2 transistor 5vsb circuit?

[370forlife]

Can't tell if this unit has it or not. The 5vsb circuit doesn't have the usual signs of being two transistor.

The only IC on the board is a EST7502B except for a octocoupler next to the 5vsb transformer labeled "CC807 EL817 Everlight"

[Wizard]

Re: 2 transistor 5vsb circuit?

I have 2 transistor 5Vsb design by FSP for acer (E190414). That also blown because of bad Fuhjyyu on the 5Vsb. This also has 4 bloaters (Teapo and Jackson IIRC). In aother PSU, I had to recap one cap (Capxon on 12V) so I can use the PSU in computer for Boss's daughter.

Does not have to be just BESTEC at all!

Cheers, Wizard

[POM_MJ]

Re: 2 transistor 5vsb circuit?

The 2 transistors 5vsb circuit is mean 2 transistors on primary side, I think.

[everell]

Re: 2 transistor 5vsb circuit?

POM MJ has correctly identified the two transistors in the 2 transistor 5vsb circuit.

Wizard - you are right. Bestec is not the only problem ridden psu with 2 transistor 5vsb circuit. I got a Antec Smart Power SP350 and have been working on it this week. After replacing a burnt looking pwm chip (KA7500), both leaky switching transistors, one leaky driver transistor, a pile of Fuhjyyu capacitors, and some other cheapo brands of leaky capacitors, and six burnt/blasted resistors, the power finally comes up with good looking output voltages. HOWEVER, the 2 transistor 5vsb circuit measures 12.13 volts. I am truely amazed that the 5vsb circuit is not blown to hell. My guess is that someone lost a mother board over this one!

Cheers

[Toasty]

Re: 2 transistor 5vsb circuit?

Need that top down shot to be a bit more to the right to see better. Where is the "failure prone" cap and the anti-frost resistor?

The "Coil?" is indeed one and along with the 2 - 470uf caps makes a pi filter for the 5vsb.

From reports, SP350's are mobo killers.

[everell]

Re: 2 transistor 5vsb circuit?

My Antec SP350 used D209L switching transistors. Both were VERY leaky. I could not find them at Mouser, Digikey, or Newark. From Google, looks like they are a cheap Korean transistor. Anyone know of a good source for the D209L transistor, or what a good substitute would be?

[Toasty]

Re: 2 transistor 5vsb circuit?

Leaky? Could they have an internal diode?

[everell]

Re: 2 transistor 5vsb circuit?

Leaky? Could they have an internal diode?

Internal diode????? They were so leaky they would not even pass for a diode. They were very leaky as in low ohms - not totally shorted - but almost!

[everell]

Re: 2 transistor 5vsb circuit?

I have been troubleshooting the 5vsb circuit on this Antec Smart Power SP350. The pc board showed signs of heat burn around the opto isolator, so I decided to start there, replacing the optoisolator and also the AZ431 regulator chip. The 5vsb went from 12.13 volts to 12.83 volts. Whoa .... going the wrong way!!! So I started tracing the entire circuit - just another typical 2 transistor 5vsb circuit much like the Bestec. All resistors measured OK. I thought I had changed out all of the electrolytic capacitors ... but oops, missed one. And it just had to be the critical one in the 5vsb, but in a location on the pc board that was unexpected. So changed that capacitor and 5vsb came up measuring 5.03 volts.

From this I am slowly learning ... ALWAYS check 5vsb circuit first to be sure you have 5 volts. If not, change the critical capacitor.

Why should this capacitor fail more often than capacitors on the output rails? The capacitor value is typically in the range of 10 uF to 220 uF, usually about 22 uF. Capacitors on the output rails have a much higher capacitance value, therefore typically have a much lower esr. So if the 5vsb critical capacitor has a higher esr value and is filtering the same switching frequency, a failure should be expected sooner. This would explain why Wizard recommends increasing the value of the capacitor in this critical location. Also, the 5vsb circuit is ON when the computer is turned OFF (unless unplugged from the wall). So I would expect problems with this capacitor before capacitors on the output rails because it is working harder and more often.

Next, none of the capacitors on the output rails failed. That includes one Teapo and all else Fuhjyyu. The 5vsb critical cap that failed was a 22 uF/50 volt Teapo. Both caps in the 5vsb output pi filter were Fuhjyyu 1000 uF/10 volt, both bulging failures. I would expect this because putting 12 volts on a cap rated for 10 volts is asking for trouble. All other failing caps were in the switching circuits which also had transistor and diode failures. They were assorted Fuhjyyu, Teapo, G-Luxon, and Koshin. They were also located in the cavity formed between the two heat sinks - which is where I would expect more heat, especially when the computer is off.

I am beginning to think an analysis of this failure, and the efforts of several who looked at the Bestec problems, leads me to believe that we are actually seeing more problems with bad design, placing capacitors next to heat generating resistors, and the cavity formed by the two heat sinks acting like an oven. The fan may provide cooling when power is ON, but does not do a thing for you when power is OFF.

At this point, I am going to challenge that blacklisting Fuhjyyu as being bad capacitors in all situations may be wrong - that they may be no worse than some of the others. This is especially true when looking at all the different brands of caps used in the critical capacitor location of the 5vsb circuit. More proof (analysis) is needed before continuing to make Fuhjyyu our scapegoat.

[Toasty]

Re: 2 transistor 5vsb circuit?

D209L - Datasheet converted from 600k PDF to JPG in a size that can be uploaded here. With ENGLISH specs. lol

Does not include the characteristics graphs page. If you want that, LMK.

Should help in the search.

Where was this critical cap found? Pics?

[Krankshaft]

Re: 2 transistor 5vsb circuit?

Was it one of the two small caps wedged between the primary semis and the transformers ?

I always had some real fun with my tweezers getting those out and replacing them. I refused to remove the transformers or the heatsink out of laziness so that was the only option.

Did 4 PSUs and it got easier each time.

My Antec TP 2.0 350W on my email server which lived it's life in my basement had Fukyuus that never failed either. Of course I replaced them immediately.

[Krankshaft]

Re: 2 transistor 5vsb circuit?

I'd run the server for 4 years before I caught them. They didn't fail because of the lower temps in the ambient air and the server was run 24/7 so the fan was always running.

As for Fuhjyyu not being any worse they ARE the WORST.

I had an Antec with a few Teapos 2 on the 5VSB and 3 on the secondary the Teapos were FINE both on the 5 VSB and the 3 on the secondary while the remaining Fuhjyyus on the secondary were bulged.

These caps are so bad another bad cap brand beats them in reliability.

Not to mention when using brand name caps they are able to take the heat over the long term.

There is no need to defend them these are subpar shoddy components heat cannot always be avoided as electronics are getting smaller and smaller. The days of the tube chassis and the large steel chassis are over. The big cap manufacturers understand this engineer caps to match.

While fly by knight operations like Fuhjyyu produce crap for the lowest bidder to make a quick buck.

The ATX footprint isn't getting any bigger and the setup in these supplies is pretty much on par with any other PSU out there.

[everell]

Re: 2 transistor 5vsb circuit?

Toasty - I have the D209L data sheet, I just was wondering what a good replacement would be and where to get them. For this psu I had to raid my junk box - to get it working with small loading, I found one D209L and one 2sc2625. Works till I can get better parts.

Krankshaft - units getting smaller and having greater concentration of heat DEMAND better engineering. Placing an aluminum capacitor directly against a hot power resistor is NOT good engineering practice. Yet we see it all the time in these CHEAP power supplies.

As for Fujyyu capacitors - yes they are cheap inferior quality - but my comment was intended to say they are no worse than the other cheapo brands. Old Chinese proverb: "You get what you pay for.....if you are lucky!"

As we move into the future, I expect to see more 5vsb circuits with pwm IC instead of 2 transistor design because it is better. Also, I expect to see future designs with less aluminum electrolytic capacitors. Problem is that during an economic DEPRESSION like what we are in now .... how many companies can stay in business and give quality? Right now I don't think the public is willing to pay for it. So we can expect to enjoy these "crap caps", and Fujyyu is only one of them. I think some of the other brands are just as bad. BUT, I also think some of the problem is poor engineering and heat instead of JUST bad capacitor quality.

I am attaching a photo of the critical 5vsb capacitor on the Antec Smart Power SP350

[POM_MJ]

Re: 2 transistor 5vsb circuit?

Could be possible to replace D209L by E13009L ==>
or better 2SC3320 ==>

[Toasty]

Re: 2 transistor 5vsb circuit?

@everell-
Try this one perhaps?

http://search.digikey.com/scripts/Dk...me=497-3555-ND

[Per Hansson]

Re: 2 transistor 5vsb circuit?

Hmm, that's interesting
I will have to check up an old Antec Truepower I recapped a few years ago
What would you say is the maximum you can increase the capacitance of this critical 50v cap, would 100uF be ok?

[everell]

Re: 2 transistor 5vsb circuit?

Thanks for the transistor substitute recommendations. POM MJ - both you recommended were not available from Digikey or Mouser. Toasty - the one you recommended by STM was about two bucks each. I found what appears to be a happy compromise, a Fairchild FJA13009 at 88 cents each in quantity of ten. Looks to me to be about the same as the STM 13009.

Do you think the 5vsb going high was the root cause of all this damage...... or do you think it took a surge?

[PCBONEZ]

Re: 2 transistor 5vsb circuit?

I had Fuhjyyu's in a TP2 550w bloat BEFORE it was ever plugged in!
PSU was a few years old spare that was STILL IN SHRINK WRAP.
- The damned Fuhjyyu's bloated -> on the shelf -> with no power applied!
There are pics of it in some thread here.
.

[PCBONEZ]

Re: 2 transistor 5vsb circuit?

Hmm, that's interesting
I will have to check up an old Antec Truepower I recapped a few years ago
What would you say is the maximum you can increase the capacitance of this critical 50v cap, would 100uF be ok?

I'm not sure what you are talking about but I'd like to know too.

[PCBONEZ]

Re: 2 transistor 5vsb circuit?

Well, that's the thought that went through my brain anyway.
-
After reading up I know about that cap.
Some where in here Wizard did a pretty good discussion about it complete with some drawings.
Been looking for it but haven't found it.
Anyone know where that thread is?
.
THX
.