28027 inside temperature sampling

Harrison,

Averaging values is always helpful in removing any potential noise in the reading.  Does increasing the sample window help to stabilize your readings?

-Tommy

the inside temperature observed in the watch window varis from 27 to 50, it seems the variation is too larger too be useful.

Anybody knows it is normal?

thank you,

I increased the sample window,ACQPS from 8 to 20, the variations is still about 10 degrees.

the reason I tried the inside temperature sensor is because I found the ADC sample is quite noisy on a simple voltage input in TI demo kit, so I thought the inside temperature sensor signal should be quite stable to be used to varify if it is the signal filter is not good enough.

it turned out even the inside temperature sensor ADC is not  stable, this is a bit strange to me, I didn't find the same problem on TI classical 430 MCUs.

the demo kit is in working mode and just power by 12v from a very good power supply.

thank you for help.

Harrison,

Are you also taking steps to account for the ADC Initial Conversion issue described in the device Advisory?

-Tommy

Tommy,

What do you mean?I didn"t find such things in datasheet and ADC guide(spruge5c).

could you please let me know a bit more?

thank you

Harrison,

Please see the Silicon Errata document in the product folder.

Direct link to document: http://www.ti.com/lit/pdf/SPRZ292

-Tommy

Thanks for the information, I didn't notice such documents exist.

I used 8 avarages and increased the hold time to 666ns, the sample is still varying from 26 to 32 degrees, is it normal?

I still think it is too large.

I don't have the initial sample problems.

thank you

Harrison,

This sounds like more variation than expected.  A few more follow up questions.

Are you using a TI development board or is this running on a custom board?

Also, are you using the controlSUITE example code or are you using your own code?

Is VREG enabled?

Are you using internal VREF reference?

Do measurements on other ADC inputs vary?

-Tommy

I am using TI HVPSFB kit;

VREG enabled

Internal Vref

other ADC inputs vary too.

Thank you

Harrison,

Would it be possible for you collect a continuous sampling of the temp sensor to see if it ever stabilizes?  It is possible for the device to heat itself up a few degrees as it is executing code.

Would it be possible for you to check for noise on the VDDA/VSSA supplies while you are sampling with the ADC?

-Tommy