Op amp input offset voltage tester

[cellularmitosis]

Heyo,

I created a board design for measuring the input offset voltage (Vos) of various DIP op amps.

Github: https://github.com/pepaslabs/opamp-vos

Gerbers: https://github.com/pepaslabs/opamp-vos/raw/master/kicad/releases/v2.1/gerbers.zip

Schematic: https://github.com/pepaslabs/opamp-vos/raw/master/kicad/releases/v2.1/opamp-vos.pdf

I'm using the circuit diagram from http://www.analog.com/media/en/training-seminars/tutorials/MT-037.pdf

I'm using the "sijosae" discrete rail splitter, so that you can run it from a single supply (Barrel jack and 0.1" header supply connections are supported).  https://tangentsoft.net/elec/vgrounds.html

It supports 4 common footprints: uA741-style single, OP07-style single, dual, and quad.

The Offset voltage is measured by inserting DMM probes into plated through-holes in the board.

For the single op amp footprints, I've included footprints for adding resistors and a trimmer to trim the offset to zero, as well as terminals to measure the resulting resistance values which were required to reach zero offset.  Additionally, the trim circuitry can be removed via jumpers.

The board can be populated with different resistance values in order to achieve different gains.  I've added some square silkscreen areas for writing on the board with a sharpie to indicate gain or other notes.

This seems to work, but I've noticed two issues so far: 1) the uA741's I've tested seem to be pretty sensitive to EMI (hand waving), which I'm guessing is due to the trim circuitry acting as an antenna, and 2) the plated through hole probe points seem to introduce a bit of a measurement repeatability issue (wiggling your probes to get better contact etc can change the reading a bit).

Board dimensions are 50x100mm, so probably too expensive to order through OSHPark.  dirtypcb's might be a better option.

Feedback is welcome, enjoy!  I'll also post a few measurements I've taken with it.

[cellularmitosis]

Oops, just realized I bypassed the feedback resistor of the first two op amps with 10nF, but used 100nF for the other two op amps.  No big deal (this is a DC application anyway).

[David Hess]

the uA741's I've tested seem to be pretty sensitive to EMI (hand waving), which I'm guessing is due to the trim circuitry acting as an antenna

741s are not any more sensitive to EMI than other operational amplifiers; be sure to bypass the feedback resistor and the non-inverting input.

If you add a couple of resistors in series with the inputs with switches to short them out, then you can also measure input bias current of each input and input offset current with the same circuit at least for high bias currents; this is a little tricky to do with FET and super beta inputs.

National application note 24 is worth studying.

[floobydust]

Maybe try adding a small cap across the 10k feedback resistor to swamp the (stray capacitance) effects of the PCB trace running from the op-amp output near pin 3 (+in).
At gains of 1,000 usually I find all kinds of RFI shows up unless bandwidth is limited.