isolated delta-sigma converters

Didn't look at this particular spec sheet but the isolated mode Delta-S***ma converters I have seen are the stream of the input modulator before any sigma part of the operation so is just high frequency delta modulation data which shouold work great.

I also remember that they aren't real cheap for whatever reason more than likely, especially AD I would suspect.

which ones have you seen that use delta modulation?

both datasheets says second-order delta-sigma modulator

We use the ADuM analog Part (adm3251 - for rs232). They work, no problems. They seem to be quiet too, the data sheet give a recomended layout for EMI. I wonder how that affects the ADC.

Cheers

The ADI is around $5 at 100, and the TI is a bit less. Looks like the ADI is a bit better electrically.

Looks like all the FPGA has to do is lowpass filter the bit stream.

Yes it's cool. I've been tracking these for a few years now, and I'm appalled at their poor DC accuracy. About 0.1% for the TI part, and only a little better for the AD part. Sheesh, the whole input switching nature of these ADCs cries out for the input-reversing schemes that drives DC offsets down into the low uV region. It's not as if they aren't charging big bucks.

I'm surprised the motor-driver blokes put up with it. They often run motors at low torques, don't they need absolute polarity control?

ADI claims 130 uV max offset with very low drift. One could always zero it when the current is known to be off.

I'm measuring AC!

First or second order delta modulation I put in the same catagory but it's just better with 2nd order.

Sigma portion done after the delta mod serial string of 1's and 0s have been transferred over the isolation barrier at high frequency.

Forgot about powering the delta modulators. Yeah, that's a pain but not too big of issue.

I did a 32 floating channel design long time ago where the 100 kHz DC/DC conversion was clocked by a timer on the CPU side of things. The transformer output on the floating side was also the delta conversion clock. Saved an 'expensive' opto-coupler per channel. Sigma parts were just 4 counters in the CPU (1:8 multiplexed to get 32 channels). Got a nice linear 14 bit conversion out of that.

Arie

delta and delta-sigma are related but different things

Yes, of course they are. Delta modulation is the input which is sent usually to a DSP for decimation to turn the HF oversampled delta modulation input to a lower frequency wider bit width.

We used, new at the time, Crystal semiconductor oversampling converters in the audio business in the 1980s. Single order Delta-mod to start with. Amazingly good stuff and I was "converted" almost immediately. Those chips had a test pin that was the delta modulator output up at somewhere around 3 MHz.

my

go

delta modulation and delta-sigma modulation is not the same thing

Meant to say the AREN'T.

The SIGMA part isn't part of the modulation though. It's processing of the delta modulation.

Actually, I'm not sure why you even mention this ?

ew

on my

an go

nce.

ut

yes it is, delta = difference, sigma = integrator

Right. There is "delta-sigma"or integration involved with the 2nd order delta-sigma modulator... But as far as this thread is concerned, isolation, it could be single delta mod or delta-sigma modulation but the downconverting, decimation or sigma that I was referring to is done after the serial HF bit isolation. Integration or addition or sigma used both places, among other processes. It's all good

But I do see why you did bring it up. Good stuff.

Some microcontrollers (at least STM32 series have these) also have SD-inputs, which make it easy to use SD-modulators for isolated ADC inputs.

There is a lot of delta-sigma modulators on the market. Adding a suitable isolator brings you the same and often makes the combo cheaper as compared to an isolated ADC. The isolator can have an isoPower-compatible built-in transformer driver, which solves your second problem as well. SOme of the modulators have Manchester output, which allows transformer-based data link isolation and you basically end up with just two tiny toroidal cores next to the chip.

And when you have an isolator and a power supply, then why delta-sigma, if LTC1407A-1 is so cute?

Best regards, Piotr

He probably means the decimating filter part.

Best regards, Piotr