topic Re: ADC self-test in one-shot mode in S32K

ADC self-test in one-shot mode

LeoZ — Mon, 30 Oct 2023 09:19:36 GMT

* bellow figure is from [S32K3xx Reference Manual, Rev. 7, 05/2023]

Question 1: "Capacitive self-test. One of the calibration steps is being run."
- not understand this statement,
- does "C0~C11" correspond to calibration steps?
Question 2: "Supply for OneShot mode and (supply + capacitive) for Scan mode only."
- does it mean: in OneShot mode, only Algorithm Support is available?
- if so, is it not necessary to use Algorithm Capacitive for OneShot mode?

Re: ADC self-test in one-shot mode

davidtosenovjan — Tue, 31 Oct 2023 08:16:24 GMT

Supply self-test executes in a row whilst capacitive self-test is interleaved into normal ADC conversion sequence as you can see in the example in RM, section 61.3.17.2.1:

Re: ADC self-test in one-shot mode

LeoZ — Tue, 31 Oct 2023 09:20:15 GMT

Hi, the flow you shown seems to be for Scan mode? in 61.3.17.2.1, the scan-mode 15-steps covers all AlgorithmS and AlgorithmC, while one-shot-mode 11 steps only covers AlorithmS, and I wonder how I do Algorithm-C in one-shot mode.

Re: ADC self-test in one-shot mode

davidtosenovjan — Wed, 01 Nov 2023 09:47:03 GMT

In single shot user defines which S or C step is executed after normal conversion.

But this way you cannot achieve requirement for S algorithm supply self-test:

In other words - it is needed to use Scan mode for executing of S algorithm.

Re: ADC self-test in one-shot mode

LeoZ — Wed, 08 Nov 2023 08:56:11 GMT

Thanks for reply ~~

* currently, in my motor-control system I need to use 3 ADC-channels to sample 3 phases current(voltage/resistance), I plan to put these 3 channels to a chain. I think I need to use one-shot mode to trigger this chain in every PWM period, and I planned to algorithm S/C to test the accuracy of Bandgap and VREFH

Based on your answer:

* If I use Scan mode, the ADC sampling is continuous and always ? --- could I use scan-mode by some-method to implement the ‘trigger-strategy’ shown in the figure
* If I can only use one-shot-mode (so I can not use algorithm S), then I plan to take Bandgap and VREFH sampling as ‘normal-conversion’ as shown in the figure -- is it recommended ?
* do you have some recommended ADC self-test strategy for for motor-control system

Re: ADC self-test in one-shot mode

john_floros — Thu, 09 Nov 2023 14:46:04 GMT

Hello,

For motor control applications during run time some methods that can be used to verify that the motor current readings are valid and therefore the ADCs are accurate are:

1. Out of range checking

2. Checking that the sum of the motor currents are equal to zero.

3. Correlate the motor current readings with the what is expected by what was commanded by the software.

Regards,

John

Re: ADC self-test in one-shot mode

LeoZ — Fri, 10 Nov 2023 02:33:48 GMT

Sorry did not describe my question clearly, I meant:
* to sample phase current for motor-control system (BLDC with FOC-control): recommend to use one-shot mode or scan-mode; and corresponding recommended ADC self-test algorithm S/C

Re: ADC self-test in one-shot mode

john_floros — Fri, 17 Nov 2023 15:33:05 GMT

Hello,

Both methods could work. It depends on what your needs are. Whether you do one shot or scan your first conversion should be the motor currents but for less timing sensitive conversions you can follow the motor current conversions. By having less time sensitive conversion after the motor current conversion can make the ADC configuration simpler than doing everything done in one shot.

Regards,

John