diffamp

J

John Larkin 8 years ago

I have a lot of SSRs that I want to measure the voltage drop across, both ON and OFF. I want maybe 25 millivolt accuracy with a common-mode voltage up to +-72.

Each SSR has two 1M precision pickoff resistors and a mux that dumps the resulting currents into a diffamp, then an ADC.

Even buying 0.1% resistors, the common-mode error stackup is big.

So this is the fix: we use a two-opamp diffamp where the upper amp is a current inverter into the lower amp.

formatting link

The lower output represents the voltage across the SSR. But the upper amp output tells us what the common-mode voltage is at any instant. So if we digitize the pair simultaneously, and have a cal table, we can take out the residual common-mode error and the opamp offsets. We can generate the cal table in manufacturing test, load it into the FPGA, and do the math in hardware. The Susumu thinfilm resistors are very stable over time, so we should be OK.

John Larkin Highland Technology, Inc lunatic fringe electronics

Vote

G

George Herold 8 years ago

I'm still starring at the schematic... and trying to get it. VCM is from some near ground feedback?

Can you say more about the 0.05% R's. I use Susumu too. (0.1%) It's nice seeing a string of 9's, or 0's. Makes a man wonder about his DMM, :^)

George H.

Vote

J

John Larkin 8 years ago

The unipolar ADC reference is +3.3 volts. VCM is 0.5 of that, +1.65. That avoids needing a negative supply.

I measued the tempcos of a bunch of the Susumus, and they were all under 10 PPM/K. They average around 15 cents each.

John Larkin Highland Technology, Inc lunatic fringe electronics

Vote

M

mkolber1 8 years ago

i think linear tech has some ics that will do this

mark

Vote

G

George Herold 8 years ago

OK then sum and difference, in theory that should work. How close is it?

You can buy through hole resistors, 0.1% for $0.20 in onesies. (yageo, mouser) Maybe smd with leads attached, but I don't care. A 1% cap would be nice.

George H.

Vote

M

Martin Riddle 8 years ago

But they are voltage limited, 72v is a tad high. Cheers

Vote

C

Chris Jones 8 years ago

ADI has some diff amps for current sensing that will do a few hundred volts e.g. AD8479, but I don't have time to look for others now. Not sure if the price would be within budget though.

If there is already a floating supply for the SSR gate drive, then another option would be to float a cheap op-amp at the SSR voltage, to amplify the voltage drop across the SSR by a factor of maybe 100. With a bigger signal relative to the common mode, you can then use a diffamp with cheap resistors to get it ground referenced again. For the floating amplifier, if you need both polarities then you could use two halves of a LM358 for example, one wired for each polarity.

Vote

J

John Larkin 8 years ago

As I noted, I want to measure the voltage across each switch when it's closed and when it's open. Open can be as much as 72 volts.

John Larkin Highland Technology, Inc lunatic fringe electronics

Vote

J

John Larkin 8 years ago

That is incidentally a cute trick: apply single-ended gain ahead of a diffamp to improve CMRR.

But code is free!

John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com

Vote

C

Chris Jones 8 years ago

Ok, I think that's new information about the voltage drop being up to 72 Volts when open. Do you still need the same resolution then?

Vote

J

John Larkin 8 years ago

The diffamp outputs each feed an ADS7866 adc, unipolar 12 bits. So adc range is equivalent to +-82 volts, and the LSB is 40 mV. We'll average a few shots and get better resolution.

Interestingly, the common-mode range in this setup is +-165, and could be higher if I reduced the 10K resistors.

The trick here is to use a simple differential circuit and improve the CMRR with software.

John Larkin Highland Technology, Inc lunatic fringe electronics

Vote

diffamp

Join the Discussion

Didn't find your answer?