Unknown Hipro... but they look good. What would you do?

[Th3_uN1Qu3]

I'll let the pics speak. Condition unknown, one missing cap on one of them, bulging 5vsb cap on one of the others, but otherwise they look worth something. The date marked on them is 1999... Standard TL494/LM339 affair, Tosin primary caps, G-Luxon and Su'Scon secondaries. Some of them are bulging but otherwise the power supplies look alright.

I got five of 'em, and they look solid enough for me to feel sorry taking them apart. What do you think? Btw, the model number on the back gives no hits, maybe you can help me with some specs.

These are "the good ones" out of the bunch of boards that i got today. I have other surprises i'll show you later.

[mockingbird]

Re: Unknown Hipro... but they look good. What would you do?

Looks similar to a Young Year PSU I am recapping. I have 4 of the same exact model, but the caps differ. In one of them, there's all the stuff that needs to be there, the X-Capacitor, extra coils, so I am doing that one.

I hope to post pics of the project when it's done, right now the caps are on their way.

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

Current status:

Hipro #1: No 5vsb, dead
Hipro #2: 5vsb ok, tick-tick-tick-tick. Secondary side short. Will be an easy one.
Hipro #3: 5vsb @ -1.54v , dead otherwise
Hipro #4 (with bad cap in 5vsb area): 5vsb @ -1.59v , dead otherwise
Hipro #5 (with missing cap): Working! Missing cap is on -5v rail, will be populated. Voltages are in spec except 12v which comes up as 10.89v. Will try and see how it behaves under load.

[goodpsusearch]

Re: Unknown Hipro... but they look good. What would you do?

Why you cut the cables?

Today I went to the university to take Pcs, ups and other stuff from the point they gather the ones to be recycled but the romas found out the place and they take everything

I'm so sad.

[mockingbird]

Re: Unknown Hipro... but they look good. What would you do?

Haha! You got "Gypped".

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

Why you cut the cables?

That's how they arrived, i didn't cut anything. No cables no cases no fans.

I'm proud to anounce that Hipro #5 is now fully functional. I found another missing cap on the pi filter of the 12v rail, on the others there is a 1000uF 16v G-Luxon, but i put a 2200uF 16v KQC instead. I bought those KQC capacitors locally, i don't know how good they are, but the end of the pi filter isn't a demanding spot.

The missing cap on the -5v rail was filled with a 470uF 16v CS. Before you shoot me for using crap caps, after it is a 7905, that thing doesn't even care if there is a cap before it or not. All the caps i use regardless if new or pulled have tested good for ESR.

The 5vsb cap that was bulged on another one is a 470uF 10v Su'scon - IMO, good cap, bad spot (too low voltage rating). Was replaced by 470uF 16v G-Luxon preventively - might as well keep the overall style. Btw the primary caps are Meritek on this one, and they are 470uF (on all of them). Except for that one 5vsb cap there's no bloaters - and this is on a PSU from 1999 remember.

I fitted the thing with LOTS of cables. There's three strings of peripheral connectors, one SATA (didn't have a cable with two SATA around), 20-pin ATX, P6 connector, and ATX12v connector (just for show). Enough that i had a really hard time pushing all of them in. Too bad the PCB is designed for 20AWG cables only, but it'll do.

I only have a generic gray housing for it, but hey, what matters is that there's a real PSU inside. It probably has one of those little boost converters for the fan, because i remember this particular fan being a lot quieter. Oh well, i can change it later.

This is a 5v heavy PSU so it should be perfect for my dual-PIII - here's to hoping i won't need 2 PSUs in parallel anymore. As soon as a small load is connected (i used a hard drive), the 12v rail comes in spec. Pics tomorrow, i'm off to sleep now.

[goodpsusearch]

Re: Unknown Hipro... but they look good. What would you do?

Good sleep!

Btw I upgrade the 5vsb caps to 2x1000uF no matter what caps they had previously.

[kaboom]

Re: Unknown Hipro... but they look good. What would you do?

Can you take close-up pix of the standby sections of the working one and the two w/o standby/aux? Maybe the critical cap overheated and the switcher got stuck in its linear region and went bang and took out its source resistor?

Are you getting the ~-1.6V readings with a high impedance meter or load resistor? Probably phantom readings- shunt it with a 1K resistor.

-Paul

[PCBONEZ]

Re: Unknown Hipro... but they look good. What would you do?

Good sleep!

Btw I upgrade the 5vsb caps to 2x1000uF no matter what caps they had previously.

Why?

[PCBONEZ]

Re: Unknown Hipro... but they look good. What would you do?

i put a 2200uF 16v KQC instead. I bought those KQC capacitors locally,

Is there a series marked or just a brand?
What is their ESR?
.

[goodpsusearch]

Re: Unknown Hipro... but they look good. What would you do?

Why?

I find 2x470uF some old psus have at 5vsb a bit low for today. Just to be safe.

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

Can you take close-up pix of the standby sections of the working one and the two w/o standby/aux? Maybe the critical cap overheated and the switcher got stuck in its linear region and went bang and took out its source resistor?

Are you getting the ~-1.6V readings with a high impedance meter or load resistor? Probably phantom readings- shunt it with a 1K resistor.

-Paul

Will do. 5vsb always has minimum load so i don't think i need to shunt anything. If i try to power up the supply 5vsb goes to 0.something volts, but still negative.

Is there a series marked or just a brand?
What is their ESR?
.

It's just a brand. I measured them a while back but forgot what their ESR is.

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

I'm back on them. The one i fixed would take a while to start up, and had developed a ticking noise that went on till it started up. Just like the other one that i thought was shorted, but isn't... 5vsb was okay though, it just took a while to start up the main PSU. I'm not excluding bad caps, but i just left it aside for a while. Now not even 5vsb comes up properly, it fluctuates around 1 volt in sync with the ticking. For 5vsb it can't be bad caps as i changed ALL of them. Weird.

Since i cabled this one with all i need, i don't feel like just fixing another one, i want to fix this one. However they're so crammed it's not funny, so i picked up one of the boards without cables to do a study of it. It was one with the negative 5vsb. After some poking around i noticed a snipped black wire in the 5vsb area. "Hey wait a minute, doesn't this go...." Yup, it was the wire that was supposed to connect 5vsb to ground. After replacing it, a quiet buzzing could be heard. 5vsb was up at exactly 5.00 volts. I wired PS_ON to ground, and hey presto, main rails came up too.

Now i have a good one to compare with the ones that don't work.

[lti]

Re: Unknown Hipro... but they look good. What would you do?

it was the wire that was supposed to connect 5vsb to ground.

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

Many times there isn't room to snake a ground trace into the 5vsb section, so its ground is brought by a wire.

Edit: Something seemed odd about the 5vsb mosfet. It was on a tiny heatsink and had a 47uF 50v 85C cap directly behind it. In the working supplies, this thing runs quite hot. The soldering was fresh on the 47uF cap (plus it was a PITA to remove), but upon pulling it i noticed it said "Su'Scon". Aka the same as in the other supplies. Mm, i thought i'd replaced it. Anyway, a replacement cap made no difference (i think that's why i left the original one in). A general resoldering of the area however did... for the worse. No more start up at all, and the voltage on the cap would sit at 1.8 volts, while it was ~20 volts with the thing running, and bouncing around 12v when it was ticking.

I decided to pull the mosfet from one of the other boards to take a look at it. I quickly noticed that my "mosfet" had 4 legs... it was an integrated switcher. A 1H065R. Mine had the SEC (Samsung) logo on it, but it was identical in all other aspects. Hmm... undervoltage lockout... 16 volts. Aha, that makes sense. Next i inspected the startup circuit. Charging the 47uF cap there's a 1N4148 and before it a 200k dropper resistor going to another resistor which goes to one of the transformer windings. There's probably a little self-oscillating thing somewhere which gets this going. Nothing wrong found with the diode and resistor. I added another 200k resistor across just in case, and all i got was that the cap voltage was now 2 volts. Not good.

Running out of any other ideas, i pulled out the 1H0165R and replaced it with the one from the other board. Tada! So it was a dud switcher after all. It also seems that my memory is wrong of the main power supply taking time to come on, as it powers up instantly.

[bigbeark]

Re: Unknown Hipro... but they look good. What would you do?

I agree with Mockingbird - I have a Youngyear and it looks very similar. My first thoughts on opening mine up were howcome this thing is so well designed, but the heatsinks and primary caps are so small. Mine was even dirtier inside than yours, but works just fine.
If it is 1999, I guess that explains it.

[Th3_uN1Qu3]

Re: Unknown Hipro... but they look good. What would you do?

The Youngyear is a different affair - are you sure you wanted to post here? Anyway, checked the 47uF 50v Su'scon again - it is in perfect condition.

After all these years near a hot little heatsink, it's funny how the 85C, general purpose, 2000 hour (!!!) capacitor survived but the switcher died. Of all of them there was only one case of bulging cap, on the 5vsb rail, but that had a pi filter afterwards so the only effect was that 5vsb was lower, at 4.72 volts. It also has a zener on 5vsb so no such thing as blowing the mobo up.

I now have 3 of 5 working. The one that has no standby probably has the same fault (dead 5vsb switcher IC) so i won't bother about it. The second one that used to tick does not tick anymore, but strangely it came up once, then again, nothing. 5vsb is okay on that one so i'll have to dig more when i'll feel like it.

It appears the integrated switcher is a current-mode controller with built-in mosfet and sense resistor (dubbed "SenseFET" by Fairchild). I'm sure it can be replaced by a UC3842 + mosfet and sense resistor. Now, how i am going to fit those... that's another deal. My hat goes off to whoever designed the layout.

Edit: On a "cold" measurement there doesn't seem to be anything wrong with this switcher. It did have a lot of fairy dust on it and between the pins - maybe that dust was of the conductive variety? I'll plug it into the board i pulled the other one out of and see what gives.