limit current or is there some form of magic with the npn transistors in the schematic on previous page that would limit current to stop needle from hitting limits hard.  Will post schematic shortly. Also side note I have a cheap ebay 35 mV galvo that measures 100r across terminals that could be used as well.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf