Dell GX280 \ LiteOn PS-5161-7DS question

[Agent24]

I've been trying to get a Dell GX280 going, but not having much luck.

Motherboard is full of good Japanese capacitors, all look OK, and it POSTs fine with a separate PSU.


PSU work so far is as follows:

REFDES	Brand	Series	Voltage	Capacitance	Dimension	Notes	Result
							
							
C203	OST	RLS	10v	2200uF	12.5 x 20 mm	FAILED - 4 Ohm ESR	Replaced with Panasonic FR
C206	OST	RLS	6.3v	2200uF	10 x 20 mm		Replaced with Panasonic FR
C283	OST	RLS	10v	2200uF	10 x 25 mm		Replaced with Panasonic FR
C285	OST	RLS	10v	2200uF	10 x 25 mm		Replaced with Panasonic FR
C223	OST	RLS	16v	1200uF	10 x 25 mm		ESR OK, left in for now
							
C503	Ltec	LZG	10v	1000uF	8 x 20 mm		ESR OK, left in for now
C504	Ltec	LZG	10v	1000uF	8 x 20 mm		ESR OK, left in for now
C510	OST	RLS	10v	470uF	8 x 12 mm	FAILED - 2.4 Ohm ESR	Replaced with Rubycon ZL

Machine starts but flashes power light orange with the LiteOn PSU. Sometimes it seems like it gets into a reboot loop, and the light flashes sporadically.

Any known issue with this model of PSU other than bad caps?

[DJduck]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Is this the sff version? I think i have one of those PSUs, but mine has mostly teapo, Nichicon and Rubycon.

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Yeah it's an SFF one.

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

I've found something else odd. The PSU won't start by jumping PS_ON to GND (even with an old HDD connected for load) - Some stupid Dell weirdness going on there or an actual fault?

Looks like something weird is going on, seems like the reboot loop is actually the PSU resetting itself, after being on for a few seconds. I can even hear a little tick when it does, and see a voltage glitch on the output.

So I guess this is some variant of a 'power cycling' problem...

Also, while the output voltages look OK, it doesn't seem like the PSU is generating the Power_Good signal straight away. This probably explains why no board connected to it will POST.

It does however seem to try to output a Power_Good signal after it goes into the ticking\cycling mode.

[retiredcaps]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

I've found something else odd. The PSU won't start by jumping PS_ON to GND (even with an old HDD connected for load) - Some stupid Dell weirdness going on there or an actual fault?

See if you have a similar/same power supply as my thread.

https://www.badcaps.net/forum/showth...highlight=grey

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

See if you have a similar/same power supply as my thread.

https://www.badcaps.net/forum/showth...highlight=grey

Yours was listed as a "5161-7DS" which matches mine, so I assume it's the same one.

But it is strange that worked for you, because according to the official ATX PSU specification:

PWR_OK is a “power good” signal. It should be asserted high by the power supply to
indicate that the +12 VDC, +5VDC, and +3.3VDC outputs are above the under-voltage
thresholds listed in Section 3.2.1 and that sufficient mains energy is stored by the converter
to guarantee continuous power operation within specification for at least the duration
specified in Section 3.2.11, “Voltage Hold-up Time.” Conversely, PWR_OK should be de-
asserted to a low state when any of the +12 VDC, +5 VDC, or +3.3 VDC output voltages
falls below its under-voltage threshold, or when mains power has been removed for a time
sufficiently long such that power supply operation cannot be guaranteed beyond the power-
down warning time.

I assume the Dell PSU does (or should do) the same thing. It seems odd that the motherboard works with a normal ATX supply, if the Dell one is non-standard.

If I understand this properly, even if it was non-standard, why would they totally change the functionality of that signal to effectively the reverse of normal? I am not sure that the loss of PWR_OK signal is not a fault symptom... after all, it starts when connected to a board, just won't let the thing even attempt to start.

And I DO get some voltage showing on the PWR_OK line, on a "normal" ATX board, fluctuating up and down when it's cycling. Very strange.

[retiredcaps]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

I assume the Dell PSU does (or should do) the same thing.

I have a personal hate-on for Dell and as such I rarely bother to research/learn what they are doing. If you weren't a badcap longtime member, I would not have even bothered to respond to a Dell thread.

I stripped all the Dells down for parts and fooled around with the power supplies just to see if they work. I recapped only the obviously bulged caps just for "fun" with used capacitors (Capxon brand ) . I have no use for them as they are non standard form factor and only keep them around for potential parts or learning purposes in the future.

Other than that, I try not to have any Dell products, even free ones, at home.

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

I don't like Dell (or any 'brand name' PC) much either. Far too much proprietary rubbish in their products and it seems like the reason is only "because they can" rather than anything sensible.

This is the first full Dell system I've had (it was free which is why I even bothered to keep it) - At first glance I thought some new capacitors would solve its problems but now looks like it might be more trouble than it's worth...

[retiredcaps]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

If you really want more ideas/suggestions, maybe head over to

http://en.community.dell.com/

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

If you really want more ideas/suggestions, maybe head over to

http://en.community.dell.com/

Thanks, might have a look, at least it may clear up how the PWR_OK signal is supposed to work on this PSU...

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

No luck on Dell forums, but that's no surprise.

Anyone have suggestions for troubleshooting the POWER_GOOD signal?

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Found a schematic.. or rather, half of it. Seems like there's a bit missing.

[momaka]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Yeah, I don't see the logic/PS_ON/PWR_OK circuitry in there.

Since the PC works fine with a regular PSU, I wonder if there is a way too fool the PWR_OK signal on that original PSU with a simple circuit. If it needs to be high when the motherboard is on, then simply connect it with a 1k resistor to the 3.3V or 5V rails. If it needs to be low, then use an inverting op-amp or something similar to invert signal.

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

That might work, but there still must be a fault, and I don't know that it would fix it.

The signal does come up, just not straight away, and when it does come up, the PSU has gone into a power cycling state.

The thing is I don't know if it's the POWER_GOOD signal circuitry that's at fault, and there's nothing wrong with the PSU, or if the PSU is really faulty, and this madness with the signal is because of it.

[momaka]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

That I can't tell you.

Do you have pictures of the PSU by any chance?

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Not at the moment, I'll take some today and post them

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Some eye candy...

For the record:
Main PWM control is a UC3845BN
5v Standby is done with a TNY266PN
Supervisor IC is a Weltrend WT7515

[momaka]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Is the PG/PWR_OK wire directly connected to the WT7515? Looks like it is to me. Check if there are any resistors between in and ground or 5V that and make sure they are alright.

[Agent24]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Looks like the startup capacitors were the problem.

C501	OST	RLS	35v	100uF	8 x 12 mm	22 Ohm ESR - 58uF
C308	OST	RLG	50v	10uF	5 x 11 mm	38 Ohm ESR - 7.3uF
C302	OST	RLG	50v	6.8uF	5 x 11 mm	23 Ohm ESR - 4.9uF

Replaced C501 with a 150uF 35v Panasonic FC and other two with 10uF LTecs from another PSU that I recapped - only 1.5 Ohm each, MUCH better. Fine for testing anyway. Now I'll see how this thing goes, if it's OK, I might do 'proper' replacements.

ALSO: It now starts on the bench by pulling PS_ON to GND, previously it wouldn't even do that.

[momaka]

Re: Dell GX280 \ LiteOn PS-5161-7DS question

Wow, that 10uF cap has nearly 40 Ohms ESR!

The again, these PSUs probably run quite hot. No surprise if even Japanese caps get cooked in there (granted, it will probably take them much longer).

I guess the small caps even in these regular PC PSUs should never be underestimated. Learning something new everyday. Good job on the repair, by the way .