Power supply build quality pictorial. part 2

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

It's on the 5VSB though.

Here's another Topower, that actually says Topower on it. It was locking up the system, and the top fan was grinding very loudly. 23,958 running hours on it powering an X-ray machine at a vet clinic, the system was Core 2 Duo with mostly low power components and the special X-ray PCI cards. All the caps looked good but I don't trust those Jun Fu. Turns out the caps on the 5VSB, 3.3V, and 5V failed without bulging. Can't really blame the one on the 3.3V after the coil, it was touching the 15Ω dummy resistor for the 3.3V (And you can see it changed the color of the glue). The one on the 12V was 215Ω, none on the 5V. All the caps are Jun Fool by the way. Hey at least it uses a Viper22A for 5VSB. Soldering looks pretty good for a Topower. Both Globe Fans were really dry.

Do I see right?? Does the label state 450W peak?? For how many ms?

350W max with those transistors.. And how about the primary caps? Are they 470uF? Then it's 300W max 250W continuous..

[Pentium4]

Re: Power supply build quality pictorial. part 2

Yeah those switchers are only 10A, so 350W would be pushing it. The Jun Fu primaries are 470uF, and they read 474uF and 476uF. I think I'm going to recap it, I'll throw in some lame 560uF Teapo I have, and replace the minimum load resistor on the 3.3V with the 215Ω resistor and leave the 12V with no dummy resistor.

[Pentium4]

Sunus Suntek 350ATXE-HP

Sunus Suntek hiding a Hipro inside This one looks similar to some of the recent Hipro's posted but it's definitely different. First of all, it has a rocker switch. Never personally seen one on a Hipro before. It's double forward, using two Toshiba 2SK2749 10A FETs. Nice input filter with MOV's. 8A bridge. Not sure what the primary controllers are. So, if I'm incorrect on in any of this, correct me because I definitely could be wrong. It looks like the 5V is rectified by four 30A schottky rectifiers (STPS3045CW). The 12V uses two 20A schottky rectifiers, and according to the datasheet, they can do 20A @ 125C! They're Fairchild FYP2010DN. As far as I'm aware, the 3.3V is generated from the 12V rail via DC-DC conversion and is output using synchronous rectification. The two FETs there are an STP40NF03L and an SSM40N03P. The 3300uF 10V Teapo SC before the coil on the 3.3V rail has failed. The soldering looks good, but definitely not Hipro's best. At least it's lead. The death ADDA was about as dry as a desert in Arizona, but it was still spinning freely, probably because the rubber seal was on there very, very tight. Also, the second lead on the fan temperature thermistor was never soldered to the board:

[Wester547]

Re: Power supply build quality pictorial. part 2

That my friend is the HP-P4017F5/RS-380-PMSR, a 350W rated Hipro, a 350W version of the 300W Hipro you posted earlier this year. +5V has two STPS3045CWs and the +3.3V rail another two. The +3.3V rail is generated through its own transformer tap and own rectifier, and hence its own winding in the toroid from which it's finally regulated down to +3.3V, hence the fact that it has its own set of filtering capacitors before and after the regulators, which are driven by the +12V rectifiers. How many power on hours were on this unit? 20,000 ~ 30,000? I take it the bulk storage is 560uF. Also looks like that unit has conductive glue. Can't help but notice that the underside of the PCB is silkscreened for a 250W version of this unit. I think the Teapo failed because it's right next to the heatsink and the heatsink probably gets warm if not hot with a heavy load on it. And that's not double forward, that's single switch forward with two switchers in parallel.

[lti]

Re: Power supply build quality pictorial. part 2

The soldering quality is horrible. There are lots of bad solder joints and long component leads that are almost touching other leads.

Hipro always screws up the fan controller in some way. Both of the Hipro PSUs I've opened didn't have the thermistor touching the heatsink. There were some Corsair PSUs (which I think were made by Chicony/Hipro) that would shut off because the fan wasn't running fast enough, and none of the components were actually bolted to the heatsinks. They just pushed the components against the heatsinks and hoped the thermal pads would hold them in place.

[Pentium4]

Re: Power supply build quality pictorial. part 2

That my friend is the HP-P4017F5/RS-380-PMSR, a 350W rated Hipro, a 350W version of the 300W Hipro you posted earlier this year. +5V has two STPS3045CWs and the +3.3V rail another two. The +3.3V rail is generated through its own transformer tap and own rectifier, and hence its own winding in the toroid from which it's finally regulated down to +3.3V, hence the fact that it has its own set of filtering capacitors before and after the regulators, which are driven by the +12V rectifiers. How many power on hours were on this unit? 20,000 ~ 30,000? I take it the bulk storage is 560uF. Also looks like that unit has conductive glue. Can't help but notice that the underside of the PCB is silkscreened for a 250W version of this unit. I think the Teapo failed because it's right next to the heatsink and the heatsink probably gets warm if not hot with a heavy load on it.

Cool, thanks for clarifying but I'm still confused. Why does it need two FETs on the secondary? Unfortunately I'm not sure how many power on hours this thing had. Yeah the bulk caps are 560μF. They read 510μF and 514μF. I think the Teapo failed because they're Teapo and they're old

And that's not double forward, that's single switch forward with two switchers in parallel.

How effective is this compared to double forward?

The soldering quality is horrible. There are lots of bad solder joints and long component leads that are almost touching other leads.

Hipro always screws up the fan controller in some way. Both of the Hipro PSUs I've opened didn't have the thermistor touching the heatsink. There were some Corsair PSUs (which I think were made by Chicony/Hipro) that would shut off because the fan wasn't running fast enough, and none of the components were actually bolted to the heatsinks. They just pushed the components against the heatsinks and hoped the thermal pads would hold them in place.

I wouldn't say that the soldering is horrible but it's pretty disappointing for Hipro. The joints on the toroid and 5VSB transformer definitely need to be cleaned and redone. I'm surprised to hear such poor build quality from Hipro, in my experience they've been very good.

[Wester547]

Re: Power supply build quality pictorial. part 2

Cool, thanks for clarifying but I'm still confused. Why does it need two FETs on the secondary? Unfortunately I'm not sure how many power on hours this thing had. Yeah the bulk caps are 560μF. They read 510μF and 514μF. I think the Teapo failed because they're Teapo and they're old

I'd just be interested in how long they lasted. It doesn't need two FETs on the secondary but using two FETs in parallel (or one on the high side and one on the low side) to linear regulate that rail is certainly more efficient than only using one.

How effective is this compared to double forward?

About the same, maybe.

[Pentium4]

Re: Power supply build quality pictorial. part 2

It doesn't need two FETs on the secondary but using two FETs in parallel (or one on the high side and one on the low side) to linear regulate that rail is certainly more efficient than only using one.

Makes sense. It must work differently than with schottky diode packs, since the specs on the two FETs are different.

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

I love that secondary heatsink!

Just like it's not bad enough that 99% of the time the fan controllers are configured to run the psu too hot and quiet, now they even forget to solder the thermistor!!

[c_hegge]

Re: Power supply build quality pictorial. part 2

I'm load testing the D2567F3P PSU at 250W again now after the recap as I type this. The ripple is about the same on most of the rails, but on the 5vsb rail, it's now down to 12.6mV - half what it was before the recap. The fan is also very noticeable at this load and is pumping out a decent amount of air.

EDIT: I re-lubed the fan too. The bearings were dry.

[Wester547]

Re: Power supply build quality pictorial. part 2

How was the voltage regulation on +5VSB before the recap, because of that Teapo? Was it out of spec then and has it improved since the recap in question?

[c_hegge]

Re: Power supply build quality pictorial. part 2

I didn't measure the voltage before, but it was about 5.05V IIRC with 1A load after. I was more interested in what effect it was having on the ripple suppression.

[goontron]

Re: Power supply build quality pictorial. part 2

100th page marker

[Dan81]

Re: Power supply build quality pictorial. part 2

100th page marker

Also your 1993th post

[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.

Update:
Model number is 225X-34.
SB540 x4 in parallel for +5V 15A rail.
UF5402 parallel pairs for each of the +12V 2.5A, +24V 3.5A and -12V 2A rails.

[Pentium4]

Re: Power supply build quality pictorial. part 2

Recapped the Linkworld. Also replaced the failing Globe Fan with that silly-named chinese one. Much more silent now. Unit looks pretty good. All the CS caps were in spec, even the one on the 12V! Canicon on the 5VSB failed, Fuhjyyu on the 5V and 3.3V failed. Fuhjyyu on the -5V and -12V were fine.

12V rail oscillates on a Haswell system, but is fine on a Core 2 Duo system. It does make a high pitch whine which brings me to the question, how can you tell if a whine is coil whine or from oscillation? I think it may just be coil whine because the pitch does not change at all at any load.

[Pentium4]

Re: Power supply build quality pictorial. part 2

I've isolated the whine to the 5VSB. I tried using caps with lower ESR and it still does the same thing. Went from nichicon PW 1000uF 10V to Panny FM 1000uF 16V. I wish I could remember if it was whining with the failed Canicon caps.

[Behemot]

Re: Power supply build quality pictorial. part 2

Maybe something on the feedback is bad…TL431, optocoupler? What about the small caps, have changed them as well, for the same type?

[goodpsusearch]

Re: Power supply build quality pictorial. part 2

What about the small caps? You should use entry level low esr caps there

[Behemot]

Re: Power supply build quality pictorial. part 2

Depends, I suggest using same type it was there before unless you are certain you can use low-ESR. Sometimes it doesn't like some drastic change and starts oscillating.