Power supply build quality pictorial. part 2

[ncovert]

Re: Power supply build quality pictorial. part 2

It's branded as an FSP.

[Behemot]

Re: Power supply build quality pictorial. part 2

FSP Group Inc.
Fortron/Source

or something else?

[ncovert]

Re: Power supply build quality pictorial. part 2

FSP Group Inc. is the brand on the label.

[Behemot]

Re: Power supply build quality pictorial. part 2

If there is no Fortron/Source and it is obviously OEM unit, than it does not come from FSP directly, somebody ordered it this way. Now I agree that this is also FSPs problem, they don't have to make crap even for somebody else, but the customer also is to blame, he wanted it to be this way…

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

Powertech ATX-450W

Those things are still available here for 18euros!

As you can tell from the label, it is an ancient design, come on, 16A on 12V?? But even more strange is the fact that 5vsb is rated 1.5A. 2.0 ampere on 5vsb became a standard more than 12 years ago! How is this thing sold anyway?

The fan stopped moving and the toroid coil burnt, then desoldered itself from pcb!

This is one of the worst power supplies I have seen:

-Inadequate input filter with fake Y caps

-4 diode treatment with tiny diodes.

-No PFC. But they placed a connector for passive pfc in the wrong spot. Notice that there is no connector on the "PFC" area of PCB but there is a connector probably for PPFC on a place after the 4 diodes! The most unexplainable, the 2 leads of the connector are soldered on the same PCB trace...

-The primary caps are only 330uF but at least their real capacity is 340 and 350uF respectively.

-ERL 33 transformer and 13007 switching transistors

-16A @ 12V, 30A @ 3.3V and 30A @ 5V

-No pi coils, except 5vsb. 3.3V and 5V output filter has only 1x 2200uF cap, 12V has 2 caps (1000uF and 470uF) and 5vsb has 2x470uF and a coil.

-No mag amp coil for 3.3V! The main transformer has an extra output for 3.3V The burnt toroid coil regulates both 5V, 12V and 3.3V! Have you seen something like that before? The small toroid coil is jumpered from PCB, just like the PPFC connector! Can anyone explain why?

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

Powertech ATX-450W datasheets:

[mariushm]

Re: Power supply build quality pictorial. part 2

There is a place for PFC on the right of the heatsink, where that varistor is located. It says PFC on pcb.
The other connector is for 110/220v selection. I think it's open-220v, shorted-110v, like in this image: http://www.seekic.com/uploadfile/ic-...7102399533.gif (the E switch)

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

But its leads are shorted permanently.

[mariushm]

Re: Power supply build quality pictorial. part 2

On the PFC connector you're basically supposed to connect a large inductor... that's what passive factor correction means. If you short it, you just skip the inductor, no problem but no pfc.

Instead of just shorting the connector and have a jumper in place, they used a varistor to limit the inrush current going to the rectifier diodes and the capacitors.

[TELVM]

Re: Power supply build quality pictorial. part 2

[B]-No PFC. But they placed a connector for passive pfc in the wrong spot. Notice that there is no connector on the "PFC" area of PCB but there is a connector probably for PPFC on a place after the 4 diodes! The most unexplainable, the 2 leads of the connector are soldered on the same PCB trace...

^ As Mariushm says (although I think he means thermistor when he says varistor).

This similar gutless wonder has the connection for PPFC coil at the same location:


However taking the trouble to solder a PPFC header when there isn't a PPFC coil to plug in is a puzzling touch .

[Behemot]

Re: Power supply build quality pictorial. part 2

Instead of just shorting the connector and have a jumper in place, they used a varistor to limit the inrush current going to the rectifier diodes and the capacitors.

Thermistor, but as you said, it limits inrush currents when you turn it on. It serves no purpose in limiting harmonics.

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

Instead of just shorting the connector and have a jumper in place, they used a varistor to limit the inrush current going to the rectifier diodes and the capacitors.

Yes, I agree about that.

I am refering to the connector they actually put after the 4 diodes:

https://www.badcaps.net/forum/attach...1&d=1388768415

See? It doesn't make any sense. Whatever you put there (PFC coil, jumper, etc) it is permanently shorted and out of the circuit. I think they really messed up with their designs. I can't find another reasonable explanation.

[mariushm]

Re: Power supply build quality pictorial. part 2

Yes, I meant thermistor. Varistors (or MOVs as some call them) are used for other purposes, like protection against excessive transient voltages.

re picture above as you posted it just as I replied: yes, that doesn't make sense.

If that's really for PFC, it only makes me think the board was designed from the start to have a fake passive pfc inductor. At some time, there were lots of power supplies with fake pfc inductors.

But if it's for 110/220v selection, then it's just an issue of connecting that to ground or to neutral, so it's possible only one wire in that connector was used and the switch was connecting the wire to neutral or ground - i forget now exactly how the connection is made and I don't feel like looking it up.

[Behemot]

Re: Power supply build quality pictorial. part 2

It's probably very stripped-down version of already not wonderful platform. It makes no sense to just make new board if you have thousands lying around so they probably just added jumper wire. Most likely there may actually be PFC coils in other versions of this PSU.

[momaka]

Re: Power supply build quality pictorial. part 2

The most unexplainable, the 2 leads of the connector are soldered on the same PCB trace...

...

The small toroid coil is jumpered from PCB, just like the PPFC connector! Can anyone explain why?

Because it's a cheap crap PSU, that's why. I really can't make any sane explanation other than that.
All I can say is, I've seen this before in a extremely cheap L&C/Deer PSU where the traces on the PCB bypassed the PI coils on the outputs. Yet for some reason, L&C must have forgotten about it, so they added PI coils in the PSU. Really funny because those PI coils did exactly nothing. I think I uploaded pictures of this somewhere, but I can't seem to find where.
...
Anyways, those PI coils did find a new home, though. I pulled them out and installed them in another L&C that didn't have any and didn't have those stupid bypassing traces.

-No mag amp coil for 3.3V! The main transformer has an extra output for 3.3V The burnt toroid coil regulates both 5V, 12V and 3.3V! Have you seen something like that before?

Yes, only once. It was in a gutless Cyberlink PSU built by Channel Well Technology (it's based on their ISO line, but extremely stripped down). I don't think I have uploaded pictures of it yet, but I definitely should (in the gutless thread, though). It powered a Pentium 4 PC that drew its power from the 5V rail. How it didn't die, I don't know.

[c_hegge]

Re: Power supply build quality pictorial. part 2

Wait a minute, If I'm tracing things out correctly, it looks like the smallest coil is actually in charge of the 12V output, and the larger burned one just does the 5 and 3.3V rails. That's very weird design, though.

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

Because it's a cheap crap PSU, that's why. I really can't make any sane explanation other than that.
All I can say is, I've seen this before in a extremely cheap L&C/Deer PSU where the traces on the PCB bypassed the PI coils on the outputs. Yet for some reason, L&C must have forgotten about it, so they added PI coils in the PSU. Really funny because those PI coils did exactly nothing. I think I uploaded pictures of this somewhere, but I can't seem to find where.
...
Anyways, those PI coils did find a new home, though. I pulled them out and installed them in another L&C that didn't have any and didn't have those stupid bypassing traces.


Yes, only once. It was in a gutless Cyberlink PSU built by Channel Well Technology (it's based on their ISO line, but extremely stripped down). I don't think I have uploaded pictures of it yet, but I definitely should (in the gutless thread, though). It powered a Pentium 4 PC that drew its power from the 5V rail. How it didn't die, I don't know.

Wait a minute, If I'm tracing things out correctly, it looks like the smallest coil is actually in charge of the 12V output, and the larger burned one just does the 5 and 3.3V rails. That's very weird design, though.

The smallest coil is shorted by PCB.

Here, have some pics that show it better:

[c_hegge]

Re: Power supply build quality pictorial. part 2

Ah, Ok. I thought the small coil was where the diode is, although this means that the 12V had no toroid coil at all! I would hate to think what the ripple would have been like. I actually feel sorry for the system it was powering.

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

All the lines, including 12V use the burnt toroid coil....

[c_hegge]

Re: Power supply build quality pictorial. part 2

Ah, yes, I see that now. Sometimes, it really does help to look at things on a 22" screen and not a tablet.