Switchmode bricks. Not totally poof-proof.

[nomadradio]

So here's something I didn't know how to break until today.

The now-ubiquitous chinesium switchmode "brick" power supplies have simplified all manner of design and repair jobs. They're efficient, pack a lot of power in a small space and they just get cheaper by the year.

We see very few cases of "infant mortality", specimens that check bad out of the box.

This one exposed a hazard I hadn't encountered yet.

The switch on the side marked "110-220". Seems simple enough, just flip it in the correct direction for how you'll use it.

This one popped the breaker on our workbench protected outlet when first installed in a 2990 base radio.
Oops. Had another on the shelf. It went in and worked just fine.

Finally had a look at the failed unit, and found the 110/220 switch was halfway between clicks.

Oops.

A look at the circuit reveals it should overload the bridge rectifier if you do this. Sure enough, a quick check of the four-legged black-epoxy rectifier-bridge module shows shorted diodes inside.

Haven't fixed it yet. A new bridge is probably all it needs.

But fair warning. The switchmode bricks are not entirely foolproof.

Proved that the other day.
73

 

[AudioShockwav]

Thanks for that Nomad.

73
Jeff

 

[nomadradio]

Well, this just gets weirder.

Replaced the rectifier bridge that was shorted internally.

No difference. Still trips a breaker.

Tore it down further, found this under one of the oscillator MOSFETs.

This is what punched the hole in the Sil-Pad.

This is one of the two oscillator (driver?) MOSFETs that pump 50 kHz AC into the tiny stepdown transformer. Just what caused a voltage spike sufficient to do this?

Or is this more like the copper trench in a shaped charge? Blown from the inside out when the MOSFET failed? There was a decent scorch mark and tiny divot in the aluminum heat sink surface where it punched through the gray Sil-pad insulator.

But the particulars don't matter. A failed bridge could maybe explode a MOSFET from the inside. Or a failed MOSFET could overload the bridge rectifier and cause it to short internally.

Don't know.

Don't care. Once this much damage was evident, the power supply was decisively totaled. If it only needed a bridge, it would be working now. But the expense of anything beyond that just can't be justified. Time to move on to the next paying gig.

73

 

[Captain Kilowatt]

Makes me wonder if a small piece of conductive material was sandwiched under the MOSFET and under clamping pressure eventually pierced the silicone pad and shorted it out to ground causing it to arc and cause that damage to the surface.

 

[kc8mob]

I've had great luck with the PSU's I get from Parts Express.

Even if you can't get their dealer pricing, the price is better then what you find just using google.

public price of 46 bucks and it's adjustable up to 13.8 if you want a slightly higher voltage unit that sweeps from 13.9 to 16 or 17 volts to.

Mean Well MW LRS-350-12 12V 30A 350W

 

[codecxbox]

I never had trusted switching psus..they rely too heavily on a clean AC input..which were I live, its mostly unlikely..all my computers have been struck by brownouts and these psus dont tolerate that...by the time a decent battery backup decides to help, its already late..