Sampling window for shunt current measurement

Timon Achtnich

Hi,

I have read a lot about shunt current measurement in the InstaSpin Foc/Motion user's guide (spruhj1d). There, the advantages and disadvantages of the two shunt technique are explained in detail. From the picture below we see that the measurement must be done in the middle of the period. The pwm module of the InstaSpin projects are configured as up down counters, as long as I know. Therefore I would conclude the the current measurement should begin close before the pwm counter reaches its period counter value (In the middle of the picture below). If we start sampling the adc channels at counter equal period we can only use the half of the possible sampling window.

Question 1:

The pwm modules of the InstaSpin Motion labs are configured to generate a SOC pulse at counter equal zero. See for that the picture below which comes out of hal.c of the InstaSpin lab 10c.

Why is the pwm module set to CounterEqualZero instead of CounterEqualPeriod? With the configuration CounterEqualZero I assume that the sampling is done in the wrong moment.

Question 2:

How can I start the current sampling with the increasing slope of that low side fet which has the shortest on time?

In other words, how can I use the full time of the sample window?

Thanks for your reply

Timon

over 11 years ago

0 ChrisClearman over 11 years ago

TI__Guru** 119680 points

sampling is started on timer = 0

this is the center of the low side switch on-time.

the high side switch duties are centered aroud timer = period

make sense now?

0 Timon Achtnich over 11 years ago in reply to ChrisClearman

Intellectual 555 points

Hi Chris

Yes, now its all clear. I assumed that timer period = 0 is at the center of the high-side switch on-time, which is wrong.

Thanks!

0 simpson bart over 11 years ago in reply to ChrisClearman

Prodigy 210 points

Hi chris,
From InstaSPIN User's Guide.pdf (In page 498,Chapter 20.5 3-shunt),we notice that there exists small sample window(which I think the adc of this phase current is not correct when sampling).
However,the code from the function DRV_readAdcData(),which tell us that the 3-phase ADC_current value is read by ADCSample all the time.So I wonder when one of pahse is in small sample window,this phase should be calculate by the another two phase.

0 ChrisClearman over 11 years ago in reply to simpson bart

TI__Guru** 119680 points

look at proj_lab10 where we show our over-modulation scheme.

notice that the normal "svgen" module is replaced with "svgen_current" module. you can see the logic used in svgen_current.h on how we use/ignore values based on the shunt sampling window.

C2000™︎ microcontrollers

C2000 microcontrollers forum

Sampling window for shunt current measurement