Dell PS-5251-2DF2 missing -12V line

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

Standby (5VSB) IC part number: TNY267pn

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

HI benyosua, welcome to the forums!

I have the same PSU, I think. Will try to dig it up and open it tomorrow, then let you know. Hope the delay in response wasn't an issue.

[benyosua]

Re: Dell PS-5251-2DF2 missing -12V line

hello! Good afternoon. Thank you for allowing me to enter this wonderful forum. Can some body tell me the part number of IC which control the stand by transformer in a dell ps-5251-2ds atx power supply, please? I need to repair it, but this IC is complete damaged, I can not see the reference part number about it in order to by it. Thanks a lot.

[pfrcom]

Re: Dell PS-5251-2DF2 missing -12V line

Over the past few years, I have fixed at least three of these Dell power supplies. In every case the A916 transistor was either shorted or leaky.

Is that located at Q251, next to (under) IC511 ?

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

Thanks for the insights, momaka. Now you've got me very interested in seeing a few computer psu schematics so I can see how these coils are connected.

If you look for JNC ATX on Google Images, you will find quite a few.

Here's some that I found (Note: All of these use the older half-bridge design and cheap group regulation. Particularly, because of the group regulation, you will see that the toroid we talked about is represented as coupled coils on the schematic.):

http://radiom.newmail.ru/atx/JNC_Y-B200-ATX.gif
These first two are pretty similar in terms of how the 5vsb is derived.
The next one is an even older design, with a 7805 regulator for the 5vsb.
http://radiom.newmail.ru/atx/JNC%20200-ATX.gif

I think the best way to understand how one of these works is to grab a cheap junk PSU, take it apart, and follow the traces with your hand.

[Longbow]

Re: Dell PS-5251-2DF2 missing -12V line

Thanks for the insights, momaka. Now you've got me very interested in seeing a few computer psu schematics so I can see how these coils are connected.

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

Apparently, current draw from the various output lines in a computer remains relatively constant, and thus the cross-coupling can be effective.

Quite the opposite, actually. The current draw from a computer varies quite a bit, especially on whichever rail the CPU is using for power. Back in the day it was the 5v rail (for AMD Athlon XP CPUs and Pentium 3) and now it's almost entirely the 12v rail. It's not unncommon at all to get output demand swings of 50W and more just from the CPU. With more power-hungry CPU and graphics card, you could get a lot more than that.

This type of transformer reminds me of trifilar wound RF coupling devices used for broadbanding in transmitter circuits. The science behind it escapes me.

See post #40 by PCBONEZ.

I might add that knowing what this transformer does, and that the windings are most likely in series with the output lines of the supply that it is even less likely to fail than a switching transformer. One reason I say this is because the voltage drop across any particular winding is nearly zero, and the voltage between any 2 adjacent windings is no more than 24 volts maximum. Of course, something next to this part could melt down and physically damage or burn the windings. All sorts of possibilities.

Actually, this inductor/toroid is a lot more likely to fail than the main transformer. Why? - I don't know, but I've seen many more burned toroids than transformers. In fact, I've never seen a main transformer go bad in a computer PSU, even after catastrophic failures.

I will confirm the schematic of the circuit, because it seems to have less components than that from everell.

Good idea. If you can draw out a diagram of the circuit, even better!

One interesting thing : without the -12V, the fan doesn´t turn.

Sounds like the CPU fan is ran from both the 12v and -12v rails. If I remember correctly, my Macron MPT-301 had a similar circuit like that - it was using 12v and -5v rails for the fan.

[Longbow]

Re: Dell PS-5251-2DF2 missing -12V line

Not used as a transformer.
It's coupled inductors. The coupling allows the field in each rail to oppose current change in all the rails.

Thanks. This type of filtering isn't found in home entertainment equipment (my area of experience). Apparently, current draw from the various output lines in a computer remains relatively constant, and thus the cross-coupling can be effective. Whereas in an audio amp or a TV set, current demands are constantly changing. This type of transformer reminds me of trifilar wound RF coupling devices used for broadbanding in transmitter circuits. The science behind it escapes me.

I might add that knowing what this transformer does, and that the windings are most likely in series with the output lines of the supply that it is even less likely to fail than a switching transformer. One reason I say this is because the voltage drop across any particular winding is nearly zero, and the voltage between any 2 adjacent windings is no more than 24 volts maximum. Of course, something next to this part could melt down and physically damage or burn the windings. All sorts of possibilities.

[rogfanther]

Re: Dell PS-5251-2DF2 missing -12V line

Hi,
I will confirm the schematic of the circuit, because it seems to have less components than that from everell. At least, it seems not to have the 2222 transistors... Will put the capacitors back today. One interesting thing : without the -12V, the fan doesn´t turn.

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

I doubt the caps would change anything.

It may also be possible that when the -12v coil overheated, it ruined the toroid. As far as I know, those toroids do change properties when they are overheated.
The only other thing I can think of is there is still a bad component somewhere.

If bad comes to worse, you could always try my earlier idea and simplify the whole circuit by disabling all of the complex circuits I mentioned. But that's probably best left as a last resort/fun experiment.

[rogfanther]

Re: Dell PS-5251-2DF2 missing -12V line

No glue involved, just that little bit in the supervisory board, that has already been removed. And the toroid was bad, because there was not continuity between the -12 coil pins, and there was some continuity between one of these pins and the other coils.

I feel it is strange that the 3.3V was altered after testing it without its filtering circuit. Going to remove and replace the caps around that again

[PCBONEZ]

Re: Dell PS-5251-2DF2 missing -12V line

Not used as a transformer.
It's coupled inductors.

Inductors oppose current change.
The coupling allows the field in each rail to oppose current change in all the rails.
[Took me a while to understand that. Folks here taught me...]
.

[Longbow]

Re: Dell PS-5251-2DF2 missing -12V line

No.
This is the main filtering toroid/inductor for the 5v, 12v, and -12v rails. Depeding on the PSU design (and age), some manufacturers will have a winding even for the -5v rail.

Thanks. I'd like to see a schematic of a toroid filter transformer used in such a circuit. I can't think of why the individual output lines from a switcher would need to be coupled to each other through a transformer, but admittedly there's always more to learn. Perhaps we are dealing with a class D supply? I looked at the data sheet for TSM111 supervisory chip, which can be seen here:

http://pdf1.alldatasheet.com/datashe...CS/TSM111.html

In the photo the transformer appears to be very small, considering the current requirements of a typical computer PSU. In any case, shorted windings would obviously have to be addressed first. The fact that symptoms are changing every time something is removed and replaced suggests to me that there is burned glue pc board contamination and not a bad transformer.

[Longbow]

Re: Dell PS-5251-2DF2 missing -12V line

I'm sure there must be a sticky on this subject, but the photos of your supply show one important thing that must be fixed. For years, manufacturers would use yellow hot glue to make sure large parts didn't work loose, or that tack-ons would not short out something. After a while, everyone discovered that the yellow glue turns brown with age and heat. After it is brown (maybe even before) you have a conductor wherever the brown glue happens to touch. It must be chipped off carefully with a pick. Sometimes freezing the glue will make it brittle and easier to chip off. Note: Brown glue: Great conductor. Get all of it off the board before you waste any more time with it. I wouldn't be surprised if the glue were actually the cause of the failures.

[rogfanther]

Re: Dell PS-5251-2DF2 missing -12V line

I´ve got some B772s. Rated for 3A. The maximum voltage is 40V, but I believe in the superv. board it would not be over 12V, isn´t it ?
As I was in it, changed the fan also.
The resistors measure ok, but I will change them also.
3.3V continues low ( 0.9 - 1.2 V ) , and I´ve observed that the 12V seems to creep up slowly. Starting at 12.23, then going up. Last I looked, it was at 12.67. Then I shut it down to rethink my strategies.

Soldered back the original caps from the 3.3V line, without noticeable effect. Seems I need to rethink my approach of this problem, or at least understand better from does the 3.3V voltage comes and why it is influenced from the -12V coil in the toroid being connected.

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

Those I had around here. I thought as this is some kind of a control circuit, that it would not be going to all that 800mA

Yes, I was thinking the same thing, which is why I thought the MJE350 with a heatsink might work fine too.

I could not find a datasheet for the 1023

Try looking for KTA1023 on google
Here's what I found:
http://www.alldatasheet.com/datashee...C/KTA1023.html

[rogfanther]

Re: Dell PS-5251-2DF2 missing -12V line

Those I had around here. I thought as this is some kind of a control circuit, that it would not be going to all that 800mA, but thinking about the burnt glue I must revise my estimatives. But if the MJEs are good just for testing, I can source the correct parts without hurry.

I could not find a datasheet for the 1023, but will try to find something rated for maybe 1A with 100V...

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

So maybe there were some problem with regulation / supervisory board that was somehow being avoided because the -12V wasn´t running.

Yes, it's possible.
That weird circuit that connects the 3.3v rail, 12v rail, and -12v rail goes to ADJ, so I woudln't be surprised if that was the case.

Any comment about the substitution of A1023 with the MJE350 ? Searching for a data sheet for the A1023 still...

I'm not that good with transistor replacements, but from what I can see from the data sheets, both A916 and A1023 are PNP transistors rated for -800mA, whereas the MJE350 is rated for -500mA.
Based on that, I don't think you can replace the A916/A1023 with the MJE350. But I could be wrong since I don't have that much experience with transistor replacements.
Of course if you really have nothing else available, it might be worth a try. The MJE350 has a mounting hole so maybe you can attach a heatsink to it (with thermal compound, of course) to make up for the greater heat dissipation due to the lower current rating.

[rogfanther]

Re: Dell PS-5251-2DF2 missing -12V line

The coil is already back. I don´t blame him 100%, because I did not measure voltages again between putting it back and dessoldering the toroid for measuring its continuity. So maybe there were some problem with regulation / supervisory board that was somehow being avoided because the -12V wasn´t running.

Will change both transistors, the 431 and both resistors ( although they measure right ) just in case .

Stupid me that I forgot to take some resistance measurings in the blue before cleaning it out... It would be nice to document it with numbers and pictures, just for some unbeliever.... After this recent learning about power supplies, and the books I read, I´ve learned about this phenomenon of old glue turning conductive. I confess I had never thought of that before.

The transistor around the black part seems to be connected with the termistor / fan, so maybe it overheated because of the fan getting stuck ( too much dust in it ) .

Any comment about the substitution of A1023 with the MJE350 ? Searching for a datasheet for the A1023 still...

Thanks.

[momaka]

Re: Dell PS-5251-2DF2 missing -12V line

One part of information : it seems that while I was thinking about the toroid, with too many people in the shop offering their "advice" [ to be honest, it was the boss ] , I tried the psu without that second coil to the left, that belongs to the 3.3V circuit. Could this have damaged something ?

Damaged something - unlikely.
But that coil is critical to the operation and regulation of the 3.3v rail. Without it, you would likely get massive waves of ripple on the output and the voltage might be unstable too.

One bit of information : without the -12V coil connected , voltages of 5 and 12V are in spec ( 5.12 and 12.2-12.4 ) , but the 3.3V is strange ( 2.7V ) .

With the complete toroid coils, voltages go out of spec ( 4.7V, 12.7V, 0.94 to 1.7V , -10.6V ) .

Interesting.
I'd say change that transistor and all other parts everell mentioned, then test it again.

Kaboom.......where are you when we need you. Tell this man about glue on power supplies!!!