How to measure small current waveforms using a uCurrent Gold and a mains grounded oscilloscope

Question

I am attempting to use a uCurrent Gold as referenced by several SE posts such as measuring microampere current without noise with a mains-grounded oscilloscope, to observe the small current into an inverting opamp input, for the first stage of a close reproduction of David Ashton's Pulsing Led for Competition.

My setup is diagrammed below, and I reason this is not the correct setup. The oscilloscope I'm using is a mains-grounded scope, so the connection to the V- jack on the uCurrent is shorted to the I- jack, and that represents a path to mains ground.

Referring to the diagram below, the points I marked as P1 and P2 (purple color) are normally shorted, but to insert the uCurrent, I've opened them up and inserted in the current inputs to the uCurrent. The ground wires for CH1 and CH2 of the scope, along with the connection between V- and I- on the uCurrent, result in shorting points P2 and P3 to ground, thus shorting the capacitor to ground, which is not intended.

Basically, I've concluded that the dashed green lines in my diagram below are all at equipotential, which is an incorrect test setup.

And, I reason that, had this circuit used a large separate mains-grounded (not isolated) power supply, versus what it uses which is a CR2032 battery, I would run the risk of damaging my scope (reference EEVblog #279 - How NOT To Blow Up Your Oscilloscope!).

Is my assessment correct? If so, are all of the following approaches viable? :

1. Buy a differential probe (EEVblog #279 offset 1214s), or
2. Use two different scope channels, one for the uCurrent V+ and a different one for the uCurrent V-, leaving all grounds at the same equipotential, and avoid hooking any of the scope's grounds at point P2, and using the scopes math function to subtract the two signals, effectively resulting in a voltage differential being plotted (only viable if I have a > 2 channel scope), or,
3. Just stop using CH2 on the scope to view the voltage on top of capacitor, as that takes out the short at point P3 (but I don't want to do do that as I want to see how the small current going into the inverting op-amp input changes (or doesn't change much at all) along with the voltage on the capacitor).

See also uCurrent with an oscilloscope for a video of someone else struggling with what seems to be a similar issue.

Here is the diagram showing the circuit under test, the latest uCurrent Gold schematic, and the front diagram of the Rigol scope I'm using with connections (high resolution image at imgur):

Answer 1

I also have a uCurrent Gold and it is super handy.

I also have run into the problem of wanting to measure the current between two points, neither of which is ground.

Since your circuit under test is battery powered, you could just connect the ground of your scope and the uCurrent to P2 - if you are willing to drop the voltage measurement at P1. The current measurement will be right.

When I really want to see the voltage and current at the same time, I use two channels on my 4 channel scope with one at P1 and one at P3 and then I difference them to generate a math trace showing the voltage at P1 relative to ground on the battery powered circuit. I then put the uCurrent output voltage on one of the other remaining channels.

Answer 2

Posting an answer that I'm not sure how good an answer this is compared to just using a costly differential probe. See the diagram below:

This approach connects the scope probes such that the arbitrarily chosen point (point P4 in the diagram) in the circuit to connect the scope grounds to should be some point that is NOT also connected to any of the scope probe signals, or to any of the scope probe grounds. This does connect mains ground (scope ground) to that same point. Since the circuit ground needs to be one of the channel signals, the scope probe grounds should not be at that same circuit ground, as then that would change the circuit to short out the capacitor.

This does require all 4 of the scope channels to pull this off:

1. The CH2 - CH1 signal -- Using the scope to subtract the voltage levels of CH1 from CH2, to read the uCurrent Gold's V+ - V- to give the millivolts that are equal to some level of current (depending upon the switch on the uCurrent).
2. The CH4 - CH3 signal -- Using the scope to subtract the voltage levels of CH3 from CH2 to read the voltage on the capacitor relative to circuit ground, NOT relative to mains and scope ground.