EEVblog Electronics Community Forum
Products => Test Equipment => Topic started by: joeqsmith on December 15, 2018, 03:59:10 am
-
Having a little fun with the PC and an ADC. The ADC is a bit sensitive to outside fields and I ended up putting it into the cookie tin with a towel over it. The reference is not very stable and would need to be in a temperature controlled housing. The attached shows how the simple ADC compares with the Gossen Ultra and my HP 344401A using the same 1.9Meg Caddock TK 10PPM part I have been using.
Second graph shows all of the meters I have looked at. I let the ADC run for a fair amount of time and it does eventually start to calm down.
-
Which software is this? link please? (I looked at the UT181a product page, but that looks different)
Thanks.
-
Which software is this? link please? (I looked at the UT181a product page, but that looks different)
Thanks.
Its something I put together for the 181A. If you want to know more about it, just go to my YT channel (link in signature) and it was the last video I made.
-
To use the ADC as a simple volt meter, I used a Caddock USVD2 for divider network. These have a tempco of 2 ppm. Again, the ADC is in the cookie tin with a towel and was allowed to warm up before I collected any data.
http://www.caddock.com/Online_catalog/Mrktg_Lit/TypeUSVD_HVD.pdf (http://www.caddock.com/Online_catalog/Mrktg_Lit/TypeUSVD_HVD.pdf)
I added support to my software to support my the HP34401A so the two could be displayed on one graph. I used the Fluke 731 to do a simple 2 point alignment.
The attached graph shows the two volt meters attached to my large 1KV power supply, set to roughly 100 volts. The HP34401A is the Red trace and the deltasigma is White. I started the test early and you can see the furnace cycling. As I walk into the room, we can see a fairly large change in the supplies output voltage as well.
******************
I should mention that the screen shot was luck. They are tight but not that tight.
-
My little lab has stayed open and we can clearly see the two meters continue to track very nicely while the supply continues to drift.
-
Had to open the cookie tin to change over to the reference standard. Short clip sped up showing the initial warmup. It will take some time to settle again...
https://youtu.be/P49chFp0-58
-
I let the board's temperature settle out and realigned it to the reference then restarted the test.
The HP is very stable compared to my little experiment which wanders all over the place.
-
The Fluke standard was set to 990uV but I had aligned the ADC thinking it was at 1mV. The HP is close to matching the reference.
I had left the cookie tin and towel and just changed the output of the reference. You can see the HP with the Fluke reference are very stable while the ADC test circuit has a poor drift. I suspect this is all thermal.
-
Into the meat packing box. I use the HP34401A to measure the boxes thermistor but we know the reference is stable and how much the ADC circuit was drifting.
I plan to run it at 35. Just enough so I can get the temperature stable. You can see how badly it drifts as I ramp the temperature.
-
What is the adc/reference?
-
Temperature was 35 and let it settle to within +/-0.05 deg. I then realigned the ADC and started collecting. The test was ran for 11 hours and the temperature remained within this window for the duration of the test.
Once again, roughly 70uV. There are no gain stages at this time. It may be time to try adding a low noise front end to it.
-
My old HP34401A still has me by a fair margin but its moving in the right direction.
Don't ask me why I was posting a uV. Not enough sleep. The Fluke 731B is still set to 1mV. Back to the cookie tin and towel. Let it run more than double the time (several hours).
-
Tracking the drift of my vintage HP power supply while running in current mode. Switching from 170mA down to 5mA. Red is my HP34401A bench meter.
-
Measuring the voltage and current to my UNI-T UT61E with a dead 9V. Dip is the backlight and then watching the battery recover.
http://www.ti.com/lit/ds/symlink/ref5025.pdf (http://www.ti.com/lit/ds/symlink/ref5025.pdf)
-
Looking at the UT61E's power draw for 4 hours w/ the same 9V battery. The wide increases are the backlight. I am not sure what the spikes are. The seem fairly periodic.
-
No more battery. Ramping down the UT61E's supply voltage until the meter drops out.
***
Two signals combined with current on graph set to 1000X (mA rather than Amps).
-
Project was getting to be a bit of a mess to work with. Put together a simple insulated and temperature controlled, shielded case. Shown with the cover being removed and installed a few times.
-
The temperature is much better now but the case could use a bit more insulation. I'll let it run for several hours to get an idea how the lab temperatures are going to effect it.
-
It's a big improvement but it's still too sensitive. I tossed a towel over the whole setup and it causes a fairly large disturbance.
-
First graphs shows switching from 10K to 100 to 1K ohms. The second is zooming into the 1K ohm area. Still a fair bit of drift.
-
Measuring the Fluke 731B and continuing to to make changes. Also included is my HP 34401A. Sample rate is 1Hz, or 2.8 hours per run.
-
With the circuit now somewhat stable, I let it run along with the HP34401A for roughly 7 hours. Standard deviation is still not as good as the HP.
Wrong plots.....
-
For the diode test the voltage and current it can supply is a bit higher than your typical meter.
Here the Brymen's battery leads are connected. I select that the meter uses a 9V alk battery. The voltage is then swept to 9V, held for some time then ramped down. Both current and voltage profiles are recorded. The nominal current and eventually the cutoff voltage are measured. I am thinking it could then provide some figure of expected life. The cutoff may be different for each meter. The Brymen will throw up some really strange numbers with a dead battery. Maybe just detect the first sign of a slope change. I am thinking the battery database would be common. The idea would be to automate this measurement to a standard.
Just playing for now.
-
Interesting thread :-+ Thank you joeqsmith 8)
:popcorn:
-
Having a meter like this being controlled from a PC means we can automate various tests. For example, I made a simple menu to walk through a battery life test for the handheld meters. In the following, the Brymen BM869s is measure twice. The voltage is set, then the meter is turned on. The backlight is then turned on. All the functions of the meter are then selected.
I then repeat the test for the Fluke 189. The 189 has two backlight settings. We can also see the current draw is not flat like the Brymen.
The software measures the nominal current, the backlight current and the cutoff voltage. It also calculates the battery life. For the BM869s which uses a single transistor battery, it estimates 80 hours. For the 189 with it's 4 X AA, 127 hours. These are just using some Wiki numbers for alkalines. All of this data would be saved for later.
I ran one other meter for fun that also uses a 9V. There are some major differences. Of course battery life for something you use in the lab may not be as high a priority as say measurement performance.
-
I was once told by someone wiser that Sigma-Delta ADCs are not really designed for constant (DC) voltages and that manufacturers usually provide absolute error numbers for DC applications.
So my (limited) understanding is that Sigma-Delta ADCs are really designed for alternating voltages.
-
Ended up swapping out the Linear Tech reference for a Thaler 0.6 ppm. The temperature is just starting to settle. I'll let it run overnight and we can see if it improves.
-
Showing the initial 5000 seconds after a half hour warmup. It's a bit early to say how much, if any, the new reference improves it.
-
Comparing the 34401A for 7hrs. While the standard deviation is higher for the HP, I just hadn't let it warm up long enough. If we ignore the first 2.8 hours of data, the old HP does a much better job.
-
Repeated a 25 minute sweep using the 1.9Meg Caddock using the new reference. First comparing it with all the meters I ran this test on, then deselecting some of the flyers and finally just with the HP34401A.
-
While I have been using this meter I was never happy with the results. The last few days I have been thinking about how it could be improved. The dominant problem has been temperature drift. To combat this, the box has a fair amount of insulation around the sensitive area with a closed loop heater control. For all that effort, the results were not very good. A few ppm hear and there adds up fast.
I am going to attempt to flatten it out using a BAV199 and resistor, similar to what I showed with the UT61E. In the attached graph, the initial ramp is with the boards temperature increasing 2.5C with no compensation. The longer section was with my first attempt at taming it with the BAV199 and a 10K ohm. It will take some trial and error but the initial results look promising.
Dave should do a video showing some basic compensation techniques. I've been a fan of using 4148s for many years as a cheap method.
***
The office can get fairly warm and to keep the internal temperature regulated, it will need to run above that. The systems drift is not linear. Next step is to sweep it over a wide range and plot drift and see is there is an area that looks like it may be easier to tame and be at least 10 degrees over my office on a hot day. It will take some time to run all these sweeps. It took several attempts to pick the combo I used on the UT61E.
***
To compensate this circuit, I would normally use my HP34401A but the Brymen is more than stable enough in the 500K mode to detect the drift trends.
That's about the only thing that is going smooth so far...
-
After several attempts, its still not very good. The two yellow boxes are roughly the same temperatures being swept. The lower is the last attempt. This is again looking at the Fluke 731B's 1V reference and again, the meter is not being aligned between mods, which is why the vertical scales do not match.
The temperature control has also been tightened down a bit which will help with the system errors.
Another option I am thinking about is to just run a temperature sweep and do a fit to this data. Just forget all this hardware mess and do it in software. It's a one off meter anyway.
-
Unless I missed it, this would be much more interesting if you you would tell us which ADC and reference you are using - all I could see was that you have swapped an LT reference for a Thaler 0.6ppm. Is this a murder/mystery thread where we have to work out what you are testing?
Are there any prizes?
-
These are just random rambles from my hobby. If you have no interest, then don't follow it. There are no prizes for the kids. I have no plans to do any sort of crowd funding or open source design.
The problem is the temperature effects are not linear. Each stage has its own signature. The attenuator for example has a different signature for each range. So attempting to collect all that data and then fit it, makes little sense even for a one off experiment like this.
After making several more attempts to tame it, I am running out of tricks. You can see it collecting at 4X speed here after warmup and alignment with the current setup.
https://youtu.be/Kx2uw9XTDqk
-
Same plots as before showing the new reference. For the less observant, the part number has been the title for some time. Green is after several attempts to compensate, compared with prior to compensation (Red) and my HP34401A bench meter (white).
While it may look like things are getting worse, it's all part of the learning process and understanding the problem. Two more tests to try.
-
I think it depends a lot on the architecture of the converter and those designed for audio use are probably not DC accurate. Was comparing a Maxim MAX1416 16-bit dual channel sigma-delta converter against an Agilent 34401A today.
2.5V ref LM4040A, 10:1 frontend attenuator 180k/20k, PGA gain = 1, unipolar unbuffered mode and 60Hz update rate. Given that there is no ground plane, at least 20mV pk-pk digital noise on the 5V supply and it's stuck on top of a noisy Arduino Mega the accuracy and linearity was surprisingly good, +0/-2 LSB for 0-24.999V FSD or 25,000 counts. Maybe the Arduino fixed point to float is rounding down the last digit.
-
Welcome to the club of home made TE. Matching the performance of the 34401A is proving to be a fun exercise. Like most of my projects, I tend to bite off more than I can chew.
I don't have any sort of input protection yet because I have only been using the source meter to measure the battery life of some handheld meters. For this sort of measurement, it's good enough as is. It seems like it's a good application for GDTs to minimize the error that path will create.
Next test is running. Clip showing the first hour, still collecting at 1sps, sped up 16X (like watching paint dry) and still using the Fluke 1V standard.
https://www.youtube.com/watch?v=xfmNWr-39PI&feature=youtu.be (https://www.youtube.com/watch?v=xfmNWr-39PI&feature=youtu.be)
-
Ran a couple more tests today. There's no low hanging fruit to pick. I think the next step is to try and work on the noise.
-
Todays sweep. Note the placing the beach towel over the unit still appears to have some effect while the control loop compensates for it. The ADCs and reference are both in inside their own insulated boxes, which are inside a box which is temperature controlled.
In the first plot, I have switched to double precision but I am still only storing the data as single. The UT181A software also works in single precision. None of the meters I have would make use of it. Using the 181A software to compare the data prior to this last week of testing various ideas, the P-P noise and drift are still about the same. This is roughly 7 hours of data each.
A harsh reminder of the time this hobby can suck of you with nothing to show for it.
-
During the same test, nothing taken apart, the gain was changed back to 4X where the original data for the VRE3025A was taken. It's improved ever so slightly. Wishing now I had stored the data differently as I can't really go back to show how much it's improved. Splitting hairs.
-
Comparing the final mods against the HP bench using both the reference and my old linear bench supply which drifts all over. Should provide a good idea how well the two track. Had shut everything down last night and I should have let it warm up a bit more. All in all, the noise and drift look decent.
https://www.youtube.com/watch?v=jzvIm025heA (https://www.youtube.com/watch?v=jzvIm025heA)
-
Hi Joeqsmith
Can we just talk calm for a while in PM ? You have my word on honesty on previous posts / comments.
-
Planning to look at the TI ADS1262 evaluation board to see how it compares.
-
Their software it seems is bit lacking but they do provide the source. As a bonus, it's in Labview which I have some basic grasp of. They used version 2014 and I am still using 2011. It wasn't a big deal to save it to the earlier version. As before, I will run the ADC at the slowest rate which has a side benefit that their motherboard is able to keep up without any problem, unlike the person posting:
https://e2e.ti.com/support/data-converters/f/73/p/800108/2962005?tisearch=e2e-sitesearch&keymatch=ads1263evm-pdk
The EVM is just sitting on the open table while I sort out the software. Here are the first 15minutes collected from the Fluke 731B. Gain of 1, 2.5Hz sample rate. Large spikes are me sitting in the chair. I'll let it run for a while and make sure the communications are solid.
-
https://xdevs.com/review/ti_ads1262_p1/
-
https://xdevs.com/review/ti_ads1262_p1/
LabView often is tricky and not very reliable in operation, as even after just ~10 minutes of fiddling with various tool settings I got it to crash with error:
So far, the stripped Labview code I put together seems solid but it will need to run for several hours before I move on.
*****
Ran for a few hours without a problem so I have gone ahead and added the ability to store the data.
-
I've been using their Labview code (actually, it just calls their lower level code) for a full day now without any problems.
Normally I would use my cookie can for the shielded box but it won't fit inside my meat packing box which will be required to get the temperature stable. So I made up a quick enclosure with some PCB material, copper wire mesh and some RF gasket. The small hole in the top is to gain access to their reset switch. For this first actual test, I'll just use the towel to keep the air flow down.
This is the first hour of data collected, played at 16X. DRATE is 2.5Hz, PGA 2. Shown with FIR but I have ran a few tests with the SINC4. This is a mistake after too many hours.. Its running SINC4.
https://www.youtube.com/watch?v=Egb-7TYUNqk (https://www.youtube.com/watch?v=Egb-7TYUNqk)
-
Comparing the TI ADS1262 eval board's current setup against the HP34401A and the my other delta sigma. The first graph, I ran a fairly long sweep collecting both the HP34401A and with the final changes I made to the source meter. I was doing something in the lab and bumped the setup. The data is stored double precision now.
The second graph, I stripped the DC, or normalized them to zero. I then look at the start and end and attempt to remove the slow drift. The old HP is doing a much better job with these small errors.
-
Graph showing nine hours. I have no idea why the sudden drop but will continue to let it collect for the day. The rate of change, considering it is wrapped in the towel, it's hard to believe it's thermal.
-
Human error. Checked it after another four hours and it still hadn't recovered. As soon as I removed the towel it snapped back. I need to add some insulation to the inside of the box before a more major SNAFU occurs.
Also thinking to try the home bundle. It looks different from the version we tested with the watermark.
https://store.digilentinc.com/labview-home-bundle/
-
Added an APEX reference to the evaluation board along with a liner supply. Drift seems to have improved but it's not a controlled test. The temperature is what it is.
Enabled the PGA's chopper to see if this would remove some of the noise or the wobbles but it doesn't appear to make a difference other than taking longer to collect. If I normalize all the five data sets and then remove the wobble, you can see how the noise compares. Ignore the p-p, but look at the standard deviation. Adding the reference increased the noise (shitty workmanship, board hacking).
I think the next step is to blob the inputs.
-
Insulating the PCB enclosure.
-
All data with the DC removed (normalized to zero). Most data sets were short, or off screen so for the most part, ignore the readouts.
Purple: Showing the ADC's input shorted with jumper wire at screw terminal. Then moved the 731B's guard from the PN to the ground which slightly improved the noise.
Dark Blue: Decided to AC terminate the power leads with a poor mans feed through (cap + coax), to the chassis. Also added braid to the reference signal's cable and terminated to the chassis. Then tossed a towel over the mess.
Light Blue: Original data after adding the APEX reference without towel.
Green: Original data of stock eval bd with shield + towel.
-
The orange blob seen through the screen is one of the small disc caps going from the braid to each power lead.
-
Finally, I have the noise sorted out and things are working well enough to try and run a somewhat constant temperature test with it. The meat packing box has the USB cable already routed. I want to combine my software that collects data from the ADS126e evaluation board with my temperature controller.
My linear regulator is inside the shielded box along with the APEX which dissipates a bit of power. At lot more than the handheld meters I have been using for a load up till now.
-
Testing out the software now. We've had USB for how many years now and the team designing this couldn't figure out how to make it work without a push button switch to reset the board.
The linear regulator is located inside the screened box. This dissipates a couple of Watts. I was thinking to run the temps low but am concerned with humidity. My plan is to use the regulator to heat the chamber and control it around 30C. The Peltiers should have an easy time keeping up.
While attempting to tune the loop, looks like I picked up a new troll. I wonder why they bother.
-
Tuning the controller then things go bad... I wonder if I am pushing the evaluation board beyond it's limits. I'll let it run and see if this trend continues.
https://www.youtube.com/watch?v=n38-smUTLg8&feature=youtu.be (https://www.youtube.com/watch?v=n38-smUTLg8&feature=youtu.be)
-
:-//
I suspect the board has some sort of thermal oddity. This is similar to what I had seen before when I thought something may have touched the case. I had the towel over the unit and after some period, the reported value started to dramatically change.
After reaching an offset of 12uV, I started bringing the temperatures back down. What's strange is the reported value becomes stable but doesn't return to the original value, it seems no matter how low I bring the temperature down.
I have a couple of K-type sensors to monitor the temperature and this appears fine. The reference doesn't appear to be the cause. There's only so much on the eval board. Time for some sleep and then a fresh look at the data sheets and manuals. I'm missing something.
-
A few bone head things to report on this TI eval board.
First, as it turns out there is no need to press the reset switch on the evaluation board. As a matter of fact, had I read their popups, it states it. Just power cycle the board. This is far better than removing the board from the chamber.
Next, my software had a major structural flaw that was causing some really strange behaviors. I was suspecting their under lying code and wrote TI today about it after I couldn't find anything in the datasheets. This problem was there from the beginning and am now certain this was the problem I saw when I use using the towel to keep the air flow off the circuit board.
I am very happy to say the system is finally working.
Shown is updated software. The chamber is programmed to run a sweep from 22 to 30C and back down where it will hold. The dwell time at each temperature will be 60 minutes. The reference's flat area is higher than this. It's more of a sanity check.
-
After setting the chamber to 22, the profile holds for 2 hours. Then 1 hour hold at each step. Total time shown is over 11 hours.
Data was collected using SIN4, PGA 2 non-chopped, DRATE 2.5, APEX 0.6ppm external reference.
Notice the second graph is showing the offset after the board was returned to 22.0 degrees. This is roughly 200nV per division or shift of 600nV. I suspect this shift is caused by parts outside the environmental chamber.
I'm pretty much at the limit of what I have the ability to look at using my meat packing box and old modest equipment. I would have no way of cross checking anything. It's a pretty impressive board by my standards which I will admit are pretty low.
-
Setting up to run one last test before putting the TI board away. The next test will be at a fixed temperature with the chopper enabled. I'll post one last video of it collecting data.
My troll buddy I mentioned above had posted one last time but I don't have their response. It was basically something about Donald Trump and their loss of respect for me after each post. They closed with something to the effect that they doubted I would leave the post, something about them making me look bad to my viewer's. I did end up blocking them like I have done with most trolls. I just don't have time for them and I doubt viewers do as well. The real twist was I went to their channel and they had one video I assume they had made about saving on your home energy bill with their device. :-DD No wonder they were so upset. If you are into the free energy thing, you should check them out. I did not watch it.
-
24 hours of a somewhat constant voltage and temperature.
https://www.youtube.com/watch?v=s_7q5hxfPB4 (https://www.youtube.com/watch?v=s_7q5hxfPB4)