Power supply build quality pictorial. part 2

[RJARRRPCGP]

Re: Power supply build quality pictorial. part 2

Here's another Dell NPS-250KB (Newton Power) A 250W max output, version 02, manufactured on the 19th week of 2002.

Back in 2002, I would have wanted that for my T-Bird!

[Pentium4]

Re: Power supply build quality pictorial. part 2

^
I would argue that the efficiency is probably the least important aspect of a PSU's performance. It's the performance on the output side that really matters, and it does very well there (as I would expect form such an overbuilt unit).

I agree with this. If a PSU has Japanese caps, I don't really care how hot the PSU runs. Lower efficiency doesn't make a huge difference unless you have a folding farm, insane amount of servers, etc. That's when you want to have some PFC to lower your bill

[ben7]

Re: Power supply build quality pictorial. part 2

I agree with this. If a PSU has Japanese caps, I don't really care how hot the PSU runs. Lower efficiency doesn't make a huge difference unless you have a folding farm, insane amount of servers, etc. That's when you want to have some PFC to lower your bill

Except, I prefer higher effeciency, because even with japanese caps, it WILL shorten their lifespan. So in the end, after a while the performance of the PSU will drop because of the increased thermal stress on the components.

Plus, that extra energy really adds up on the electric bill!

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

Well

125/0,7 = 178,5W, 53W Heat wasted in the PSU
125/0,85= 147W, 22W Heat wasted in the PSU
125/0,92= 135W, about 11W wasted

And with 250W it's double that amount. So it's worth looking for a more efficient PSU

[ben7]

Re: Power supply build quality pictorial. part 2

Well

125/0,7 = 178,5W, 53W Heat wasted in the PSU
125/0,85= 147W, 22W Heat wasted in the PSU
125/0,92= 135W, about 11W wasted

And with 250W it's double that amount. So it's worth looking for a more efficient PSU

Totally true.
And, for some people, it can increase the temp of their room significantly... like mine... either that or I have too much hot air xD

[Pentium4]

Re: Power supply build quality pictorial. part 2

Yeah, computers are getting very efficient nowadays though, and even high end 80 Plus units can go under 80% with low loads. Even the average decently made half bridge unit gets at least 75% efficiency

[Behemot]

Re: Power supply build quality pictorial. part 2

Well, with no load the PSU has negative efficiency - minus infinity actually No load and some 10 W power draw

[Pentium4]

Re: Power supply build quality pictorial. part 2

Let me tell you, I never get tired of these Macron PSU's. I've been seeing a lot of them lately, probably because all the caps are failing in them. As you can see they didn't quite cut that last zip tie Caps on 5VSB were out of spec and high ESR. Other caps were good on ESR, high on capacitance, except the caps on the 12V, strangely.

Well, with no load the PSU has negative efficiency - minus infinity actually No load and some 10 W power draw

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

Was it wet or where did that corrosion/rust come from?

[Pentium4]

Re: Power supply build quality pictorial. part 2

It was used in a computer close to the ocean, so that could be it

[c_hegge]

Re: Power supply build quality pictorial. part 2

If you're talking about the stuff all over the PCB, I'm pretty sure it's just flux. If the corrosion were really that bad on the bottom of the PCB, then the casing would be falling to pieces.

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

Naa, it looks like corrosion. Flux should look different.
And take a look at the cable outlet. It's almost through...

How near to the sea does this computer stand?

[japlytic]

Re: Power supply build quality pictorial. part 2

Power Systems Inc (from a Tektronix TFP2A FiberMaster)

820uF 200V Nichicon 105C primary unit pair.
2200uF 35V Nichicon PF on all rails; +5V rail has 2200uF 35V Nichicon PF before and after output inductor.
820uF 16V Nichicon PF on primay side control B+ rail.
MJW16010A primary switcher with BUK454-800B synchronous clamping.
KBU6J primary bridge rectifier.

---

Mascot 9260 24V-12V Converter

47uF 63V Chemi-Con bipolar unit before the fuse and first input inductor with 470uF 63V Elna RSG after the second input inductor with 470uF 63V Elna RSG after the third.
3300uF 16V Philips LD before and after 12V 12A output inductor.
470uF 16V Chemi-Con SME on regulated control B+ supply.
IRF540 switching regulator pair.
BYV72EW-200 output rectifier.

My unit had a leaky Zener diode on the 12V output (between the +ve and -ve output terminals) causing an overcurrent condition.

[PeteS in CA]

Re: Power supply build quality pictorial. part 2

P4, that Macron looks like a credible design dragged down by crappy caps (and maybe a cheap fan). Am I missing something? Looks like a nice power line filter, nice heatsinks (including on the I/P bridge rectifier), -52 material O/P inductors, large main transformer, active PFC. It looks like it really could do 400W or very close to it (unlike some 450W-550W P/Ss that would probably put out their rated power for a few seconds before launching).

[Pentium4]

Re: Power supply build quality pictorial. part 2

P4, that Macron looks like a credible design dragged down by crappy caps (and maybe a cheap fan). Am I missing something? Looks like a nice power line filter, nice heatsinks (including on the I/P bridge rectifier), -52 material O/P inductors, large main transformer, active PFC. It looks like it really could do 400W or very close to it (unlike some 450W-550W P/Ss that would probably put out their rated power for a few seconds before launching).

Yeah, it's a great design! Very reliable with some good caps. I've already recapped it with all Japanese caps except I left the big 270uF 400V G-Luxon cap. It read 259uF 0.27Ω ESR so I assumed it was still pretty healthy. The Power Logic fan is definitely high quality. Their sleeve bearing fans have been extremely reliable in my experience, so BB is even better If you look on the transformer it says "MPT500" so it looks like with the right parts, it could do 500W

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

Something to barf about:

Reason: Too little space, to many components...
That Layout should go through another round of layouting...

[ben7]

Re: Power supply build quality pictorial. part 2

Something to barf about:
Reason: Too little space, to many components...
That Layout should go through another round of layouting...

Double barf with the CapXon and Teapo caps there....!

[Stefan Payne]

Re: Power supply build quality pictorial. part 2

Double barf with the CapXon and Teapo caps there....!

Interestingly the CapXons are polymers so they should be fine. The Teapo on top of that is what worries me a little...

[Agent24]

Re: Power supply build quality pictorial. part 2

Interestingly the CapXons are polymers so they should be fine.

CapXon makes rubbish as it is. What makes you think their polymers would be any good? Because polymers are a better design?

Never underestimate how some Chinese manufacturers seem to have a special ability when it comes to taking something and making a bad quality version using whatever substandard materials and corner-cutting tricks they can get away with.

[_chiller_]

Re: Power supply build quality pictorial. part 2

Hey guys, I looked inside this power supply. His name is ATX780HM and it has 780 Watts Oo

Well, i think it has 350 Watts in reality, what do you say?

Some other information of the power supply:
OEM is HKC

Primary caps: two Caps named by the company "Cheng", I´ve never heard of them.

Secondary caps: most of them are from "ChengX", a few of them are from a Company named "FCon" and three caps are from "BH"

Protection-IC: Weltrend WT7520, which has OVP and UVP at 3,3V an 5V. It also has this on 12V, but I dont think that it does work. The paper to the technical datas says that OVP stops the PSU at 5V, wtf? Oo

Oh and it has passive PFC ^^

The company that sells this PSU says, that it has four rails at 12V. In reality, the circuit board says it has only two rails Oo On the bottom, I´ve seen that they soldered all rails to only one. This is a Single-Rail PSU ^^

This is my second post in this forum, I hope the picture works ^^