Removing DC offset from ADC Buffer

[blue_diode]

I'm using a STM32 Cortex M0+ to read an AC signal from a CT. I'm sampling at 6kHz and storing 400 samples. The signal has a DC bias equal to Vcc/2 = 1.65V. In the digital domain this is 2048. When there's no input signal, I only read the DC bias which is OK. Have a look at the dc_bias_raw.txt file. As you can see, the readings are not so bad. It varies from 2044 to 2052. The DC bias is very precise in hardware so I think this variation is due to the noise floor. Now if I want to do RMS in that set of data, I need to find a way to deal with this DC bias variation. In the file adc_raw_current.txt you can see the samples when there's an input signal.

I have been thinking the following:

1- Subtract a fixed value of 2048 from each ADC reading. This is no so good as I said above this value may vary slightly. Also, if I want to read zero cross it may cause errors to choose exactly 2048 as reference.
2- Sample the DC bias and average it.
3- Don't remove the DC bias. If I calculate RMS then I would get the DC value when there's no input signal.
4- Use a more sophisticated software high pass filter

I appreciate any help or suggestion

 

[KlausST]

Hi,

"More sophisticated high pass filter".
This sounds rather difficult.
Good news: it's way less difficult than RMS calculation.

I'd adjust the high pass filter with your RMS algoritm.
--> How do you do the RMS calculation?
Blockwise? Reading n samples in a buffer, then processing RMS?
If so, then first do an average on all the buffer values. Sum up all buffer values the multiply by "1/n".
Then do the RMS by "buffer_value(n) - average"

Other methods are for continouos measurement
* biquad high pass ( read_ADC --> high pass --> store in buffer)
* low pass to get DC value, subtract this value from ADC_values

Klaus

 

[danadakk]

Some ARM Cortex parts have differential input capability for SAR or DelSig.
So you could take your precision Vref and feed that to - input such that
you are digitizing a +/- range about that Vref, eg. the diff input subtracts
out the offset. Some parts even have a fairly precision Vref on them, +/-
.1%.

Shown above is a Cortex part with both a SAR and DelSig and +/- .1% Internal
Vref, all single chip. The Vref can also use onchip PGA to G up its value, but
the PGA G accuracy probably would not support 12 bit absolute accuracy.
Or you can use external Vref. Note OpAmp also on chip (there are 4 of them).
You could use onchip OpAmp with external precision Rs to G up the Vref and
get the accuracy, just a thought. Note the DelSig is up to 20 bits, so you can
manage errors better. If you need more channels also onchip analog muxes
to manage that.

Adding a filter or a subtractor adds error and noise.


Regards, Dana.