Power supply build quality pictorial. part 2

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

Yep, Power Integrations is the PWM vendor. Its P/N might be a special-for-Samsung number (if you can sell 100s of 1000s to one customer, a semi-custom part and custom P/N are well worthwhile!).

[Behemot]

Re: Power supply build quality pictorial. part 2

Delta at it's best. How much do you think?

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

Built mid-2000 (D/Cs on PCB and transformer), rated for 300W, active PFC (220uF, 450V, 105C Nichicon GQ I/P cap) ... probably for an HP server. Good caps (UCC LXV & Nichicon PJ?); lots of heavy metal (= heatsinks); as such things go, beautiful PCB and soldering. One curious thing, the AC filter board looks like paper-epoxy while the main board looks like glass epoxy (FR-4) - possibly saving a bit of $$ on the AC filter board?

[Behemot]

Re: Power supply build quality pictorial. part 2

That's about right. But hey, reading the labels is called cheating you know.

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

While at Delta - 1998-2001 - I did some support work at HP, so I might have gotten my hands into prototypes of that model (DPS-300 JP).

[Behemot]

Re: Power supply build quality pictorial. part 2

Oh come on, you're just talking yourself out of that

Anyway, I am curious, is that some kind of DC-DC converter for the +3,3 V rail? It has some transistors down there…

Also it's quite intereseting that while being only ATX 2.0, it uses active PFC…

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

Anyway, I am curious, is that some kind of DC-DC converter for the +3,3 V rail? It has some transistors down there…

Also it's quite intereseting that while being only ATX 2.0, it uses active PFC…

It's possible, or it may be that the design used synchronous rectification (MOSFETs used as rectifiers) to improve efficiency.

Active PFC has been around in commercial use since the mid-late 1980s that I know of; Unitrode's PFC IC, the UC3854 came out ca. 1990. APFC is not exactly new. My guess is that HP waved the right number of $$ bills and quoted a high enough volume estimate, so Delta said, "Whatever you want."

[Behemot]

Re: Power supply build quality pictorial. part 2

It's possible, or it may be that the design used synchronous rectification (MOSFETs used as rectifiers) to improve efficiency.

Don't think so, the rest uses diode rectification, I think it's technically more complicated to use synchronous rectification for one rail and diode for the rest.

ADD// it MAY actually be some kind of synchronout rectifying. Acording to tracing, there goes the yet unrectified +5 V onto those transistors.

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

It's just a guess on my part, and it is more complicated, but if the $$ is right and efficiency is critical, that would be an option. And around that time the load on the +3.3V O/P was taking off. Save 200mV-300mV on the forward drop of the "rectifiers" and multiply that by 20A or 30A, and that's a big efficiency improvement (and less demand on cooling).

[Pentium4]

Re: Power supply build quality pictorial. part 2

Ultra LS600. Gets standby but won't power on. Looks decent, not 600W material though, and it only comes with 1 PCI-E x6 connector

It has an 8A bridge, 820uF HEC primary caps, KSH13009H switchers, 39 transformer, 20A schottky for 3.3V, 30A schottky for 12V, dual 30A schottky for 5V. ALL the caps besides the primary are CS. SG6105DZ controller.

I was looking on the soldering side....and look at that lead in the picture, how sloppy is that. And that's for the fuse leg

I guess it's worth noting that when this was given to me, the voltage selector switch was set at 230V. If they tried running it on 230V, what would be the first thing to check, input caps? I fixed the fuse joint but still doesn't power on

[Behemot]

Re: Power supply build quality pictorial. part 2

At 115 V it most likely won't even turn on with 230V setting as it leaves it with 1:1, that is 115 V or whatever you have there

Just to be sure, check bridge rectifier for shorts and also primary side (switchers, diodes). If all goes well, turn it in and check if you get +5 V SB.

[Pentium4]

Re: Power supply build quality pictorial. part 2

Oh good to know it wouldn't just blow up It gets 5.10V on the 5VSB, so I'll start with checking the primary side

[Behemot]

Re: Power supply build quality pictorial. part 2

Well it there is +5 V SB, problem would more likely be on secondary because if you tried to turn it on with something blown on primary, you'd get fireworks which you haven't as I understand it.

Blown secondary rectifier maybe? Them shottkies do not seem that beefy, maybe somebody tried to pull more than they handled? I've just yesterday repaired some Xilence thing made by Andyson with strange startup behaviour. Got shorted diode on -12 V.

[everell]

Re: Power supply build quality pictorial. part 2

Delta at it's best. How much do you think?

Back to the Delta........The booster coil in the APFC circuit looks like it is made up of TWO coils in series. I have a Delta that uses two booster coils in series - runs HOT. Does your Delta booster coils run hot?

[Pentium4]

Re: Power supply build quality pictorial. part 2

Interesting. I guess I should start there. It just seems strange since there is absolutely no discoloration and everything looks fine. Looks are deceiving though

[Behemot]

Re: Power supply build quality pictorial. part 2

Back to the Delta........The booster coil in the APFC circuit looks like it is made up of TWO coils in series. I have a Delta that uses two booster coils in series - runs HOT. Does your Delta booster coils run hot?

I haven't loaded it actually but there are no traces of anything overheating. Also the fan seems to be spinning more than it could so…

You sure yours are not bad?

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

At 115 V it most likely won't even turn on with 230V setting as it leaves it with 1:1, that is 115 V or whatever you have there

Not necessarily. One time at Apple's "Open Field" EMI test site we did a full scan with the P/S set for 230VAC but with 115VAC applied. The computer worked and was running some sort of program to keep it busy during the test.

There are three smallish lytics between the two heatsinks, closer to the primary side heatsink. One of those, used in starting up the PWM, may be the problem with that supply. Just because it isn't spilling its guts or preggers doesn't mean it's good. I'd suggest using a PW, LXZ or FC series part for replacing it if that's the problem.

[everell]

Re: Power supply build quality pictorial. part 2

I haven't loaded it actually but there are no traces of anything overheating. Also the fan seems to be spinning more than it could so…

You sure yours are not bad?

YES they were bad.....see this thread

https://www.badcaps.net/forum/showthread.php?t=9660

[Behemot]

Re: Power supply build quality pictorial. part 2

Oh yeah, it's the one with the hole, I remember that.

[370forlife]

Re: Power supply build quality pictorial. part 2

Ultra LS600. Gets standby but won't power on. Looks decent, not 600W material though, and it only comes with 1 PCI-E x6 connector

It has an 8A bridge, 820uF HEC primary caps, KSH13009H switchers, 39 transformer, 20A schottky for 3.3V, 30A schottky for 12V, dual 30A schottky for 5V. ALL the caps besides the primary are CS. SG6105DZ controller.

I was looking on the soldering side....and look at that lead in the picture, how sloppy is that. And that's for the fuse leg

I guess it's worth noting that when this was given to me, the voltage selector switch was set at 230V. If they tried running it on 230V, what would be the first thing to check, input caps? I fixed the fuse joint but still doesn't power on

Made by ShenZhen RuiShengYua aka RSY.