Edge Collective

DIY Conductivity Sensor

Temperature Measurement

RTD amp breakout from Adafruit.

Signal Sampling

Burr Brown App Note on precision absolute value circuits.

Signal generation

Arduino Waveform Generator

Prior Art Among Us

Don's last version in January, based off Craig's 4pstat is here

Craig's latest 4pstat is here

Don's simplified 4 probe based on an earlier design is here

Craig comments

What I found that seems to work well on the driven shield lines was to use some RC snubbers to ground after the 50 Ohm output resistor. Values like 10 Ohm and 100nF give you a corner frequency of 160kHz - if your bandwidth is 10kHz that should likely work out.

Don, if you use my design, there are lots of mods needed to stop it from oscillating.

If you are doing a virtual ground, I suggest using the in-the-loop compensation circuit I got from AoE3

fig2
In-the-loop compensation scheme suggested by Craig here.

Actually I wonder if the corner frequency is actually 6X lower because of the added 50 Ohm resistor? If that is the case, it might be better to back off the capacitance a bit And what about output resistance from the opamp (Mike, I'm looking at you)? If that's like around 100 Ohm maybe we'd be smart using an even smaller cap. This could be why I did not make my intended bandwidth of 100kHz on the bodged up 4pstat.

Revision Notes


As of 10-02-2020:

Don's current attempt at the above is here.


As of 10-06-2020:

As of 10-07-2020:

'Adder inputs' section of Craig's 4PSTAT schematic:

fig2
From Craig's schematic here.

As of Oct 9 2020:

fig2
Revision A of the 0.4 EC-4P circuit: attempt at a simpler 'conductivity' measurement with adder amp feedback coming directly from the the amp output. See discussion here.

Adder circuit explanation on wikipedia.

Oops. Above places C16 incorrectly:

Craig:

It looks like you are shorting out C16. Instead C16 should go across R21 to the summing junction.

Think of it this way, your gain at DC for each adder pathway (e0-e3) should be R21/Rn. What the feedback capacitor C16 does is to short out R21 for high frequency signals where its impedance is small; thus it rolls off the gain at high frequencies, the corner frequency being 1/(2piR21*C16). Just set that, say 5X, above your highest intended signal frequency and you will have smoothed DAC signals and more stability.

Correction:

fig2
Revision B of the 0.4 EC-4P circuit: attempt at a simpler 'conductivity' measurement with adder amp feedback coming directly from the the amp output. See discussion here.

References / Datasheets

INA331

OPA831 / OPA2381

LMC6484

Useful misc info

via Stackoverflow: Command to get latest URL for remote commit:

git ls-remote git://github.com/<user>/<project>.git