Blown Current Ranger - Potential easy fix? [Fixed!]

[tealbrains]

Hi,

First of all : Thanks for the great work!

[Before I start describing the problem: I did read the instructions before using the CurrentRanger]

I have been using a current ranger for a few days. The only part I was using was the Load side of the Current Ranger. No issues so far.

I started using an USB oscilloscope to monitor the voltage (3.3V bus) on my DUT. Then I wanted to see the correlation between current and voltage of the DUT so I used the output of the Current Ranger and it got damaged, but it does not seem to be too serious.

Felix, if you see this post, I am providing all the information I have on hand, if you need more please let me know. I would attempt a fix right away but prefer to ask you first to make sure I don't completely destroy it. I think I have been lucky.

Here is a diagram of my (problematic) setup:


I saw smoke when I plugged Channel 1 of my oscilloscope.

• Note that my DUT is running off a battery.
• I think the issue is potentially due to having connected Out- (Ch1) and GND (Ch2) [see purple arrow in previous image].
• It looks like I allowed 12V to be connected to GND of my DUT via oscilloscope). I should have left one of my GND reference of my probes disconnected. (see blue lines in picture below))
• The result is a blown trace. (see Orange highlighted trace in image below)

Other observations:
-No chip seems to have released magic smoke.
-The SAMD21 is working, the device turns on, the OLED is displaying the usual stuff.
-BUT the readings swing between 0 and max value (looks like it is due to a loss of reference because of the blown trace) and the output follows the reading of the display, swinging rail to rail.



Fortunately it looks like the thinnest trace in the path of destruction blew up, which fortunately left the rest of my gear undamaged. Luckily this was on the CurrentRanger and the fix may just be a wire to replace the trace (and avoid making the same mistake after that). Am I correct or should I be checking something else?

Kind Regards,
Gabriel

[james_s]

This is a classic noob mistake with oscilloscopes, at least be thankful that it was low voltage without a lot of current available, some people blow up their scope by doing something similar. The ground on both probes is connected to the same point so using standard probes you MUST only measure signals relative to the same ground. If you want to measure something the way you were doing, you need differential probes.

[tealbrains]

This is a classic noob mistake with oscilloscopes, at least be thankful that it was low voltage without a lot of current available, some people blow up their scope by doing something similar. The ground on both probes is connected to the same point so using standard probes you MUST only measure signals relative to the same ground. If you want to measure something the way you were doing, you need differential probes.

I thought the input on the current ranger would be isolated from the output for some reason, just measuring current flowing through a sense resistor in differential mode. That does not excuse my mistake of course, I should have checked the schematics and applied the usual approach with the probes.

[Felix]

The most important thing to avoid is having 2 of the same earth ground or ground reference attached to the CR on 2 sides - either input+output, or input+USB.
For USB its easy to add isolation as shown in the guide.
For the output, you'd need differential probes or an earth isolated power source for your scope, to float it's ground from anything else in the circuit.

If the blown trace didn't take anything else with it, then just find 2 points/SMD pads where it connects, and add a short thin jumper wire and you should be good to go.

[tealbrains]

Thanks for the reply, I ended up biting the bullet yesterday as I had very little to lose anyway. I soldered a wire to replace the trace and everything is back to normal.

[Felix]

Glad to hear that, thanks for the update.

[Jakk]

Hi Felix, could you please confirm that the OP was correct in that the problem in the image, purple connection, was the fault?

Thanks

[Felix]

confirm that the OP was correct in that the problem in the image, purple connection, was the fault?

Most likely. Instead of studying what might be faulty, I will point you to what is correct, and all of that is explained in the guide: https://lowpowerlab.com/guide/currentranger/safety-and-proper-usage/