Examination of Golden Field ATX-S628 550W peak power PSU

[Newbie2]

Hi again,

I've recently made a purchase of a new Golden Field ATX-S628 550W peak power PSU that was being sold at a local computer shop I know for $38 CAD. It actually had some weight to it, along with a few other things that made it attractive to me at that price. I'll have it on hand as a backup PSU, in case my Antec PSU fails.

It appears to be a great PSU for the money I spent on it, as it has beefier heatsinks than normal cheap power supplies, has a 120mm fan, nice dark gray finish, and comes with a decent secondary side and a transient filtering stage (two Y caps, two X caps, and a coil). The caps in this PSU are BH brand, which I haven't heard anything bad yet about them (but they do look like cheap Taiwanese caps, IMO).


The complete package.




Three shots of this PSU.


This PSU has one 24-pin ATX connector, two P4 12V connectors (that can be joined together for EPS12V), one PCI-E 6-pin connector, three 4-pin peripheral (molex) connectors, and two SATA power connectors. This arrangement I believe will suit most basic computers.


A shot of the label. Note the "Oiliness Bearing Fan" remark.


Overhead shot of this PSU's internals. Notice that it is not completely gutless.


A shot showing the transient filtering stage, and the two primary caps.


The primary caps are two BH 680uf 200V.


The best picture I could get of the KBL406 4A 800V bridge rectifier. It being 4A, the most this power supply could draw from a 120V AC outlet would be 480W.


A shot of the primary side silicon. One CET CEF02N6A 600V 1.4A n-channel transistor, along with two 13009 NPN power transistors.




Three shots of the secondary side.


A shot of the secondary side silicon. The one on the left is an S20C45C 20A 45V Schottky rectifier, the one in the middle is an S30D40C 30A 40V Schottky rectifier, and the one on the right is a U20C20C 20A 200V power rectifier. I doubt that this PSU has two 12V rails, as on the label, since there is only one power rectifier for the 12V. Correct me if I'm wrong.


The 120mm fan used in this PSU is a Yate Loon D12SL-12 12V sleeve bearing fan, which operates at 1350RPM, pushing out 47CFM of air at a noise level of 28dB.


The fan connects to this additional board in the PSU. I'm not sure what it exactly does, but I'm guessing it's some sort of fan controller, as the package indicates an "Adjustable Fan Speed Controller". If you know what it is, tell me. It has an LM339N on it.


This PSU uses an SG6105 PWM controller chip.

I haven't even powered on this PSU yet. The rating on the label is 550W, but that's the peak power rating. I would guess that this PSU's continuous rating would be at 400W, considering how the PSU looks inside, and the rating of the bridge rectifier.

If you would like to view the webpage on the Golden Field website on this power supply, click on this link: http://www.goldenfield.com.cn/en/ProductView.asp?ID=146

Anyway, feel free to comment about your opinions on this Golden Field ATX-S628 power supply.

[370forlife]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Goldenfield uses two OEM's, one is Huntkey and the other is Sunpro.

This looks like a huntkey.

[c_hegge]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Re-cap it. It's probably not too bad after that.

[370forlife]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

It's probably good for 300w. 13009's are good for about 300w continuous, 350w peak if cooled well. Only 1 20A ultrafast on the 12v, and yes, it's only single rail as it doesn't even have OCP.

[PCBONEZ]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

UL mark has no number => Fake.

[momaka]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

It's probably good for 300w. 13009's are good for about 300w continuous, 350w peak if cooled well. Only 1 20A ultrafast on the 12v, and yes, it's only single rail as it doesn't even have OCP.

What he said ^ .

As far as fan controller - look for a thermal resistors near the secondary side of the PSU - it's usually marked on the PCB as NTC, PTC, TR, or THR. If you see one, there is a thermal fan controller.

Aside from the BH caps, it doesn't look at all that bad. For $38 CAD, though, I would have expected to see Teapo or at least Ost in there. Even LTec would be preferable to BH.

[Scenic]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Goldenfield uses two OEM's, one is Huntkey and the other is Sunpro.

This looks like a huntkey.

i've had a Golden Field PSU that was CWT OEM.. (yes.. including Fuhjyyu)
interesting enough, only a few of them were actually Fuhjyyu.. most of the caps in there were Crapxon (KM and GL).

refused to turn on one day.. so i junked it..
was too crammed (lots of daughterboards + jumperwires across the whole PSU) and shoddily built (thin wiring, underrated components (if not missing)) to be worth fixing..
label wise it was a 420W IIRC. inside looked more like 350W.. at best.. with lots of tailwinds from the dual 80mm fans..

edit: by the way.. "BH" is a common abbreviation for bra in german lol
(büstenhalter -> BH .. could be roughly one-to-one translated as breastholder)

http://www.dict.cc/?l=e&s=BH

[370forlife]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Interesting, on System General's website they have a schematic for a power supply that is 300W that utilizes the SG6105.

It does have OPP though.

[Khron]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

I think i can just see a second TO-220 pack rectifier for the 12v, on the other side of the heatsink in the overhead shot. Quite a common arrangement, from what i've seen.

[Newbie2]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

I think i can just see a second TO-220 pack rectifier for the 12v, on the other side of the heatsink in the overhead shot. Quite a common arrangement, from what i've seen.

Yes, there is actually another U20C20C 20A 200V ultrafast power rectifier on the other side of the heatsink, something I overlooked in my initial examination.




So Golden Field actually wasn't lying about this PSU having two 12V rails.

I do not plan to recap this PSU, as I don't have the time and money to do so right now. The BH caps should be okay, since if I ever need to use this PSU, it'll only be for backup purposes only and wouldn't be used for a long time.

[370forlife]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

So Golden Field actually wasn't lying about this PSU having two 12V rails.

I do not plan to recap this PSU, as I don't have the time and money to do so right now. The BH caps should be okay, since if I ever need to use this PSU, it'll only be for backup purposes only and wouldn't be used for a long time.

Thats good that there is 2 diode packs on the 12v rail, but it is only a single rail unit, as it dosen't even have OCP.

The two diode packs run in parallel to double the capacity. This does not mean it has 2 rails. Usually multiple rail units only have 1 or 2 12v outputs on the transformer and multiple rectifiers running in parallel. This is then split into multiple outputs after the chokes with shunts to monitor the amount of amps running through each rail.

[Newbie2]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Thats good that there is 2 diode packs on the 12v rail, but it is only a single rail unit, as it dosen't even have OCP.

The two diode packs run in parallel to double the capacity. This does not mean it has 2 rails. Usually multiple rail units only have 1 or 2 12v outputs on the transformer and multiple rectifiers running in parallel. This is then split into multiple outputs after the chokes with shunts to monitor the amount of amps running through each rail.

I see, thanks for the correction.

[PCBONEZ]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Thats good that there is 2 diode packs on the 12v rail, but it is only a single rail unit, as it dosen't even have OCP.

The two diode packs run in parallel to double the capacity. This does not mean it has 2 rails. Usually multiple rail units only have 1 or 2 12v outputs on the transformer and multiple rectifiers running in parallel. This is then split into multiple outputs after the chokes with shunts to monitor the amount of amps running through each rail.

Parallel diode packs [and some other IC's] are often used to reduce heat dissipation rather than to increase capacity.
Heat dissipation -to- amps is logarithmic in silicon.
1 @ 10 amps gives off much more heat than the total heat dissipated by 2 at 5 amps each.
Means less heat-sink area is required and a smaller fan can be used.
.

[370forlife]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

In this case it was probably a heat vs capacity issue. They needed to make as least heat as possible while still at least meeting the specs on the label.

Though, I don't know. Ultrafasts generate more heat than schottky's, they could have just gone with 1 to-247 40A schottky and it probably wouldn't have gotten any hotter than 2 20A ultrafasts in parallel. Might be a cost issue there, I don't know how much more a single 40A to-247 schottky costs versus 2 20A to-220 ultrafasts.

[ratdude747]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

I do not plan to recap this PSU, as I don't have the time and money to do so right now. The BH caps should be okay, since if I ever need to use this PSU, it'll only be for backup purposes only and wouldn't be used for a long time.

oh yeah, its also i good idea to use an old rotten tire for a spare... just when you need it, it fails...

BH is a very lame brand, i would never trust 'em... do yourself a favor and recap it when you have the funds...

[Newbie2]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

oh yeah, its also i good idea to use an old rotten tire for a spare... just when you need it, it fails...

BH is a very lame brand, i would never trust 'em... do yourself a favor and recap it when you have the funds...

Should last for at least a month or two since this power supply's brand new and unused

[momaka]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Interesting, on System General's website they have a schematic for a power supply that is 300W that utilizes the SG6105.
It does have OPP though.

It's possible. If I remember correctly, the SG6105 IC does support OCP/OPP protection. It doesn't necessarily has to be used, though.

Should last for at least a month or two since this power supply's brand new and unused

Or a bit more if the PSU is well cooled and there aren't any big loading resistors near the caps. Many cheaper PSU manufacturers have them, though.
Deer PSUs, for example, are known to put a 2W, 100 Ohm resistor on the 12v rail. That's (12^2)/100 = 1.44 watts of heat from that resistor. Doesn't seem like a lot, but it is - usually enough to darken the PCB and the sleeve on many caps.

[Khron]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Ultrafasts generate more heat than schottky's, they could have just gone with 1 to-247 40A schottky and it probably wouldn't have gotten any hotter than 2 20A ultrafasts in parallel. Might be a cost issue there, I don't know how much more a single 40A to-247 schottky costs versus 2 20A to-220 ultrafasts.

Yeah, schottkys DO have lower forward voltage drops than ultrafasts (and thus less heat dissipated for a certain current going through them), but i think i read somewhere that there are / might be peaks higher than 40v on the 12v line, rendering the use of (even 40v) schottkys somewhat risky.

And regarding the price, a quick browse through OnSemi's site shows that a to-247 60A 45v "L-series" schottky is 1,46$ (2$ for a "normal" 30A 45v), while a to-220 16A 200V ultrafast is 0,44$ (unit price for 1000pc, i believe).

[Newbie2]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

Or a bit more if the PSU is well cooled and there aren't any big loading resistors near the caps. Many cheaper PSU manufacturers have them, though.
Deer PSUs, for example, are known to put a 2W, 100 Ohm resistor on the 12v rail. That's (12^2)/100 = 1.44 watts of heat from that resistor. Doesn't seem like a lot, but it is - usually enough to darken the PCB and the sleeve on many caps.

I did not see any large resistors near any of the caps, so it should be alright.

Also, the Yate Loon 120mm fan should also provide ample ventilation for this power supply.

And regarding the price, a quick browse through OnSemi's site shows that a to-247 60A 45v "L-series" schottky is 1,46$ (2$ for a "normal" 30A 45v), while a to-220 16A 200V ultrafast is 0,44$ (unit price for 1000pc, i believe).

That price difference should explain why two ultrafasts were used instead of one TO-247 Schottky.

[Khron]

Re: Examination of Golden Field ATX-S628 550W peak power PSU

That price difference should explain why two ultrafasts were used instead of one TO-247 Schottky.

In fact, that could make one wonder why they "spoil" us with schottky's at all