• State
  • Locked Locked
  • Replies 3 replies
  • Subscribers 48 subscribers
  • Views 729 views
  • Users 0 members are here
Support feedback
Options
Options
Related

This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

PGA309: PGA309EVM-USB

Joseph Letkomiller
Prodigy 100 points

Part Number: PGA309

Regarding programing set points for the "USB-DAQ EVM" when calibrating the PGA309.

 

I have been using the "USB-DAQ EVM" to calibrate the "PGA309" signal conditioning IC for use with a strain bridge with good success. I have used the "USB-DAQ EVM" successfully as a single unit calibrator and as the brain of a Multi-Cal system calibrating up to 24 PGA309's. We have purchased 8 of the "USB-DAQ EVM" boards again with good success. About 6 months ago one of the "USB-DAQ EVM" units started to insert a slight offset to the calibration point. Enough to cause unacceptable calibration errors. This was a unit used only for testing. Not sure to the cause, I replaced the onboard AtoD thinking there was a problem with it. Then I reloaded the software. The offset persisted and the software reload didn't help. To continue using the unit for calibration experiments, the offset was incorporated into the set point to get a better calibration. (The offset was about .093v high at the F.S. or 4.593. So the set point was reduced by .093 or 4.407. it wasn’t perfect but close enough for my testing.) When I used a different "USB-DAQ EVM" the offset went away. I believed this to be a one off and didn't pursue it further.

One month ago a similar offset started occurring with our 24 channel calibrator. Note this unit had previously been working fine for some time. This is a serious problem as it is part of our production line. Note that the Multi-Cal system no longer uses the "USB-DAQ EVM" onboard AtoD as a voltage measurement it uses the HP34401 multimeter. The symptom is still the same. 

Has there been any other problems like this? Do you have any ideas how to fix it? 

I have attached some images showing an independent meter monitoring the actual set point and the indicated voltage in the software. They are clearly different and I have used many other meters to check this. They all show the same offset. This offset is real and it is programed into the PGA309. Again, if I use a different "USB-DAQ EVM" the problem goes away.   

 USB-DAQ EVM offset error problem.docx

  • 0
    TI__Mastermind 30358 points

    Joseph,

    For production PGA309 calibration you should use a digital multi meter.  If you look on page 34 of the users guide, it describes how to do this ( see PGA309EVM users guide) .  The AtoD on the USB-DAQ EVM are intended for proof-of-concept and not production testing.  Furthermore, these AtoD are calibrated when the USB-DAQ EVM are built, so replacing the AtoD will not resolve calibration issues.  The multi-cal has a similar feature that allows for use of an external DMM.  The software is designed to accept a 34401 DMM.  It can work with others, but this will require some driver development.  Do you have a 34401 DMM?

    I hope this helps.  I suppose it is possible that some other issue is causing the offset, but you should be using a DMM for production calibration anyhow.

    Best regards,

    Art

  • 0 in reply to Art Kay
    Prodigy 100 points

    Please note that we are using the HP 34401 DMM for calibration. I only brought this issue up when we started to have a problem with the calibration system. Note that I stated that I let the problem slide on the development kit. This issue is now manifesting on the calibration system that is using the HP34401. It did not do this at first. Please note the second paragraph in my first message.

    thanks

  • 0 in reply to Joseph Letkomiller
    TI__Mastermind 30358 points

    Joseph,

    Sorry, I missed the point that you are using the 34401 with the multi-cal.  With this in mind it is not likely that you are seeing issues with the actual measured value out of the PGA309.  Here are some possibilities:

    1. Sensor non-repeatability -  some sensors have a pressure or temperature hysteresis.  That is, when pressure is applied to the device from low to high back to low again, it does not return to the same output when it returns to low.  The same phenomena occurs with temperature transitions.  
    2. Temperature gradients in the sensor - in some sensor modules, the electronics and/or sensor have self heating that causes a gradient.  This gradient may change with time and / or applied power or signal.  During calibration if the temperature measured by the PGA does not properly track the actual sensor temperature this leads to no-repeatability of calibration results.
    3. Some bridge sensors will have a required soak time to account for self heating.  A soak time is provided in the Multi-cal system.  This applies power to all the sensors and allows them to reach equilibrium.  The required soak time is different for different sensors.  
    4. The temperature in the test chamber may not be stable.
    5. Noise pickup could impact results.
    6. There are a few ways to understand if you are having some non-repeatability issues.
      1. If you take the calibrated module, and repeatedly apply different pressures and temperatures, do you see the same results or different results.  If the results are always consistent than this is probably not due to items 1 or 2 above, but could be due to 3.
      2. You could use a sensor emulator evm to apply a stimulus to your sensor and test the sensor module calibration procedure without a real sensor.  This eliminates items 1, 2, and 3 above.   The PGA309evm includes a sensor emulator, but when using a complete sensor module the old style sensor emulator is better:  https://www.ti.com/tool/SENSOREMULATOREVM#tech-docs 
    7. The PGA309 and the calibration algorithm have a good history of success.  When there are issues similar to what you mention, it usually relates to a sensor repeatability issue.  
    8. I missed your attachment the first time I looked at this question.  In the pictures in the attachment, it looks like you are comparing the PGA309EVM to a hand held DMM.  The A-to-D on the EVM is usually fairly accurate, but as I mentioned it does get factory calibrated. 
    9. Note: the PGA309EVM is designed to work with the associated PGA309 Test board.  When you connect to an actual test module, the test board should be removed and the signals to your module should directly connect to the 25-Pin D-SUB connectors.  You may already be doing this but, it is a common issue that people try to calibrate with the test board still attached.
    10. I suppose it is possible that the hardware is damaged, but it seems unlikely that both the multi-cal system and multiple EVMs have a similar issue.

    Suggestions/questions: 

    1. Check a sensor module that has bad calibration results multiple times at multiple pressures, and temperatures.  Do this over the course of a day or two.  Do you get very repeatable results?  If so, it is probably not sensor non-repeatability.
    2. It sounds like you have calibrated many devices successfully, and now it is not working will.  Has something changed in the procedure?  Has the sensor changed?
    3. What is the sensor output and drift.  I ask this so that I can know what gain and offset correction is required.  High gain and offset are more sensitive to small variations.

    I hope that I can help you work through the problem soon.

    Best regards, Art Kay