µCurrent Offset Voltage Fix

Demonstration of a fix for the potential µCurrent "321" marked opamps, without replacing the opamp.
Not the best solution, but it works.

Another method here:

[MrNukKKT]

I would love an in-depth Fundemantals Friday on poles and opamp response.

[Zeigren]

If you don't have one already a video about poles in regards to op-amps would be very cool

[tenmillionvolts]

2 minute tear down - in Dave's Aussie accent: "Look at this! What a bodge! Ahaha. Instead of re-designing the circuit, they just hung a couple of caps on there. Must have had some sort of problem with oscillation"
Just kidding Dave.
Hi from New Zealand

[dcelctrl]

Absolutely enjoy the detective work. Keep bringing us videos. Thank you.

[allesklarklaus]

opamp stability and poles, frequency compensation and such stuff would be a very cool video!

[Arrowtake]

I am waiting your reply Dave about what you are going to do with my faulty uCurrent.  As you are not answering this question I assume you are not doing a thing. Not happy!

[jerrywh]

Just asking don’t crucify me. Why don’t you just suck it up a replace the faulty product?

[johnfrancisdoe]

Given the layers of indirection between cause and effect, please test that your supposed "worst case" load is also worst case for the problem you are solving. Proposed test: repeat the test on an unpatched 5000-series board to see how it behaves.

[themaconeau]

I'm not sure I want to see what happens when you *do* have the time to build it to scale and paint it :P

[chrismr]

You throw away notes! Still have mine from 1980s, kind of nice to go over once in a while.

[Audio_Simon]

I would love to learn more about opamps and poles. Fundamental vijayos are the best!!

[pault]

Some ST opamps show an R vs C stable / not stable plot, at least for low values of C

[LutzSchafer]

Well Dave in general, an internally compensated op amp can be most critical at unity gain e.g. as a follower because the closed loop is at max. Gain. It needs to be ensured that the poles of the internal stages combined do not cause a phase shift that gets close to the Nyquist point in the complex transfer function. Not sure if you are familiar with the book from Herpy (Hungarian author but book is in English ). There are some examples shown on a classic 709 where it can get critical. Also the very popular (in my time) 741 or jfet 740 version which are known to be stable may oscillate on reactive loads. But your case should be fine except for the internal compensation. I suspect the internal phase shift cap is too small. You could build a test rig and plot the closed loop gain as a Bode or Nyquist plot. In the Bode you should be able to see each (might only be one) pole which will shift with capacitive load closer to the critical (-1, 1) Nyquist point. With the schematic of the op amp based on the location of the poles you should be able to estimate the actual capacitance of that internal cap (which often is a mos cap). Yeah but after all on such popular known to be reliable opamp this should not happen.

[tsusec]

what about the method that is used in audio amplifiers, where you put Resistor and Capacitor in series from the output of the op-amp to the ground?

-In many datasheets they " say " that this will prevent oscillations at higher frequencies

[v.kritin]

Please do a video on op amp stability and how poles effect it etc. It would be very interesting

[bernardosilva]

Why not add those two capacitors on the Revision 6 of the board to avoid those manufactured ICs issues?

[krissolshaq]

@EEVblog2 Is there any chance that you can make a video (as some sort of continuation of your opamp tutorials) about loop stability, phase margins etc.?

[gdj]

Hi Dave
I've been where you are with the same chips but from different manufacturers .
The purchasing department saved £ 0.10 p per chip .
It cost ££££££££££££ in testing & rework !

Looking at the map my starting point would-be change C1 for 1 or 10 uf capacitor &
add a 100 nf across R7 .

The problem could be battery condition dependant
( testing required here ) .

Don't grab the wagging tail it may turn & bite !

[mdrew]

Why not just replace the op amp or is the older one not available anymore?

[hormiguero]

My uCurrent works in microAmpe and nanoAmp but doesn't work in miliAmp, I used the first schematic. I read a solution but i don't know, i don't want crash this....What happens if I put a 100nF cap in the output?