Antec SP500P "resurrection"

[Khron]

Greetings everyone,

A couple weeks ago, my dad got a semi-fried SmartPower 500W from a cousin of mine. First-look "diagnosis": all the big "Fuhkyuu" bloated (even the primary ones), one resistor "well-done" (from the 5vsb area to the PFC controller chip, most likely its Vcc), one "tanned" zener in the same area, two pins of the KA7500 controller somewhat burnt.

I've attached photos of the "culprits". From what i've seen around this forum, the pcb is identical to the 350-400w versions, with the PFC section non-populated.

I'm going to chip-mod the 5vsb section, most likely with a TNY266's i've scavenged from a Dell PSU. I think/hope i've correctly "transplanted" everell's DM311 schematic to the TNY266, just too bad his photo-guide isn't quite as clear as i'd like it

Now, i intend / hope to revive it sometime soon. My hunch is that the PWM is quite surely fried, by the looks of things. With a bit of luck, i think i can source a replacement from a friend (it's pin-to-pin compatible with TL494).

I'd appreciate any tips you might have. It's a CWT-built unit, so it looks sturdy, to say the least, and it'd be quite a shame to go to waste

PS: Of course, it WILL get a re-cap (Nichicon, UCC, Rubycon and whatever else i can find in my stocks, of the right values and can sizes)

Thanks in advance,
Chris

[everell]

Re: Antec SP500P "resurrection"

Good morning Chris, and welcome to the forum.

That power supply looks like it is going to be quite a project - but not impossible. I worked on a Antec SP350 with similar problems, and can offer some tips.

The pwm chip, KA7500, is most likely bad. The burnt pins are Vcc power coming in and output to driver transistor #2. This means the the driver transistor #2 is probably shorted. On mine, I replaced the 7500 and both driver transistors, H945. If you have other junked out boards around, the 7500 chip and the 945 transistors are very common. One of the driver transistors also had a leaky diode which prevented proper startup. I replaced the diodes on both driver transistors.

The main switching transistors were leaky, and the resistors R6 & R8 were fried beyond recognition. Replacement value is 2.7K. I also replaced a few other resistors in the switching circuit that had signs of burning. Also important to replace the 4.7 uF/50V capacitors in that circuit.

As for the 5vsb circuit, mine was working when I started, measuring 12.15 volts. That is why I gutted that circuit and replaced it with my DM311 mod. Important to understand that the DM311 is NOT the only chip that can be used, and you can probably make other chips work. The DM311 operating frequency is 67 Khz, which worked with the transformer in my psu. Check the data sheet for whatever pwm chip you plan to use. If it runs at a higher frequency, such as 132 Khz, the transformer might not work properly with it.

Keep us posted with your progress. If you have any questions, I will be happy to try to help. Toasty has also done a lot of work on the Antec SP series.

[Brian C]

Re: Antec SP500P "resurrection"

Is the MOSFET switch for 5V standby Transformer OK ? If not, please check the primary winding of standby 5V trans.

[Khron]

Re: Antec SP500P "resurrection"

Regarding the switcher IC, i also have an ST Viper12A @ 60KHz (the TNY266 switches @132KHz).

I'm ready to do a full re-cap of the board, even those many small electrolytics. One thing though, i'm wondering if the 0.1uF and 0.22uF ones could be replaced with film caps (and maybe with a series resistor, to "emulate" the higher ESR). Those are the only values of 'lytics i don't have in stock.

everell , i can post the schematic for the Viper12A-based switcher, which i modeled after your DM311 design, if you'd be willing to "double-check" it

BrianC , i intend to remove / bypass the discrete 5Vsb switching circuit, and replace it with an IC.

[everell]

Re: Antec SP500P "resurrection"

Regarding the switcher IC, i also have an ST Viper12A @ 60KHz (the TNY266 switches @132KHz).

You could try both, then let us all know your results.

I'm ready to do a full re-cap of the board, even those many small electrolytics. One thing though, i'm wondering if the 0.1uF and 0.22uF ones could be replaced with film caps (and maybe with a series resistor, to "emulate" the higher ESR). Those are the only values of 'lytics i don't have in stock.

I used Nichicon VZ aluminum caps.

everell , i can post the schematic for the Viper12A-based switcher, which i modeled after your DM311 design, if you'd be willing to "double-check" it

Would love to look at it - perhaps can make recommendations - but the final proof is how well it performs.

BrianC , i intend to remove / bypass the discrete 5Vsb switching circuit, and replace it with an IC.

Brian is correct. Take the time to check that big 5vsb switching transistor. If it has emitter to collector short, or drain to source short, you may have a shorted primary winding on the 5vsb transformer. Another check, two of the windings should read almost short. The third winding, which is the primary, should measure a few ohms, distinctly higher in resistance than the other two windings. If all read about the same, that is like a short, then its time to start looking for another transformer. You will need a good transformer to use with your pwm integrated circuit.

[i4004]

Re: Antec SP500P "resurrection"

brian said to check primary winding of trafo...
then again, that's not so easy without trafo ring tester...like this one
https://www.badcaps.net/forum/showthread.php?t=7817
OR unless you have a reading from a good one, like everell just gave...heh
(everell was a bit faster to reply...)

but trafo is probably ok, unless you see traces of excessive heat in +5vstby circuit.
and even then, trafo will probably be the last one to go...mobo will probably be destroyed prior to that happening...

--------------

60khz is probably better idea than 120....one wonders how would same trafo act with 2x the frequency(as everell says), so it's probably better to play safe...

[Khron]

Re: Antec SP500P "resurrection"

Ok, just checked the trafo and the 5vsb switcher. The trafo windings read 2.7ohms ("main" primary), 1.0ohms (two primary taps in the middle), 1.1ohms (last two taps, towards the optocoupler), so no shorts there. The switcher seems ok too, all three pin-to-pin measurements showed "out of range", so no shorts there either.

So far, so good, i suppose

[Toasty]

Re: Antec SP500P "resurrection"

>>all three pin-to-pin measurements showed "out of range", so no shorts there either.
<<

On Diode Test function? I assume DMM. Should show something.

D10 is likely shorted taking out R11. Check D8 and Q3 and solder side Q4 (1P), ZD1, R13(chip 24Ω) on Source leg of Q3. All of this likely caused 5vsb to go high taking out the SMPS controller KA7500C. That crack in the top of the chip is -not- an ion ram scoop.

[Khron]

Re: Antec SP500P "resurrection"

Previous test was on 200ohm scale. On diode test, it says "534" (red-drain, black-source), and "634" (red-drain, black-gate), other combinations show "out of range". This is for Q3 (main 5vsb switcher).

Both D8 and Q3 look ok visually (diode right up against trafo, and the main 5vsb switcher), D8 measures in at 541 (mV, i'm guessing).

Q4 measures ok, but i'm not sure (618 red-B black-C, 625 red-E black-C).

ZD1 says 707 (red-edge of pcb, black-inside) and goes "out of range" the other way around.

R13 looks ok, and measures in at 25 (says "240" on it, but i blame the crappy DMM for the 1ohm inaccuracy ).

[Krankshaft]

Re: Antec SP500P "resurrection"

A resistance check of the transformer will only show if its shorted completely it won't show a few shorted windings FYI. The only way to do that is with a ring tester.

But the chances of a dead 5 vsb transistor is rare.

Teapos on the 5 VSB nice I see they kicked the bucket too I had a TP 2.0 350W i recapped with those on the 5 VSB thought they would fare better guess I was wrong.

You have to use Panasonic FC on the two 1000uf 10V on the 5 VSB they make a 10mm diameter version that fits perfectly if the caps are any wider in diameter the rear exhaust fan will rub against them when it decides to come on. Yes it's that tight.

[Khron]

Re: Antec SP500P "resurrection"

I don't have any FC's in that case size, but i think i have some Chemicons or Nichicons that'll fit the bill (somehow). I believe a "lesser" cap of a good brand can outperform / outlast a crap-cap anyway

[Khron]

Re: Antec SP500P "resurrection"

Quick update: i just sourced a free TL494, so the PWM replacement's assured now

[Khron]

Re: Antec SP500P "resurrection"

everell , here's the schematic and the pcb for the VIPer12A 5vsb board, and HERE is the application note i used to "transplant" this, to make it look like your DM311 circuit.

[Khron]

Re: Antec SP500P "resurrection"

Yep... I think it's safe to say this chip has seen better days...

Possibly noob-ish question: if i were to power up the psu without the pwm in it (as in just plug it in the wall socket), it shouldn't blow up, right? Or at least not before i'd have a chance to check the 5vsb, for instance...

[everell]

Re: Antec SP500P "resurrection"

everell , here's the schematic and the pcb for the VIPer12A 5vsb board, and HERE is the application note i used to "transplant" this, to make it look like your DM311 circuit.

Looks good to me. Give it a try and let us know how it works.

[everell]

Re: Antec SP500P "resurrection"

Possibly noob-ish question: if i were to power up the psu without the pwm in it (as in just plug it in the wall socket), it shouldn't blow up, right? Or at least not before i'd have a chance to check the 5vsb, for instance...

If no other problems with the psu, that is correct. If you see or smell smoke, turn off immediately because the main switching transistors are bad. You can resolve that problem by unsoldering them and taking them and the heat sink out, then continue with evaluating your 5vsb circuit.

[Khron]

Re: Antec SP500P "resurrection"

I'll get to that later today, when i get back from school / work (7:35am right now, over here).

Although I must admit, i'm still a bit worried about that "well-done" zener (or whatever it is, visible on the solder-side photo), and that "coverless" resistor in the PFC Vcc (first photo)...

[Krankshaft]

Re: Antec SP500P "resurrection"

PFC Vcc??? Don't you mean PWM?

These supplies only have passive PFC.

[Khron]

Re: Antec SP500P "resurrection"

Not this one. From what i noticed, my 500w version and the 350-400w versions share the same pcb design, but those two don't have the places filled; mine does. The PFC coil's opposite the input filter, and it has an 8-pin KA3844 controller, if my memory serves me right. Also, the 350-400w ones don't have the resistor that's blown on my board (which goes to the KA3844's Vcc pin).

[everell]

Re: Antec SP500P "resurrection"

Before power up - would you please post a few more pictures - showing that pfc circuit, and a top down showing entire circuit board.

As for your new 5vsb circuit board, how are you going to mount it? I didn't see any mounting holes. It needs to be securely mounted to something.

Looks like your time zone is about six hours ahead of here. Are you in UK or Europe?

Photo of your 7500 chip with "ion ram scoop" looks cool!