El-cheapo 5V "LED" PSU - Yay or nay

[floobydust]

I suggest a Koppla USB charger from Ikea 5V 3.4A
and it looks like a very good design, polymer caps, good heatsinking, good clearances, transformer construction etc.

But more than $2.

OP's power supply, at least it has a fuse. But cheapest electrolytic capacitors so it's not going to last. Missing CM choke and inrush PTC, flux everywhere, crappy transformer quality yeesh.

[Moondeck]

Alright, thanks guys. I am not going to use it then, and go for a different solution (perhaps a 12V plug-pack with a regulator). I would still like to see some testing of this though, just to see. Any ideas of what i can do at home? I've checked for shorts in the transformer with a multimeter, but i guess that's not enough for isolation.

[Moondeck]

To make sure (high possibility, not absolutely) the insulation barrier is not killing you, you need to do at least PI/DAR test with a Megger meter.

That's some expensive equipment Not quite something i would have at hand. Maybe it's a stupid question, but would it be possible to use a electrostatic spark generation device to test if there's arcing anywhere?

I guess i just won't plug it in, and go for a different solution.

[Cyberdragon]

That's some expensive equipment Not quite something i would have at hand. Maybe it's a stupid question, but would it be possible to use a electrostatic spark generation device to test if there's arcing anywhere?

Electrostatic generator's output voltage is way higher than required, and can cause more insulation damage, besides it's not reliable, which means you can't back out insulation resistance values, hence no PI/DAR.

This POS is probably so bad that even an isolation tester would cause it to have a Bob Ross level of a "happy little accident" making burn art on your table.

[Shock]

I order my new and replacement power supplies from a wholesaler that supplies to quality brand manufacturers (as a local replacement) under it's own brand. They have been using the same manufacturer for many years and are locally certified.

I suggest going for something like that or find a reputable brand and purchase through a legitimate re-seller. Over the years I've had several smps fail but none of the new or replacement ones I ordered.

Lately I've seen some really crap supplies on cheap electronics in the country, places like dollar stores and mobile phone kiosks are the worst offenders. I still wish every power supply came with screwed enclosure, would make them easier to evaluate and keep the good ones running indefinitely.

[Moondeck]

Thanks for advice everyone, i know much more about PSUs now. What about a module like this one? Will THAT burn my house down too?
https://www.ebay.com/itm/HLK-PM01-AC-DC-220V-230V-to-5V-Converter-Adapter-Mini-Power-Supply-Module/272811461072

[Shock]

At least they have a website, looks like you can buy direct from them as well. I'm not sure if they are the original manufacturers of their products.

This is a part of buying shit over the web you have to learn to do the research yourself. When you buy from the major (component) distributors in the EU they should have already jumped through the regulatory hoops for you. Buy from China or Ebay and anything goes it's like the wild west, for all we know the Ebay seller is making lookalike HLK Tech products.

http://www.hlktech.net/product.php?CateId=10

[Cyberdragon]

That's some expensive equipment Not quite something i would have at hand. Maybe it's a stupid question, but would it be possible to use a electrostatic spark generation device to test if there's arcing anywhere?

Electrostatic generator's output voltage is way higher than required, and can cause more insulation damage, besides it's not reliable, which means you can't back out insulation resistance values, hence no PI/DAR.

This POS is probably so bad that even an isolation tester would cause it to have a Bob Ross level of a "happy little accident" making burn art on your table.

Not possible. I've tested 0.15mm spacing tracks with 1.1kv and everything is fine. To actually reach breaking down is hard, what we would like to see is the trend of non breakdown leakage, and how it changes over time. If it increases, then carbon deposition is happening and that post is gonna fail pretty soon. If the leakage reduces by a factor of at least 1/2, then it should be at least safe for that test voltage.

Direct breakdown at below 2.5kv dc is really rare even for poorly designed and built power supplies.

I wasn't talking about the tracks, they aren't a problem with this one. I meant the transformer where the primary and secondary could be touching and only insulated by enamel. Bigclive has found this happening from ripped tape or poor winding and turns touching the output leads.