Using CR with scope while measuring voltage of DUT

[innomkr]

Hello,

Excuse me if the question is stupid but I am having hard time resolving the issue.

I have the following test setup:

Current Ranger(powered by the Lipo battery) is measuring the current flow to a supercapacitor on a battery powered DUT.
The output of the Current Ranger is connected to one of the channels of my scope and second channel of the scope is measuring the voltage across the supercap.(GND of the probe is connected to GND of the DUT of course)

I first connected the probe that measures the voltage across the supercap I and could see 3ma current on the OLED display of the Current Ranger which is what I expect.
But once I connected the second probe to the output of the Current Ranger i got over 3A or a dead short...

Can someone with more experience explain to me why this happens? I thought I will be able to use my Current Ranger as a current probe while also measuring voltage on another channel of the scope.

[Felix]

Ground loop, you're overloading the CurrentRanger by means of connecting external earth GROUNDs at two different potential levels in the CR. Thanks to design changes the CR is most likely going to survive this overload although it's not designed to be subjected to long overload events.

I suggest you use a floating/isolated DMM for the voltage reading to separate the GND domains.
Also, it would be recommended to read the guide safety page where ground loops are explained in a bit more detail.

[innomkr]

Thank you for your reply. However, as I mentioned in my post, the DUT is battery powered and the CR is battery powered as well. The GND of 1st probe is connected to the DUT's floating GND and the 2nd probe's GND is connected to the CR's output. How can a ground loop happen in this situation.

PS: I read the safety and proper usage section multiple times and I cannot understand the problem here...

[Felix]

From the guide:

To avoid damage to CurrentRanger and/or an oscilloscope connected to the output, it's important to understand that the CurrentRanger was not designed to be used with the same ground reference from input to output (ie with the same ground connected at both I/O ends). This can cause ground loops and in certain cases can damage this instrument or your oscilloscope/test equipment.

You said you are connecting earth GND to measure the input voltage and the output of the CR. So you are using 2 probes that are both EARTH GROUNDED.
IOW you connect:
- one at the output, biasing the input of the CR circuit to EARTH GROUND potential
- one at the output, biasing the output of the CR circuit to the same EARTH GROUND potential

=> Violation of above statement. You are shorting the input/output. Now I hope it's more clear?

[innomkr]

Now it makes sense. Thank you for the comprehensive answer!

[robber57]

Just read this as i am planning to do more or less the same.

I conclude that it would be possible to use 2 channels of my scope in differential mode to measure voltage at the dut while at the same time measure the current with the CR?
I guess this would work without the groundloop problem or any other problem?,  dont want to blow up my very nice brand new CR 

[Felix]

robber57,
It would not work because internally the scope shares the same Ground to all probes. Even if this was 2 separate scopes, the same applies, they are sharing the same galvanically non-isolated earth ground.
Hence you are forcing the same ground between the CR output and the DUT common Ground which is at a different potential than the CR output (ground loop). You would need to isolate the Grounds.

[robber57]

But if i use two channels of my scope in "differential mode" i would not use any of the ground connection of both channels but just the two signal connections(the tip of both probes), so no ground wire connected.
in case you are not familiar: in differential mode you use only the two signal connections of two separate channels to measure a voltage difference between the two channels, the setup is done by using the math function of the scope where you let the result of ch1-(minus)ch2 write to the screen.
It is also called " a poor mans differential probe"
Another solution, i guess, would be to actually use a real differential probe?
Or am i missing some factor i dont see?

On a four channel scope i could do both current and voltage in differential mode so none of the measurements would be ground referenced.

[Felix]

I think I read your post too quickly.
If you are sure there is no galvanic path to the same ground then that is fine.
Having any external ground reference between 2 points at CR input/output will essentially short the CR internally and cause stray unintended currents through that path which can cause damage mainly to the MCU and stress other components like the opamps, that is what this comes to and we're trying to avoid when there is no isolation.

[robber57]

Ok, i will soon make a setup to do the measurement in diff mode and see what happens, i think i fully understand what you are saying about the double ground path and shorting the cr.
I will think about an initial test with fused protection on both sides just be sure no funny things happen, i will have to find a supplier with very low value fuses, something like a few milliamps, i know 10 mA exists but is rare.
If am sure nothing funny happens i can remove the fuses and do the real measurement, i will have to dig in as the last time i did such measurements were at school with a philips crt scope, long loooong time ago 😅

Thanks for you time and advise and developing this very nice device 👍🏻
Gr richard