Hello, I have a pcb from an e scooter which had a faulty diode but even after removing it from circuit, it still has a short to ground on the three components circled. I have removed every cap on the board and tested one by one and the short is still there and all the caps test ok. Could this be because I haven’t put a diode back on it yet or where do I go next please?

 

Hi,
Can we have a picture of the underside of the boards please.
The short is likely to be a semiconductor.
Can you indicate where and how you are measuring the short.
Good luck......

 

This is the underside of the board and could you point out the semiconductor please and how I would test it?

 

I am measuring with black probe on the ground plane and red probe on the three circled components, all three beep both sides of the components.

 

Welcome to AAC

Without removing parts from the board its very hard to diagnose such a fault with just a multimeter; you're looking for resistance differences to ground of a few milliohms. The usual way to do this is power the board at, typically 2 to 3v from a constant current supply of an amp or so and view it with a thermal camera - the faulty part gives itself away by glowing hotter than other parts. Without a thermal camera the other, less effective, way is some freezer spray. Spray the board until well frosted, then power it up and see what defrosts first. Its a bit more hit n miss but does work...

Hope that helps.

 

Where do I put the red and black wires from the power supply on the board please to use the frosty method?

 

Where do I put the red and black wires from the power supply on the board please to use the frosty method?

Start with the normal input... if that doesn't give results we'll have to do more digging..

 

Is that black wire to ground and red wire to 12v pin?

 

Start with the normal input... if that doesn't give results we'll have to do more digging..

I have tested the 117 regulator and Between ground and output measures 0.828K ohms and ground to input measures 51.8k ohms

 

Where do I put the red and black wires to inject voltage on the pcb please?

 

Finding a short across power rails is one of the most challenging repair situations. As @Irving says, you need some special debugging techniques.

1) HP 546A pulser and HP 547A current tracer which fortunately I happen to have.
2) Thermal imaging camera.
3) Cold mist spray to find hot spots.
4) Cut PCB traces at strategic locations.

 

On the 12v input, but I wouldn't use a full 12v. Looking at your pictures it seems that U3 is a switching regulator generating the 5v and feeding the LM117 which generates the 3.3v. What are the markings on U3?

 

The markings on U3 are: IF 8GD also I removed the second to last resistor to measure the resistance as shown in the picture and it’s showing O.L instead of a reading.

 

Is this correct that I connect the power supply wires like this in the picture?

 

The 514 resistor measures out of circuit 0.510milliohms is this a bad component or is that what it should be please?

 

Is this correct that I connect the power supply wires like this in the picture?

Yes

 

Yes

Should I connect at 3v and 1amp

 

Should I connect at 3v and 1amp

Yes

The 514 resistor measures out of circuit 0.510milliohms is this a bad component or is that what it should be please?

I would expect that to be 510k

 

Yes


I would expect that to be 510k

So that is probably bad then judging by the measurement

 

So that is probably bad then judging by the measurement

Maybe, though it is extremely rare for a resistor to fail short, 99.999% of the time they fail open circuit. What did you measure it as? 510milliOhm ie 0.51 ohm? I doubt your multimeter can measure 0.51 milliohm, ie 0.00051 ohm