the gutless, bloated, and fried power supply hall of shame

[Topcat]

Re: the gutless, bloated, and fried power supply hall of shame

Here's a super junky PSU, reported here at Per's request!

The junker in question is a "Manhattan PUF405S" ATX power supply.





While this supply was working when I pulled it from a system that was given to me, it's definitely a cheapo. The first thing I noticed were the very thin/flimsy wires. I peered inside and noticed about 1/3 of the space inside the case was empty....and this thing weighed less than a happy meal....



Opened up.....and even more embarrassing!! Rated at 405w, look at those puny sinks....





Then WTF is this goofy air deflector on the fan?!? Never seen anything like that on all my years....







More of the puny sinks. Secondary caps are Samxon GF series, known to be a bit problematic....but these appear to be good...



Now to scope the outputs.... They were a little noisy, but not the worst I've seen. When warmed up, the noise did settle down a little.....





The top is CH-A; the 12v rail...the bottom on CH-B; the 5v rail.



Recapping it would likely smooth that out....but this one is just not worth it....I just wanted to autopsy it.

[Agent24]

Re: the gutless, bloated, and fried power supply hall of shame

This guy was ridiculous. He'd even say it outright regularly in class when anyone asked questions about doing something in a different way,
"Don't argue with me, this is how it is because I tell you so."

Did he offer the class Kool-Aid as well?...

[momaka]

Re: the gutless, bloated, and fried power supply hall of shame

I don't need to look at HE series datasheets to know the available voltage ratings don't go as high as 450V.

^ Exactly!

Haha, fake teacher, fake caps!

Well, I wouldn't go as far as calling him a fake teacher. He actually knew his stuff, at least when it came to programming microchips (what that class was.) Problem was his ego - if you didn't learn or do everything exactly the way he wanted you to learn it or do it, he'd disregard you and even become mean to you... which is what happened in my case, as I found a way to do one of the projects without using his software button de-bouncing routine (since it wasn't stated anywhere on the project requirements sheet that this had to be exactly there). Instead, I implemented my own way, which worked perfectly... if not better, since my code ran in predictable timed loops rather than going haywire with interrupts and crazy line calls/goto's like his.

Anyways, long story short, he didn't want to admit that my code actually worked and even refused to look at my project based on that assumption. This would have meant to re-take the class... and I was about to graduate, so "I ain't 'aving it" (as PhotonicInduction would say in some of his videos, minus the hammer part lol... though I almost got mad enough that time to kick his a**), so things got pretty ugly between us (verbally) until I got the dean involved. But anyways, glad that's over, and I guess this POS power adapter is the only thing that will remind of what I had to do to get done with that silly class... or silly teacher, rather.

At least half the teachers (in my experience) try to be more like slave-masters than anything else. If you don't do exactly as told, you'll get a bad grade. "Critical thinking" be damned!

Interesting.
For me, it was really just that one teacher for the most part. I mean, I've also had a few others with a somewhat elevated ego's too, but not in that uni/school (mostly just in high school) and not anywhere close to that level. This guy was ridiculous. He'd even say it outright regularly in class when anyone asked questions about doing something in a different way,
"Don't argue with me, this is how it is because I tell you so."

[zunasthegreat]

Re: Another Supernight

Hmm that "cap maze" looks familiar, its probably some low quality OEM. Most common now as ChongX or ChengX.

[PeteS in CA]

Re: the gutless, bloated, and fried power supply hall of shame

But what's "special" here, is the primary 68µ 450V cap.
It's another fake Nichicon, with a "Y" shaped vent, concentric-circle bung, "HE" series, silver print (instead of white), welded-on leg extensions, and wildly incorrect date code.

I don't need to look at HE series datasheets to know the available voltage ratings don't go as high as 450V. The HE series is very good, but that is a very poor counterfeit, and a recycled used part to boot.

[ben7]

Another Supernight

Another day, another Supernight!

This one is also popped - blown fuse, shorted primary FET, and blown caps (see below). The case was slightly melted, but I didn't bother photographing it as it's nothing unusual now.

Surprisingly the 68µ cap measured about 70µ. I wouldn't be surprised though if the ESR is a bit high - it's likely dried up somewhat.
The PWM electrolytic (rated 47µ) measured about 2µ, and after just a moment the meter began freaking out, autoranging back and forth.
The popped 1000µ measures only about 130µ and its sibling measures ............ ~4300µ.

But what's "special" here, is the primary 68µ 450V cap.
It's another fake Nichicon, with a "Y" shaped vent, concentric-circle bung, "HE" series, silver print (instead of white), welded-on leg extensions, and wildly incorrect date code.
However ......... This one has hemorrhoids!

(Dunno why it pooped itself, looks like the bung melted and bubbled only in a localized area. The top of the cap is not bulged, and the legs are still firmly in place; I've seen this before on high voltage electrolytic caps, but the legs always corrode through and fall off.)


https://www.youtube.com/watch?v=CKjaFG4YN6g

~~~~~~~~~~~~~~~~

... I suspect these adapters were bought in bulk for the class and modified by the teacher who teaches (and manages) that class, since he also designed the programming boards too (which were full of cheap Chinese parts and what seemed to me fake Rubycon caps, BTW…) This teacher was also the only one I've ever had problems with in my academic life (calling my project “bull----“, literally)… but that's another story. (It ended OK once I had the dean involved, that's all I'll say here.)

Haha, fake teacher, fake caps!
Lemme guess, the project had too much toxic masculinity?
It's no wonder why our schools are crap these days - we pay the highest prices for the least amount of "teaching."
At least half the teachers (in my experience) try to be more like slave-masters than anything else. If you don't do exactly as told, you'll get a bad grade. "Critical thinking" be damned!

-Ben

[Behemot]

Re: the gutless, bloated, and fried power supply hall of shame

Perhaps not as terrible as much of the other stuff in this thread, though the c(r)ap selection mainly consisting of CrapXon and Teapo isn't exactly great... reason this one is fried is that one time I was trying to patch up some traces on an old broken Slot 1 board and used this PSU for testing; after doing so it turned itself on just upon plugging in the ATX connector and promptly released some hefty magic smoke . I was actually hoping to find which part had exactly burned up but no luck.

I'll note this unit received overall very positive reviews locally back in the day, praising its build quality amongst other things. That little screwed on PCB does look pretty rough though... in the end this was only a ~40€ unit though so can't expect too much I guess.

Yeah it's an old FSP platform so on the better side, just their capacitors choise is dubious since…ever

[momaka]

YT-1208 power adapter

Here's a really lame one:

It's a “YT” –branded (read: no-name) power adapter that was included with the programming boards we had to buy for one of our university classes (some years ago back when I was in uni.)

Let's look at the label:

Well, there's not much to it. A bunch of symbols on the bottom indicate that this is a double-insulated device ( ), for indoor use only, and not to be opened (risk of electric shock.) OK… what about safety approvals? - Bah, ain't nobody care about those!
This aside, the adapter is rated for 110-240 VAC input and 12V, 1 Amp.on the output. The red ink marker was actually someone crossing out the “12V” print and writing “15V” below it… which is correct – the adapter does output 15V. I suspect these adapters were bought in bulk for the class and modified by the teacher who teaches (and manages) that class, since he also designed the programming boards too (which were full of cheap Chinese parts and what seemed to me fake Rubycon caps, BTW…) This teacher was also the only one I've ever had problems with in my academic life (calling my project “bull----“, literally)… but that's another story. (It ended OK once I had the dean involved, that's all I'll say here.)

Anyways, let's look inside this no-name adapter…




Well, at least it was easy to open with just a single screw! Case is quite crappy and made of rather cheap, thin plastic. Input filtering is… where? We just get a small glass fuse and thin (24? AWG) input wires acting as a secondary fuse.
“Bridge rectifier” is just four 1N4007 diodes. Input cap is a 400V, 10 uF unit made by Chong (CD11x series… or whatever that means.) Startup cap is a 35V, 47 uF and also Chong CD11x. But hey, on the positive side, at least there is a PWM-FET IC in there: an ST VIPer22A. As for whether or not it's a genuine ST part or not… I heard you like Russian roulette?

That aside, isolation is probably the biggest concern with this adapter. First, note how small the transformer is. Of course the PCB silkscreen has marks for a bigger one, but a smaller one was used. And to make matters worse, there is a 1 nF 1 kV -rated ceramic cap between primary ground and secondary ground. Such cap is used for reducing EMI and RFI. Because of the way it's connected, it has to be rated Y1 or Y2 class… which this one is NOT.

Let's move to the output/secondary side…
Output rectifier is an SB3100 (3 Amp 100V Schottky), with the output DC filtered by a 16V 1000 uF 10x18 mm cap and a 16V 470 uF 10x13 mm cap - both being Chong “LOWESR” CD110x series. A PI coil is also present between them. Finally, the output wires: 20 AWG, 600V. – Yes, “they” used 600V –rated wires on the output but only 300V on the AC input. Perhaps this is to protect the AC line from getting shocked by humans?

Last but not least… remember that red marker I talked about on the label that indicated the adapter as being 15V? Well, have a look here:

^ I don't know if this “work” was done by said teacher above (or he might have had students do this for him “for experience purposes”, as from what I learned he also had certain parts of the programming boards done this way.) But either way, that's a craptastic job for sure. I wonder how many adapters arced over and burned out from this hack job.

So the above mods were done to convert these adapters to 15V for the programming board we used. On that note, I haven't tested what output current it can sustain. But if I remember correctly, I had no problems pulling up to 500 mA a few times out of it. I also remember a few slightly more careless students who happened to short the output on these power adapters on their breadboard. Miraculously, none of the adapters blew and only shut-down gracefully. So looks like that VIPer22A PWM-FET IC is doing its job properly. Too bad it can't do anything about the lame small transformer and poor isolation with the 1 KV ceramic cap. And while I'm complaining here, I must say that red LED indicator lamp is brighter than a 1000 suns! This may not be a bad thing, though – at least you get a bright RED (warning?) light you that the adapter is plugged in, as if saying DANGER! DANGER! Don't leave me plugged in!

On that note, I haven't been using this power adapter for anything other than a few brief tests and nothing where I was grounded. I simply don't trust it. With those caps, the non-existent input filter, and that lame pri-sec isolation, this thing probably won't be used in anything critical. However, I might re-build it one day just for fun - that is, get new caps, replace the 1 KV ceramic cap, install input filtering + new wires, and see if the transformer primary and secondary windings are isolated properly from each other, along with remove those resistor hack-jobs. Oh, I mustn't forget taming the output of that red LED. No idea when (if ever) I'll try that, though. But just throwing the idea out there, as the design with the VIPer22a IC is otherwise decent. At least this one is not a silly single or 2-transistor self-oscillator without any protections.

[momaka]

Re: the gutless, bloated, and fried power supply hall of shame

It turns out, they INDEED are using recycled/re-sleeved capacitors!

Sad to see what we've come to in this world

I am all about re-using and up-cycling when possible... but some things just need to be buried and left to RIP. Those crappy recycled caps should go under this category.

That said, the bung on one of them with the extruded "T" stamp looks like it indeed might have been some legit known-name brand before, as you noted.

I guess you should give it to your local electronics recycler for them to send it back to China... or wherever else they still send these things nowadays... and after some time, maybe it will be back in another crappy no-name adapter

The output caps measure about 1800-3000uF, which suggests to me that they are failing (despite not bulging) due to high leakage current. (They're rated 1000uF 25V, KSJ brand.)

Yeah, that's most definitely high leakage current / electrolyte "eating" the aluminum foil. Give them the full rated voltage printed on their sleeve and watch them pop. I bet very few (if any) will survive. (And that's if their sleeves weren't telling a lie about the rated voltage, which could be the case too.)

^Unit overheated, melted case (not shown), switching FET shorted and the re-sleeved capacitor has evidently seen better days.

In other words, nothing is salvageable from this adapter.

Did a little writeup on a dead bequiet PSU over in the other thread, but I figured I might as well post this other one from the junk box as well before disposing of it... this is a bequiet Pure Power L7 430W bought in 2011. This one never saw much use other than for occasional testing.

Save the X2 and Y2 caps, along with the line chokes - these can come in handy when fixing other PSUs. If you have a desoldering iron with vacuum (or just manual vacuum pump), try saving the rectifiers too - they make nice upgrades for monitor power supplies that have undersized rectifiers.

[Xan03]

Re: the gutless, bloated, and fried power supply hall of shame

Did a little writeup on a dead bequiet PSU over in the other thread, but I figured I might as well post this other one from the junk box as well before disposing of it... this is a bequiet Pure Power L7 430W bought in 2011. This one never saw much use other than for occasional testing.






Perhaps not as terrible as much of the other stuff in this thread, though the c(r)ap selection mainly consisting of CrapXon and Teapo isn't exactly great... reason this one is fried is that one time I was trying to patch up some traces on an old broken Slot 1 board and used this PSU for testing; after doing so it turned itself on just upon plugging in the ATX connector and promptly released some hefty magic smoke . I was actually hoping to find which part had exactly burned up but no luck.

I'll note this unit received overall very positive reviews locally back in the day, praising its build quality amongst other things. That little screwed on PCB does look pretty rough though... in the end this was only a ~40€ unit though so can't expect too much I guess.

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

Thank you for the warm salutations, momaka and goodpsusearch.

Which is interesting, because at a glance, the PCB appears to be well-designed in terms of spacing and isolation distances. But that's where the good ends. The bad: secondary heatsink is extending over to primary side and is not covered/isolated and sitting to close to primary-connected components. Same appears to be going on with the primary heatsink over the secondary side. And to top it all of, the blue EMI/RFI cap between primary (-) bus and secondary ground appears a bit too thin, suggesting it may just be a regular 1 or 2 KV -rated ceramic instead of proper Y2-class. So most likely, this adapter would fail a proper hi-pot test.

It uses a real Y1 type capacitor, according to the cap itself. It's only 2.2nF, which may explain the small-ish physical size.
The transformer also doesn't use doubly-insulated wire for the secondary, so it still likely wouldn't pass.

~~~~~~~~~~~~~

I was "gifted" two more of those "SUPERNIGHT" W-T5000 adapters.

It turns out, they INDEED are using recycled/re-sleeved capacitors!
The sleeves are the same and they all have the welded-on lead extensions, but the bungs are different. On one unit, the bung is like that of an NCC (Nippon Chemi-Con) cap. Interestingly, it hasn't bulged, but it seems to have gone high-impedance or has been abused (?overheated externally?), as its sleeve cracked. This PSU also had a slight melt/droop on the external case, and the PCB was browned, but not as badly as the PSU-1250-D4 (Not "SUPERNIGHT" branded) in one of my previous posts.
They all measure 60-70uF. One cap has slight corrosion around the negative lead.

All the other caps are the same across all 3 units. I measured them also:
The PWM chip electrolytic is 47uF, but they all measure about 20uF or so. The output caps measure about 1800-3000uF, which suggests to me that they are failing (despite not bulging) due to high leakage current. (They're rated 1000uF 25V, KSJ brand.)


^ Three different bungs. One black-bunged cap has an "X' vent and the others have the "Y" vent.


^Unit overheated, melted case (not shown), switching FET shorted and the re-sleeved capacitor has evidently seen better days.

-Ben

[goodpsusearch]

Re: the gutless, bloated, and fried power supply hall of shame

welcome back ben!

[momaka]

Re: the gutless, bloated, and fried power supply hall of shame

BEN!!!!!!!!!!!
You're back!
Haven't seen you around here in ages. Always good to see your PSU posts.

This PSU is not that bad, however, it came with a fatal factory defect ...
...
Has screwed-together case (great for repairing!), synchronous rectification, polymer capaci- oh... a popped polymer!
The cap was installed the wrong way around in the factory! (Yes, this PSU was new, worked for a few minutes then popped, according to the guy I got it from. Poly cap pushed out the bung and short-circuited.)

That's a shame.
It really don't look to be too bad of a PSU otherwise.

P.S. Popped poly-cap odor is interesting! It smells like a combination of burnt fireworks and new car tires,... but with a notable sweetness.

Reminds me of when the judges comment on the food they are served in Iron Chef / Chopped TV shows.

Channel Well Technology
KPL-060F
...
Secondary caps: 2x 16V 1000µ SAMXON GK (blown)

Not surprised here.
Samxon aren't really that bad compared to other caps. But flyback converter PSUs really do stress output caps a lot more than regular continuous output designs. So for flyback, it's gotta be Japanese caps if you want the thing to last.

This one is not the worst out there, but it's pretty darn shameful I'd say!
...
A fake Nichicon HE!
https://www.badcaps.net/forum/attach...2&d=1610833288
^Horrible print quality, done in *silver* ink (unlike white on real Nichies) on black sleeve.

Concur.
If it uses counterfeit brand caps, it sure must be shameful. I'd rather see even Chang/Chong/Cheng/X than these... though the difference may not be that big really, and the result is guaranteed to be the same in the end - i.e. blown caps sooner rather than later.

^Does this date code mean anything to you?!
1930, 2030? There aren't even 58 weeks in a year!
Maybe it is a secret code for "BULLSHITE!"

Now that's funny!
Those Nichicon date codes are sooo easy to decipher and understand... yet the copycats failed to do even that. Then again, given the obviously bad fakery, it looks like they didn't try very hard to begin with.

UL: No safety certification markings at all!

Which is interesting, because at a glance, the PCB appears to be well-designed in terms of spacing and isolation distances. But that's where the good ends. The bad: secondary heatsink is extending over to primary side and is not covered/isolated and sitting to close to primary-connected components. Same appears to be going on with the primary heatsink over the secondary side. And to top it all of, the blue EMI/RFI cap between primary (-) bus and secondary ground appears a bit too thin, suggesting it may just be a regular 1 or 2 KV -rated ceramic instead of proper Y2-class. So most likely, this adapter would fail a proper hi-pot test.

^Does this date code mean anything to you?!
1930, 2030? There aren't even 58 weeks in a year!
Maybe it is a secret code for "BULLSHITE!"

"AC/DC ADAPTOR"
"TS5-6W"

Made for your MONITER

^ It was difficult to take a picture of this, but you can faintly see the spot where the case has warped/melted/sunken. In fact, it is melted all along the heatsink for the primary side transistor, and especially so just above the transistor! Ouch!

That's scary. The primary must have been drawing a crazy amount of power to do that. Surprised the primary controller didn't catch that. Then again, if OPP and other protections are not set properly, perhaps that's why.

^ Mains bridge rectifier solder joints cracked, plus splattered solder balls everywhere!

Maybe the guy who set up the wave solder machine read some "Western" guide with temperature instructions in Fahrenheit and then copied the same values on a machine that has settings in Celsius?

Now for the last one ... [properly] hold onto your soldering iron!
https://www.badcaps.net/forum/attach...5&d=1610848475
^ Mains input jumper link. Do you even solder?!

Oh, it's fine! Mains electricity can jump over small gaps like that, so why does it even matter?

When will it end?!... When your business burns down, in fact.

I was told that the fire department was called out for this one, as the employees heard a "strange noise" and smelled something burning!

Amazingly, the fuse is not blown, despite *every* electrolytic cap being blown, and the case seriously warped/melted!
It appears the large capacitor also blew, and sprayed electrolyte, which dried/crystallized on the case and some components.
...
UL: E473260

I guess that goes to show that UL-listed doesn't necessarily mean designed well. In this case, the primary PWM controller really should have sensed something is going on.

In addition, the one very bulged 1000µ cap has imprinted the shape of it's vent on the cover, pushing it out a little - ouch!

Nice!

Thanks for posting these. Was a really fun read for sure.

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

When will it end?!... When your business burns down, in fact.

I was told that the fire department was called out for this one, as the employees heard a "strange noise" and smelled something burning!

Amazingly, the fuse is not blown, despite *every* electrolytic cap being blown, and the case seriously warped/melted!
It appears the large capacitor also blew, and sprayed electrolyte, which dried/crystallized on the case and some components.

"DC ADAPTOR"
"PSU-1250-D4"

100-240V 50/60Hz 1.6A
12V 5A

"Trained Service Personal Only"

Rectifier: KBP310
Controller: GR8875N
Primary FET: SNF8N60
Secondary rectifier: MBRF20100CT
Primary cap: 120µ 400V, "Richcap" (Blown)
Controller cap: 22µ 50V "Richcap" (Bulged)
Secondary caps: 2x 1000µ 25V "Richcap" TM (Both bulged, one slightly)
UL: E473260

It smells charred ... just look at that PCB turning dark brown! It cooked the PCB around the mains bridge rectifier! (Excess RMS current due to bad cap(s)?)
On top of that, the output cable has two conductors, one red and one black; here you can see it has gotten so hot that the red wire has turned dark brown. (The barrel jack was not melted, FYI)








"WARNING:
CENTRO POWER
PS-D1205-D4-R1
2017/07/22"

BTW, keen eyes will note a damaged trimmer pot there... I did that during disassembly. But you can see, the electrolyte caused it to rust under where the plastic adjustment head was.
In addition, the one very bulged 1000µ cap has imprinted the shape of it's vent on the cover, pushing it out a little - ouch!

Also, sorry for the bad pics on this one, the lighting was especially bad here!

-Ben

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

Yet another faulty 12V 5A PSU ... this one has some absolutely filthy solder!

"AC/DC ADAPTOR"
"TS5-6W"

100-240V 50/60Hz 1.8A
12V 5A

Rectifier: 4x 1N5408
Controller: OB2269CP
Primary FET: Toshiba K11A60D
Secondary rectifier: SIRECT MBR20100FCT
Primary cap: 100µ 400V, "LH.Nova" RX
Controller cap: 22µ 50V "LH.Nova"
Secondary caps: 16V 1000µ "LH.Nova" LE (popped), 16V 680µ "LH.Nova" LE
UL: E329462


^ This is how the PSU was from the case. Very warped PCB.


^ It was difficult to take a picture of this, but you can faintly see the spot where the case has warped/melted/sunken. In fact, it is melted all along the heatsink for the primary side transistor, and especially so just above the transistor! Ouch!


^ Popped cap. Disappointing - no crazy Super-Nova here...


^ Mains bridge rectifier solder joints cracked, plus splattered solder balls everywhere!
...baa-a-a-aad! :sheep:


^ Mosfet has turned PCB a different color, and the joints are cracked. More splattered solder balls, too.

Now for the last one ... [properly] hold onto your soldering iron!

^ Mains input jumper link. Do you even solder?!
(It's in the place of an NTC thermistor. It's not a fuse; there is a fuse on the other mains leg.)

-Ben

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

This one is not the worst out there, but it's pretty darn shameful I'd say!
(Will give you a "super-night" when it burns your house down!)

"Supernight"
"AC/DC ADAPTER"
"W-T5000"

Input "VOLTAGEDE": 100-240V 50/60Hz 1.6A
12V 5A

Rectifier: KBP310GL
Controller: UC3843AN
Primary FET: FQPF5N60C (Looks like it's been tossed around...)
Secondary rectifier: STPS10H100CT
Primary cap: 68µ 450V, "Nichicon HE" (FAKE - See pics!)
Controller cap: 47µ 50V "KSC"
Secondary caps: 2x 25V 1000µ "KSC"
UL: No safety certification markings at all!

Note that the LED in this unit is the unusual oval type...!


"CAUTION:
INDOOR USE ONLY.
FOR USE WITHIT EQUIPMENT ONLY."
Did they mean "without" equipment?




^ Burnt SMD resistors.

Here is the interesting bit:
A fake Nichicon HE!

^Horrible print quality, done in *silver* ink (unlike white on real Nichies) on black sleeve.


^Does this date code mean anything to you?!
1930, 2030? There aren't even 58 weeks in a year!
Maybe it is a secret code for "BULLSHITE!"


^ Wrong vent stamp, and poorly shrunken sleeve.


^ Wrong bung, and the leads appear to have been messed with.
Perhaps this was removed, re-sleeved, re-leaded (lead extensions welded on), then sold as "new/genuine"?
It measures about 63uF, for what it's worth!

Also, BTW... I opened this PSU with my bare hands.
Yes, my bare hands. No tools, no dropping nor throwing. Just a squeeze.
It was THAT easy to crack the plastic! (It mostly broke along the seam, but some of the actual shell broke along with it!)

-Ben

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

Channel Well Technology
KPL-060F

100-240V 50/60Hz 1.7A
12V 5A

Rectifier: ??? (Hidden, didn't bother to unsolder heatsink, but looks like same package as KBL40x types)
Controller: "MWP 36R 55"
Primary FET: 10NK60Z
Secondary rectifier: SBR20U60CT
Primary cap: 120µ 400V, Su'scon
Controller caps: 22µ 35V and 4.7µ 50V, Su'scon
Secondary caps: 2x 16V 1000µ SAMXON GK (blown), 1x 16V 220µ SAMXON GF (Left it in as it's after the PI coil; dunno about the ESR, but it measured OK capacitance-wise.)
UL: E161451

Was a pain to crack open, unlike the cheapo PSUs. I slightly damaged the case, but after recapping with NCC KYB (what I had on hand), it is good now for field testing/diagnosing purposes. After all, this PSU has a special 4-pin connector, and costs ~$60 new from a reputable source.





-Ben

[ben7]

Re: the gutless, bloated, and fried power supply hall of shame

This PSU is not that bad, however, it came with a fatal factory defect ...

"AC ADAPTER"
"TS5-6W"
"Trained Service Personal Only."

100-240VAC 50/60Hz 0.8A
12V 5A

Rectifier: GBU808P
Controller: CR5842B
Primary FET: AGM12N60F
Sync Rect Controller: 1('I'?)9HMA
Sync Rect FET: 085N10
Primary cap: 82µ 450V, (Brand is "H." in a circle)
Controller cap: 22µ 50V, "BH"
Secondary caps: 16V, 470/560/820µ ( ), "RLE2"
UL: E329462 (UL now requires you to register with them in order to see the listings! Bummer! So, MFG is unknown to me...)

Has screwed-together case (great for repairing!), synchronous rectification, polymer capaci- oh... a popped polymer!
The cap was installed the wrong way around in the factory! (Yes, this PSU was new, worked for a few minutes then popped, according to the guy I got it from. Poly cap pushed out the bung and short-circuited.)




^ You can see that the capacitor was installed backwards, based on the silkscreening (and PCB traces - not shown here).

P.S. Popped poly-cap odor is interesting! It smells like a combination of burnt fireworks and new car tires,... but with a notable sweetness.

-Ben

[pc7fan]

Re: the gutless, bloated, and fried power supply hall of shame

I think the second "+5V" column is for the -5V output. IMO, the chances the -12V and -5V outputs are capable of 1A are pretty low if they are regulated by a 78xx 3-T regulator.

The Semko and Nemko marks are pretty badly done, .

I'd guess the realistic output power at 250W-300W. Do NCC KMH series large-can lytics have black sleeves? Or the usual NCC brown?

Always brown. Same with KMM, KMQ and KMG.

[momaka]

Re: the gutless, bloated, and fried power supply hall of shame

Best part: Damn thing can't even do 100mA without the voltage tanking. Output sagged down to 3V at just 125mA.

Dang, what a piece of garbage!

I also have never quite seen a design like this before.
It's flyback, of course. And it appears to have feedback, as evident from the optocoupler. But generally designs like that use 2 transistors: a bigger one for driving the transformer and a smaller one for driving the bigger transistor. This one has only ONE transistor doing everything on the primary side. I initially even confused it with one of those single-transistor oscillator designs with no feedback. Then again, it appears there is no difference between this one and the one-transistor oscillator designs when it comes to regulation... or lack thereof, I mean.

If the output caps are good and that thing is still only outputting only 3V with ~100 mA load, then it surely belong to the scrap pile. I mean, with such poor loading, you'll be lucky if you can power more than a few LEDs with this thing. Heck, if you remove the case LEDs, you can probably almost double the adapter's output capacity.