Help with failed rectifier

Thread Starter

Mastiff37

Joined Mar 14, 2014
3
I'm trying to debug a broken welder, and I see that there's a (4 diode bridge) rectifier circuit that is no longer working. I'm at home with a multimeter trying to figure out where the problem is. I found this on-line, but this is basically the circuit:



The fact of the matter is that it was working when I was debugging another issue and I shorted something and broke the rectifier. The AC is coming in from a transformer and was about 20 volts. I was measuring DC voltage at the output, but now am only seeing AC. The cap "seems" okay in that it is not shorted, and I can watch resistance readings go up and down from the charge coming from the multimeter.

The thing I can't explain is that all four diodes read open in one direction, and about 500 ohms the other. I expected them to read almost no resistance in the forward direction. Does it makes sense that good diodes would have so much resistance? Blowing them all to the same state seems improbable.

Thanks for any help.
 

inwo

Joined Nov 7, 2013
2,419
Do you have "diode test" on your meter?

Diodes will not "turn on" with most ohm meters.
Diodes fail "shorted" normally.

Rather than wait! ps.

Do you have ac volts across tx output?
Are you checking for dc across cap.? (not from ground)

Diodes would show damage from short.
Damage tx winding or burnt traces are likely.
 
Last edited:

Mike33

Joined Feb 4, 2005
349
Right, like Inwo suspects....with bridge removed (transformer secondary open) you can measure the voltage at the secondary, see what is going on. AND you can smell for burned insulation and listen for crackling etc.

With a welder, I'd suspect an issue with the windings - someone ignored duty cycle and got it very hot, melted some wiring insulation, caused a short, blew the rectifier, etc...
 

Thread Starter

Mastiff37

Joined Mar 14, 2014
3
Thanks for the replies. I'm not sure I follow though. All this stuff is on one little PCB. I can see the AC voltage coming in (across 1,2) same as when it was working, I just see no DC (at 3) like I did before. The transformer must be working, I think, or I'd have no low voltage AC like I'm measuring.

It's such a simple circuit, all I can think of is bad diode, bad cap, bad resistor (measures 1k) or maybe shorted load. That last point I didn't think of for some reason, but I think I can disconnect most of the big loads and see.
 

inwo

Joined Nov 7, 2013
2,419
Thanks for the replies. I'm not sure I follow though. All this stuff is on one little PCB. I can see the AC voltage coming in (across 1,2) (what value do you measure from 1-2?) same as when it was working, I just see no DC (at 3) (you must measure DC across capacitor)like I did before. The transformer must be working, I think, or I'd have no low voltage AC like I'm measuring.

It's such a simple circuit, all I can think of is bad diode, bad cap, bad resistor (measures 1k) or maybe shorted load. That last point I didn't think of for some reason, but I think I can disconnect most of the big loads and see.
We aren't there. If you don't answer our questions, we can't help you.

1 and 3 are mixed up on the bridge. It makes discussion dificult.
 

Thread Starter

Mastiff37

Joined Mar 14, 2014
3
We aren't there. If you don't answer our questions, we can't help you.

1 and 3 are mixed up on the bridge. It makes discussion dificult.
We must be talking past each other. I said I measure 20V AC at the input. I measure no DC at 3, i.e. across the capacitor. You said across the cap not to ground, but the cap goes between ground and 3, so what's the difference? Thanks.
 

#12

Joined Nov 30, 2010
18,224
A bit late, but my 2 cents: Ohm meters don't work well measuring diodes. They have random voltages and currents, depending on the manufacturer. You need to use a, "diode" setting if you have one. Otherwise, you're stuck measuring voltages like inwo has you doing.
 

inwo

Joined Nov 7, 2013
2,419
We must be talking past each other. I said I measure 20V AC at the input. I measure no DC at 3, i.e. across the capacitor. You said across the cap not to ground, but the cap goes between ground and 3, so what's the difference? Thanks.
Language sometimes fails. Reason for being specific!

1 & 3 are not labeled correctly on schematic. Makes it necessary to use points other than schematic labels.

No diode check on meter? What meter do you have?

Check with low ohms scale from the bridge plus to cap plus and bridge minus to cap minus.
 

Richie121

Joined Jan 12, 2014
27
If the 12V supply is from the main welding transformer, then there is a lot of current available - enough to blow wires joints and traces like a fuse. Check them all - particularly anything through-hole.
If the diodes are small then I would replace them as a matter of course. Cheap components that cost you time. If large then test with a torch bulb to establish if they are open or blocking properly.

If you like small projects then do as I did, and epoxy a 40° (NO) temperature switch to the coils, and use an old fan from a microwve oven to cool the coils when they get warm. Now I have the turbo model :D
40° sounds a bit low but you want the fan on early. There should already be a high rated thermal switch for forced shut down if it gets too hot.
 
Top