Self test or test?

Started by fizzyade, November 24, 2021, 08:41:44 AM

fizzyade

What's the easiest and most foolproof way of checking if the current ranger is working within spec?

I've spent days trying to figure out why a later revision of my hardware is consuming too much current.  As a sanity check, I've gone back to a different revision and I'm now seeing odd currents on that as well.

It might be code related, but I need to check if the current ranger is still functioning as it should and that it hasn't been damaged.  If it has been damaged, I have no idea how because I only ever connect the DUT via battery, and if I use USB, I also use an isolator.

I'm baffled as to what the issue is, whether it's the CR, my board, something changing, me mis-reading measurements or anything, I'm so confused right now and looking to prove that the CR is still working correctly, as if it isn't, I'm facing a losing battle from the start!

Edit:

I mean is there a more scientific way than measuring a resistor and a voltage source and then checking the current measurement is correct, I've just grabbed a 680K and have 3.7 volts and I'm going to check the current on my Bremen first and then the CR.

fizzyade

I measured my battery at 4.12V
I measured the resistor at 684000

And both measurement devices agree with the calculation I made and that it's 6uA of current!

So at least I've not been chasing my tail, but for the life of me I cannot figure out what the actual problem is, I'm a bit perplexed why I'm suddenly seeing different currents to when I last performed the tests on the previous revision of hardware, neither the previous prototype nor this one are showing the correct answer, or maybe they are and I was seeing the wrong answer before.

Back to the drawing board.

fizzyade

Just as a follow up, this is where the CurrentRanger has really proved itself.

The hardware I'm working on has been through a number of design stages, its a monitor for filament dry boxes, it's total overkill but I designed it for myself so it was never designed with cost as a factor.

The first prototype I used matrix board and got the hardware up and running and wrote the software for it, I was able to use the CR to get a handle on battery life, the unit spends most of its time in deep-sleep (esp32e) consuming around 12uA.

This latest prototype is taking 70uA, which is a massive hit to battery life, the CR has allowed me to verify through my stages of design if there was a problem, I have eventually tracked down the issue to the voltage regulator which I suspect is a fake, I was only able to verify this because of the accurate and resolution of the readings that the CR gives, if I was in mA range most of the time, I'd never have spotted this problem, but in uA range, the problem stuck out like a sore thumb!

've only had the unit for a short amount of time, but it's already become my most important device.

Felix

I did follow the thread and I'm happy to see you've found the issue. And more so that the Current Ranger has been helpful in identifying it, which is exactly one of the main goals I've designed this tool for. I consistently hear similar testimonies on various channels so thanks for adding to the collection of success stories. And nice board you have there  ;)