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.

PGA970: Required the Schematic for LVDT half bridge probe.

Part Number: PGA970
Other Parts Discussed in Thread: OPA1662, OPA1664,

Hi Scott,

  I am working on PGA970 LVDT signal conditioner IC.

 In that i have used OPA1662 IC and PGA 970 IC .

 In my hardware i am getting small amount of noise in excitation waveform both P1 and P2.

can you share me the schematic of PGA970 with OPA1662 and LVDT half bridge probe.

Also i am using solartron AX/1/SH probe for communication. 

Awaiting for your reply,

Thanks and regards,

VAI

  • Dear Scott,

    Above issue has been solved.

    Attached below are the displacement we are getting at the 

    We have calibrated at 1.000 , 0.000 and -1.000.

    here we have used LVDT probe drump for calibration and X-CTU for software reading.

    the behavior we are getting is non-linearity if this behavior we are getting below as per below reading then we are used for further development.

    If there is any solution then please let us know.

    Kindly reply it on urgent basis. 

  • Hello Vai,

    What are the accuracy targets? It looks like it's behaving as I would expect it to with the largest deviations being at the points that are furthest away from the calibration points, and the best accuracy at the calibration points. But it depends on what your accuracy targets are if that is acceptable.

    Regards,

  • Dear Scott,

    Accuracy which we required is 1 micron.

    here we used external amplifier for secondary input using OPA1664 IC with gain of 5.

    the probe which we are using has +/- 1.000 mm linear displacement.

    Kindly solve the above issue or give the suggestion to get the linear waveform with accuracy between +/-1 micron..

    Awaiting for your reply,

    VAI

  • Hi Scott,

    Is their is any updated on above message.

    also i want to know that if i bypass the external OPAMP then i get the linearity reading when calibrating at +1.000,0 and -1.000.

    with internal software gain as 1. but the count is very less also the reading instability issue is there.

    But if i applied the secondary input to OPAMP the i get more count but the their is linearity in reading after calibration +1.000,0 and -1.000.

    with internal software gain as 1.

    I have check the linearity of OPAMP it is working fine.

    Is the internal software gain as 1 for ADC is Affecting the linearity of PGA is Yes then please tell how to rectified it or some additional component we need to placed to get the linear count for calibration?

    Kindly solve the above issue or give the suggestion to get the linear waveform with accuracy between +/-1 micron.

    Awaiting for your reply,

    VAI

  • Hello Vai,

    Can you share your schematic? Why have you decided to use an amplifier before the secondary input? Is the output of your sensor too low? Typically the primary signal would be amplified rather than the output of the secondaries. The primary excitation of the PGA970 may need some external amplification based on your application, but that is typically where I would start as opposed to on the secondary side.


    Regards,

  • Hi Scott,

    i have amplified the excitation waveform with more gain i have got the total count reading 17000 in secondary waveform.

    but i required total count of 30000 in secondary waveform. 

    here i cannot increase the supply voltage according to my application requirement.

    also to get more stable count what additional component need to be placed in primary or secendory side of LVDT probe.

    Regards,

    VAI

     

  • Hello Vai,

    What is the amplitude of your signal to the primary? And what is the voltage you are supplying to the PGA970 itself? Can you not use the same power supply from the PGA970 as the supply for an amplifier on the primary excitation output?


    Regards,

  • Dear Scott,

    Sorry for the delay.

    I have manage to amplify the amplitude of primary waveform by changing the gain of primary excitation waveform.

    Currently i am facing 1 issues.

    1) How to get more stable sine wave from secondary side any extra component need to be added or any precaution need to take care.

    Regards,

    VAI

  • Dear Scott,

    Awaiting for your reply on above issue.

    Regards,

    VAI

  • Hello Vai,

    Just to clarify, is your issue still with the nonlinearity of the output after calibration? Or are you experiencing a fluctuating input to the PGA970 from noise on the output of the secondaries? Do you have any filtering on the secondary outputs? A simple RC lowpass filter should be used to filter out any noise above the expected operating frequency of the device. Ideally a bandpass filter could be used. Please see this application note regarding nonlinearity in LVDT sensors: http://www.ti.com/lit/an/slyt582/slyt582.pdf

    Regards,

  • Hi Scott,

     your issue still with the nonlinearity of the output after calibration? Issue Solved

    Or are you experiencing a fluctuating input to the PGA970 from noise on the output of the secondaries? Yes Their is fluctuation in sine wave variation. 

    Do you have any filtering on the secondary outputs? No I have not used any filter.

    A simple RC lowpass filter should be used to filter out any noise above the expected operating frequency of the device. Ideally a bandpass filter could be used. : While using RC filter my raw counts is also decreases is their is any way to get the same raw count with filter.

    Also to save power i have directly turn OFF the VCC of PGA and then power it after some times.

    But the count which i was getting before PGA turn OFF and  the count when PGA become ON is not matched.

    their is about 100 to 200 counts variation when PFA turn ON.

    Please guide us to solved above both the issue,

    Regards,

    VAI

  • Hello Vai,

    If the filter is decreasing the signal too much, then you may need to adjust the cutoff point of your filter higher so that it does not eat into the operating range of your sensor.

    When trying to save power it is better to use the functions built into the PGA970 instead of completely removing power to the device. You can disable the analog power within the PGA970, the excitation waveform, and you can disable the modulators to save power. 

    The most likely reason for your change in modulator output on powerup is due to self-heating. When the PGA970 is first powered on, and your LVDT is supplied with an excitation waveform there will be a small amount of self-heating. This will affect the measurement for a short time until the temperature stabilizes. I would recommend disabling some of the internal functions as described above, but continuously supplying power to the PGA970.

    Regards,

  • Hi Scott,

    I have tried with PGA 970 commands as per your suggestion to reduce the power consumption.

    My supply voltage is stable to PGA 970 but still i am getting variation in reading before going to sleep and after waking from the controller.

    What is the issue,

    regards,

    VAI.

     

  • Hello Vai,

    The variation you are seeing is still most likely due to some amount of self-heating. This could either be in your LVDT sensor or from the PGA970 itself. If you have an external excitation source one way you could test if the LVDT is causing the variation. When you power down the PGA970 use the external excitation source to power the LVDT until the PGA970 is turned back on. If there is still variation then the self-heating of the PGA970 itself is causing the variation, but if the counts are now consistent before and after removing power from the PGA970 then the variation is most likely due to the LVDT's self-heating.


    In either case, there will be some settling period before the temperature equalizes and the counts become consistent. How long does it take before the readings stabilize?

    Regards,

  • Hi scott,

    Now i have tried with continuously power ON PGA970 and not turning off through commands in PGA970 to observed the stability in reading,

    But the stability in reading is still not there, 

    I am not getting the root cause why it is occurring and how to rectify it.

    It has been supply for 1.30 hrs and the count variation which i got is at starting till end 400 counts it is slowly decreasing the counts.

    Kindly suggest the way to rectify this problem.

    Regards,

    VAI

     

  • Hello Vai,

    Does the count drift occur on all of the ADCs in the device? Do you notice a drift in ADC3 as well as both of the secondary input ADCs?

    Can you apply a fixed voltage (not from the sensor) to your ADC inputs and record the counts over time? In that case do the counts still drift?


    Regards,

  • Dear Scott,

    I had given fixed voltage from variable source supply to PGA970 But still i am get variation in count.

    Can you sent me reference circuit to verify.

    so that what are the importance component which i  have not place.

    or which are the necessary component to achieved the stability. 

    Awaiting for our reply,

    Regards,

    VAI

  • Dear Scott,

    Mention below is  the setting done in evaluation board,

    While reading from ADC the variation is shown in below image.

    Also i have attached the Excitation waveform video and secondary excitation video,

    Their is slight variation in both primary and secondary waveform.

    I am getting variation in read ADC also the supply is Fixed Voltage.

    Kindly Guide to solve the issue in ADC reading.

    Awaiting for your reply,

    VAI

     

  • Hi Scott,

    Any update for above 2 message.

    Awaiting for your reply.

    Regards,

    VAI

  • Hello Vai,

    Thanks for the additional information. I'm going to be in the lab tomorrow to take a look at this to see if I can observe the drift on my EVM setup, and I will get back to you once that is done.


    Regards,

  • Dear Scott,

    Awaiting for your reply,

    to achieved stability  which factor need to be take consideration  in hardware please share the document.

    i have checked all the connection as per given in PGA970 EVM user guide.

    also i have checked by using 2 different half bridge LVDT probe but behavior remain same  also using oscilloscope i have checked the voltage level it is also remain the same. 

     

    Regards,

    VAI

     

  • Hello Vai,

    Have all of your tests been on the same PGA970EVM? So far with my initial tests in the lab using my LVDT I have not seen a drift over a long period of time as you have described. There is some initial drift when adjusting the position, but then it settles to a relatively stable value. I will perform some more tests with a fixed supply as well to see if I can better replicate the issue you are facing.


    Regards,

  • Hi Scott,

    Yes all my test have been on the same PGA970EVM with 3.6V DC supply.

    I had check the variation it will start from PI and PE pin of PGA 970 and remains in continuous drift .

    also i have made bare PCB and then i have seen the secondary waveform 

    the drift variation is more as compared to PGA970 EVM.

    What are the factor which causes the primary and secondary waveform drift in hardware so that i can rectify.

    Regards,

    VAI

  • Hello Vai,


    Can you please show the ADC capture with the register set to "Both ADC" over a period of time with a fixed position to show how the ADC counts drift?

    Also, have you tried any different supply voltages? Do you still see the drift with a higher VDD? I would try 5V and 12V. This should not affect the drift, but it is another variable to test.


    Regards,

  • Hi Scott,

    Mention below is the image file as per your requirement.

    have you tried any different supply voltages?

    Yes for 5V it has been the same drift observed and for 12 volt the waveform in primary is clipping

    Do you still see the drift with a higher VDD?

    The Voltage does not change in CRO but the reading changes.

    is their any way ti direct communicated or video calling to solve this on urgent basis.

    Please guide us to solved this issue 

    Regards,

    VAI

  • Dear Scott,

    Awaiting for your reply,

    Regards,

    VAI

  • Hello Vai,

    Sorry for the miscommunication. I would like to see a larger time period. Currently your graph window is set to scroll which only shows a short period of time where the drift is not apparent. Can you run the graph for several minutes, and then click on the magnifying glass in the bottom right corner and select the bottom left option. This will show the entire set of data in a single plot and the drift should be easier to see.

    Regards,

  • Hello Scott,

    Mention below is the image file as per your requirement.

    Kindly give the solution for stable sine wave without any drift.

    This is an serious issue we are facing we need to rectify this solution on urgent basis.

    Kindly share the Reference circuit or factor for stable sine wave in hardware or software.

    Awaiting for your reply,

     

     

    Regards,

    VAI

     

  • Hello Vai,

    At the moment I am still not able to replicate what you are describing on my EVM setup in the lab. I do not see any significant drift in the output of the secondary over time.

    Can you please provide your schematic? You are working off of the EVM, but what additional components are part of your system? Can you provide the datasheet for the half-bridge LVDT that you are using, and show the connections that you have made to the PGA970?

    To further clarify, you mention that you are seeing variation at the waveform generation output of the PGA970. Is that correct? If you connect nothing to the waveform generator output and measure directly at the PI pin do you see any drift? Finally, is the drift a change in the DC bias in only in one direction, or do you see it shift both up and down?

    There is no reference design outside of the EVM itself at this time.

    Regards,

  • Dear Scott,

    I have made the bare PCB using PGA970 and OPA1662.

    In this i have observed the variation in reading is due to temperature changes.

    also i have put the PCB in chamber for heating  to test the stability of reading.

    their is 1 mm variation in reading. 

    what is the reasons we are facing this issue?.

    Also i want need solution on mention issue.

    Schematic is as per PGA970 EVM also wiring diagram is 3 wire diagram.

     

    Awaiting for your reply.

    Regards,

    VAI

  • Dear Scott,

    Awaiting for your reply,

    Regards,

    VAI

  • Hello Vai,

    In order to assist you better, I will need some answers to the questions I asked in the last post. I'm still trying to better understand exactly what is going on when you say that your reading is drifting. Saying that there is a 1mm variation in the reading is not particularly meaningful without an understanding of the sensor you are using and the exact way that it is being hooked up and driven.

    If you are not comfortable sharing your schematic publicly, please add me as a friend and send it through private message.

    Also, please let me know what you are seeing with respect to these questions that I asked in my last post: "To further clarify, you mention that you are seeing variation at the waveform generation output of the PGA970. Is that correct? If you connect nothing to the waveform generator output and measure directly at the PI pin do you see any drift? Finally, is the drift a change in the DC bias in only in one direction, or do you see it shift both up and down?"

    Thanks!

  • Hello Scott,

    Plz let us know how to send the message private for schematic diagram

    Regards,

    VAI

  • Hi Vai,


    I received the schematic. I'll take a look and reply to you after.

    Thanks!

  • Hi Scott,

     

    We required solution on urgent basis i hope we have given reply to all your question.

    Awaiting for your reply,

     

    Regards,

    VAI

  • Hello Vai,

    Thanks for the updated information. We can continue to discuss through private message. Once resolved I can post any publicly shareable information back to this thread to help other users.


    Regards,