leak or glue?

[hinisa]

hi. psu works, but it was loud and filthy so I opened the case to dust it.

2 caps in the picture has a transparent yellow stuff on top of them but I'm not sure if they leaked or it's some kind of glue.

[R_J]

Re: leak or glue?

That is glue, remove the glue from the one next to the coil to see if it is bulged under the glue, if the top is flat it is likely ok but you can not tell just by looking.
What was loud? noisy fan?

[hinisa]

Re: leak or glue?

That is glue, remove the glue from the one next to the coil to see if it is bulged under the glue, if the top is flat it is likely ok but you can not tell just by looking.

1, 2: Conicon 1009 - 1000uf 10V Low ESR
3: Conicon 1005 - 1000uf 10V

(1) and (3) looks slightly bulged, 2 isn't.
I bought 2 no-name 1000uf25V caps from local store but I'm not sure they'd be fit as low esr replacements.
Also wondering if I can install something in place of C42 which was left empty in factory.

What was loud? noisy fan?

yes, fan runs very fast and loudly. If I can't find a silent replacement fan I'm thinking of undervolting it by splicing from the 5V line.

[hinisa]

Re: leak or glue?

Found an analysis of the psu here: https://www.overclock.net/threads/li...ld-fly.769505/

[R_J]

Re: leak or glue?

Yes those two caps are bulged, and are bad, that is usually due to being close to the inductor that will run fairly warm, The cap at location C42 is likely parallel to one of the other caps, and not required. You may want to replace those two small electrolytics next to the coil as well, they won't bulge like the others but will just dry out from the heat.

[PeteS in CA]

Re: leak or glue?

The cap you labeled #1 is next to an inductor that could get hot and a pre-load resistor that is also dissipating heat. I suggest leaning the resistor a bit farther away from the cap.

Since L4 and L5 have been replaced with jumpers, that P/S probably has very poor spike noise. I recommend looking on the primary side at the capacitors Line-to-chassis and Neutral-to-chassis. If they are there, they should be safety agency approved parts. If they are rated for 1 KV or 2 KV DC, that is a safety problem.

Glancing through that linked review, calling it a turd may be an insult to turds. The review sample had the dangerous DC rated Line- and Neutral-to-chassis - "Y" - capacitors. If the primary side capacitors really are 470uF, they might be adequate for 250W. The MJE13007-clone transistors and heatsinks might be adequate for 250W. It sounds like the main transformer is small for 250W, though, so 200W would be a more realistic maximum continuous output power.

Safety agency approved Y capacitors could replace the cheapies, 560uF or 680uF capacitors could replace the 470uF types, the switch transistors could be replaced with MJE13009-clones, more honestly rated output rectifiers could replace what is there, good quality output capacitors could replace the Canicons, and ferrite rod inductors could replace the jumpers at L4 and L5, but the main transformer would still limit the realistic continuous output power to ~200W.

[Per Hansson]

Re: leak or glue?

I recommend looking on the primary side at the capacitors Line-to-chassis and Neutral-to-chassis. If they are there, they should be safety agency approved parts. If they are rated for 1 KV or 2 KV DC, that is a safety problem.

Glancing through that linked review, calling it a turd may be an insult to turds. The review sample had the dangerous DC rated Line- and Neutral-to-chassis - "Y" - capacitors.

I always wondered why the Chinese do this, who do they just not skip installing capacitors there at all, have they not heard of Muntzing?
Is it a game to try to make the device as dangerous as possible?
Certainly when you watch some videos from Big Clive you might start believing so...

[hinisa]

Re: leak or glue?

Here I present CY1



I'm starting to think maybe it's better if I'd re-assembled this without any recap and only use it as power source for the tire inflator from now on

[Per Hansson]

Re: leak or glue?

Here I present CY1

What happened, is it corrosive glue or has the capacitor shorted?
Using ceramic caps for this type of circuit is just insane!
I'm genuinely interested if someone can answer the question in my previous post!

[hinisa]

Re: leak or glue?

it has only 1 leg left other one looks burnt. psu still works splendidly, though.

[Per Hansson]

Re: leak or glue?

Yea now I see it, there is one behind it that appears to have exploded too.
This is a prefect example why safety rated capacitors are required here.
But also why it is so idiotic to put regular ceramics: they can just be left out and the thing will work fine anyway!
Chinese crap.

[PeteS in CA]

Re: leak or glue?

A PWS can work OK without all those common mode inductors, X caps, and Y caps, but other electronic products, like TVs and radios, might be affected by conducted EMI.

[Per Hansson]

Re: leak or glue?

I know I know, but you think these Chinese seller test for conducted EMI before putting the thing on the market? Lol of course not!

Here is a link hinisa for proper safety rated capacitors.
But you should check that they have the correct lead spacing, and also capacitance rating.
But it gives you an idea what they should look like
https://www.digikey.com/en/products/...2M-VKA/6124900