IWR1443: Under what circumstances should we re-run a full calibration when partial calibration is usually used?

HI, there:

You can find detailed information on calibration at https://www.ti.com/lit/an/spracf4a/spracf4a.pdf?ts=1622663480625&ref_url=https%253A%252F%252Fwww.google.com%252F

In general, it is suggested to do a factory calibration (an interference free environment) and restore it during power up in the field.  After that, no full calibration is needed.  But run-time calibration (section 5 in the above document) is suggested to compensate the temperature effect.  Note that there will be some phase and gain jump before and after the run-time calibration.  If your signal processing chain assume continuous phase/gain cross frames, then it will be better to disable run-time calibration and instead call one-time calibration at your convenience.  For run-time/one time calibration, temperature will be checked, and different gain may be applied only if the temperature is changed 10 degree (in Celsius).    If you never run the calibration, remember that the device will experience gain change over temperature, roughly 1dB down for every 10 degree of temperature increase.  

Best,

Zigang

Hi,

Our IWR1443 device work as below,

1  power on 

2  restore the the full calibration data,

3  partial calibration,

4  run 1 chirp and calculate the distance 

5 power down

we meet the problem some time, the adc data very strange, seems the RF part doesnt work properly.

But  if the IWR1443 work as follow,  the 1443 works well. 

1  power on 

2  run the full calibration,

3  run 1 chirp and calculate the distance 

4 power down

HI, 

Can you try to skip partial calibration in your first approach, and see whether the problem is in restore or in partial calibration.

1  power on 

2  restore the the full calibration data,

3  run 1 chirp and calculate the distance 

4 power down

And what do you do in partial calibration?

Best,

Zigang

Thanks for your advice.

To skip partial calibration, can we just set partial caibration mode and mask all to 0?

We just tried skipping the partial calibration, but the strange adc data was still there sometimes.

Thanks Xiushan

I'll need more time to look into this. Will have a response for you by Friday.

Regards,

AG