I am working for a guy who has some PCBs he designed. He has asked if I can 'stress test' his pcbs to see if the traces will sustain a fault current. His 24v AC enters the PCB on one side and leaves on the other.
Basically, he asked me to stick a resistor (he gave me some 1/4w ones lol) across the 24v AC supply to the board.
The actual AC supply on his bench is 28.2v AC, which if you said a fault current of say 5A.... isn't that about 140W? Toasty.
I was thinking of a simple 150W, 5.6R resistor but that is basically sticking worst case straight across the board.
A much better solution would be some kind of adjustable load with maybe a readout to get some idea of what is happening.
I found this...
Wondering if I can 'tidy' this up into a nice bench unit. Unless, anyone can point to an off-the-shelf option that doesn't cost more than my last divorce
Then the test is easy. Run 5A through the traces for 8 hours and see what happens.
You will have to come up with some acceptance criteria for pass/fail
My first inclination would be to determine what the fault current could be. i.e. what would fail to cause this current. Then I would use the constant current capability of my power supply to send that much current through the traces.
That of course assumes there is such a predictable failure mode. I would think a short that takes down the supply but not before generating a large current spike.
Also you need to understand the over current trace failure mode.
A current going through a trace is high enough to generate heat in the weakest trace.
As this current heats up the trace (I2R loss) the temperature of the trace increases.
At some point, if the trace temperature exceeds Tg of the board (glass transition temperature) the bond between the trace and board is compromised.
At this point, the trace likely lifts from the board removing the heat sink effect of the board allowing the trace to get even hotter....... this cascades until the trace finally gets hot enough to melt open.
Yea. This whole request from him seems a little rough around the edges
He had 1000 PCB's made.... probably should have done a bit more research and testing before getting them made.
Makes me wonder if any have failed.... otherwise, why is he asking.
There are loads of testers on Amazon, sold as battery discharge testers. Same concept I assume (in constant load mode).
Are you product testing, or testing the design. I.e. will every item shipped be subjected to this load, or will you take one, and test it (perhaps to destruction)?
He already installs them. For some reason, he has asked me to test one.
I asked to what degree and he just said roughly 5 amps. I suppose that is what I do.
There's little point in 'testing' traces for a current of 5A. You already know how that will pan out based on copper thickness and trace width. You can still verify if theory holds true, but there's little point in it.
EMC testing etc, I can get behind that. But seeing what happens if you run 5A through those traces...nah. Either he chose an appropriate trace width or he didn't. In case of the latter, he'll have to bin those 1000 PCB's and start over anyway. Or get insurance that doesn't ask any questions and get rid of his conscience.
Yes.. the relay load traces are part of the traces design. And not big enough in my opinion for 5A.
I will have to try and get the Gerber file and take a closer look.
Well, even fairly narrow PCB traces can carry surprisingly high currents as long as they're not too long. Even without a Gerber file you can get some insight with a caliper and a calculator.
I use one of my lab power supplies and control it by setting the voltage at some nominal value and adjust the current to zero. This puts the power supply in current mode. I then connect it to the DUT (Device Under Test) and slowly adjust my current to where I want it. This is typically a DC test but I can do it with an AC lab supply but normally do not bother the DC is good enough for me. You can easy do TTD (Test To Destruction) using this method.
Why are you focused on 24VAC? Ohms Law applies. If you use a 48V supply and drive 5A through the traces using a LARGE 10 ohm resistor, it's the same as 5V, 5A with a 1 ohm resistor, and 24VAC(presuming RMS) with a 5 ohm resistor - the only possible difference would be if there were traces SO close together that 48V might jump the gap, where 5 might not. That gap is infinitesimally small, so if that's what he's looking for, there's a more fundamental design/assembly issue.
