PGA305: PGA305 Temperature Compensation Works Sometimes

Hello Lori,

Do you notice the same issue when using a 3T calibration, or is it only when doing a 4T? What is the error typically with the first calibration and what is the error with the second or third? Can you show the coefficient results of a successful calibration vs a failed calibration?

The calibration can be a bit sensitive to gain as well (although if that is not changed between calibrations it may not be a cause of this issue). In the results you posted, the digital PADC gain is set to 3. I would recommend increasing the PGAIN in the analog EEPROM. Ideally, the digital PADC gain should be 1 after the calibration.

Regards,

I have only performed a 4T calibration.  The error on the first calibration run seems to be on the order of 2% fso.  

The coefficients in previous email were from a first run and the results were poor.  I have attached the coefficients from the second run and the temperature results. 

Although the zero and span shifts through temperature did very well on the second run,  I have not been able to improve the linearity for any units tested.

Hello Lori,

Can you please post the ADC and DAC data from the successful calibration? I would like to compare it with the unsuccessful calibration.

Out of curiosity, I tried entering the data from your first post manually into the GUI to see if I got the same coefficients that you generated. Just to clarify, does the coefficient page from your first post correspond exactly with the ADC data posted below it? I've attached my results, which are somewhat close, but not exactly the same as yours. Can you please try manually entering the data from your first calibration and generating coefficients from that? Does it match the coefficients that were generated the first time?

Regards,

I am in the process of running a new prototype.  I have run it once and began verifying but it failed at the second data point, 110 degF.  I did not continue verifying after it failed at 110 degF. 

Attached is the data for the failed run.  I plan on running it again in the morning and will send the data if it passes. 

Attached is the data from the second compensation cycle of the prototype unit.  It compensated very well this time. Again,  verifying that the first pass at temperature compensation does not work and the second one does.

Hi Lori,

Can you run a 3P3T calibration with your device? I'm curious if the issue you are seeing is related to the GUI's settings for the number of calibration points. This is not something I have encountered before when using 3P3T calibration.

Regards,

Hi Scott,

We temperature cycled the pc board from -40 to 185 degF are currently running a test to see if this eliminates the problem.  If this does not work,  we will try the 3P3T calibration to see if it works. 

I will update this post when the first test is complete.

Thank-you for your help,

Lori

We have compensated two prototype units using boards that were temperature cycled.  Both prototypes compensated well the first time through.

Hi Lori,

That's interesting. Were the PGA305 prototypes powered and running while the temperature cycling happened?

Regards,

No, they were not powered during the temperature cycling.