LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

[jazzie366]

So I was planning on piecing together an Intel Core 2 Quad with this PSU and a tiny graphics card just as a "beater pc" so to speak. Nothing special, just did it's job and could run small games like TF2 without lag and such. I have a main gaming pc, this one is just for when I'm not home.

Case pic:

Some Ltec primaries, they test fine. There's also a Rubycon right next to them, it's the only japanese cap in this entire psu.




Top view with bad cap by the fan.



Obviously shit on the meter.



Some Nichicon PL series, they're rated for 0.07 ESR at 20C. They FAT though for this application, 12.5x20mm case. However, there was enough space and these were a little higher capacitance so what the hell, why not.





Installed and dwarfing the other caps, and most other components on the board.



Alright, it now turns on nice, now lets look at some of the voltages.

3.3v rail

5v rail

12v rail

Well, that's odd.

I put a 10 ohm resistor on it to see if it'd go up, and instead it went down to 11.71. 6 ohms made it go down to 11.61, fluctuating between high 11.5's and low 11.6's. 2 ohms = lasts for about 10-20 seconds and then died. I let the caps decharge and the LED go out, and it came back and would stay on at 11.4-11.5 volts at best.

None of the heatsinks got really warm enough for me to be worried or show any obvious signs of where the problem might be.

There's only a few other Ltec caps on the board, the rest are Taicon, which imo is fine for being low wattage.

Any ideas as to what's causing this?

[Per Hansson]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

You need to load all rails to get the unit to behave.

[PeteS in CA]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

23 year old Nichicon PL series caps in tape-and-reel? Where'd you find those? BTW, the PL series was replaced in the mid 2000s with the ROHS-compliant PM series.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

23 year old Nichicon PL series caps in tape-and-reel? Where'd you find those? BTW, the PL series was replaced in the mid 2000s with the ROHS-compliant PM series.

They're just stuff I have around the shop. I work with my father in a vintage amplifier repair shop. The thing is, he's really limited in what he can teach me due to him working on mainly vintage equipment. Stuff like Fender guitar amplifiers, we restore things like those, I have so many tubes it's not even funny. When it comes to older tube related things, I know them very well.

I've got a stock of around 3000 tubes, and lots of older capacitors, can style ones wax caps, you name it I've probably got it.

However, when it comes to new equipment like with switch mode supplies, I'm not that well versed as you guys can probably tell. To me, a power supply is a transformer, rectifier tube, and a lot of high voltage caps. There weren't many amps until the later 90's that used actual rectifiers (except for an extremely small number) Basically what I'm saying is I didn't work with much of the newer electronics when learning.

I'm more old school, than new school.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

You need to load all rails to get the unit to behave.

Just tried that, won't hold voltage still.

Loaded the 12v rail with just over an amp, 4v rail with 1/2 amp, and 3.3v with an unknown ammount of current, but there was a load there (hard drive) and it still wouldn't hold voltage. Should I just put it into the pc and see what it does?

UPDATE: I put it into the pc, got a memory error (bad RAM) replaced the ram and it posted just fine with an 11.67V on the 12v rail. So it's something else, however it seems to run just fine?

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Alright so after letting the pc run for awhile:
It runs fine, no instability and an almost perfect dc sine wave on the output.
The PC now has 8GB of RAM in it, all matched Samsung sticks and is running beautifully, just updated windows and it still ran just fine. I'm not really sure what's causing the PSU to act up.

When under load, I can bring the voltages down to around 11.5V, sometimes touching high 11.4's yet the system acts as if this is totally normal.

It's running the 8500GT I recapped a few weeks ago, with an intel Core 2 Quad q8300. I just can't get this thing to fail.

Anybody know what's the deal here???

[ChaosLegionnaire]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

the atx specs for the 12v rail say it can have a ±5% variance so anything from 11.6v to 12.4v is within spec. at idle, the 12v rail is barely within spec at 11.67. but at load it drops below 11.6v which is out of spec.

fortunately and unfortunately (why i'll explain later), u appear to be using it on a "modern" lga775 board so the board is probably well designed to handle out of tolerance voltages so it continues to run normally. however, this can shorten the lifespan of the switching regulators on the mobo by forcing them to work harder and produce more heat. therefore, i recommend u stop using the pc until u iron out the undervoltage issue on the psu. improperly functioning psus can kill mobos.

from the power label on the psu, it appears that this psu is actually designed to work in a pentium 3 or older system when the load was primarily on the 5v rail. "modern" lga775 systems are primarily heavy on the 12v rail, therefore this psu is not designed to handle that (this is called crossloading) or "modern" loads (it was designed to handle "vintage" pentium 3 or older loads) so the voltage on the 12v rail sags and goes out of spec.

from what happened when u lowered the load resistors on the 12v rail, what i suggest u can do to "mod" this psu for "modern" 12v heavy usage as a temporary bandaid measure is to increase the load resistor resistance for the 12v rail to reduce the current draw on this rail and decrease the load resistor resistance for the 5v rail to increase the current draw on this rail. this should force the psu to bring the 12v rail voltage up by tricking it into thinking there is a heavy load on the 5v rail. try that and report back what voltages u get. also a word of caution: DONT use too low of a resistance for the 5v rail load resistor. it may fry the psu pcb due to too high current draw through the traces.

there is also another, more efficient way of modding this psu for 12v heavy usage by replacing the rectifier/diodes for the 12v rail but that is for expert users only. if done incorrectly, it may fry the psu AND motherboard! maybe another more experienced member can help u with that.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

the atx specs for the 12v rail say it can have a ±5% variance so anything from 11.6v to 12.4v is within spec. at idle, the 12v rail is barely within spec at 11.67. but at load it drops below 11.6v which is out of spec.

fortunately and unfortunately (why i'll explain later), u appear to be using it on a "modern" lga775 board so the board is probably well designed to handle out of tolerance voltages so it continues to run normally. however, this can shorten the lifespan of the switching regulators on the mobo by forcing them to work harder and produce more heat. therefore, i recommend u stop using the pc until u iron out the undervoltage issue on the psu. improperly functioning psus can kill mobos.

from the power label on the psu, it appears that this psu is actually designed to work in a pentium 3 or older system when the load was primarily on the 5v rail. "modern" lga775 systems are primarily heavy on the 12v rail, therefore this psu is not designed to handle that (this is called crossloading) or "modern" loads (it was designed to handle "vintage" pentium 3 or older loads) so the voltage on the 12v rail sags and goes out of spec.

from what happened when u lowered the load resistors on the 12v rail, what i suggest u can do to "mod" this psu for "modern" 12v heavy usage as a temporary bandaid measure is to increase the load resistor resistance for the 12v rail to reduce the current draw on this rail and decrease the load resistor resistance for the 5v rail to increase the current draw on this rail. this should force the psu to bring the 12v rail voltage up by tricking it into thinking there is a heavy load on the 5v rail. try that and report back what voltages u get. also a word of caution: DONT use too low of a resistance for the 5v rail load resistor. it may fry the psu pcb due to too high current draw through the traces.

there is also another, more efficient way of modding this psu for 12v heavy usage by replacing the rectifier/diodes for the 12v rail but that is for expert users only. if done incorrectly, it may fry the psu AND motherboard! maybe another more experienced member can help u with that.

Well here's a twistI'm just gonna throw in there: This came with the PC I put it in, the pc had been sitting on the bench for awhile with a no power on issue that I just didn't care to fix.

Well, now that I wanted to use it this issue came about.

The thing is though, we get 19A @ 12v on the rail, so that's almost 240W of 12V power. Now, this CPU that is has has a max TDP of 95W. As for the GPU, it has a tiny TDP, it only uses a molex connector to power it, and it doesn't get very hot or do much unless I stress it.

What I'm trying to say is that this thing should be able to handle the load it's getting, because the load isn't anywhere near where it's supposed to handle.

Now, I'm down for replacing the rectifier schottkys, but if I was gonna do that, I'd like to rebuild the supply entirely.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Looked into the issue a bit further, these might work but I have to check the pinouts and resistances of this chip.

http://www.ebay.com/itm/4pcs-Interna...UAAOSwLVZV3kJR

[ChaosLegionnaire]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Well here's a twistI'm just gonna throw in there: This came with the PC I put it in, the pc had been sitting on the bench for awhile with a no power on issue that I just didn't care to fix.

Well, now that I wanted to use it this issue came about.

The thing is though, we get 19A @ 12v on the rail, so that's almost 240W of 12V power. Now, this CPU that is has has a max TDP of 95W. As for the GPU, it has a tiny TDP, it only uses a molex connector to power it, and it doesn't get very hot or do much unless I stress it.

What I'm trying to say is that this thing should be able to handle the load it's getting, because the load isn't anywhere near where it's supposed to handle.

Now, I'm down for replacing the rectifier schottkys, but if I was gonna do that, I'd like to rebuild the supply entirely.

ok, after u said that, there could be a few more possibilities too. one, there could be a partial short on the 12v line somewhere on the board or video card drawing a lot of current thus pulling it down. the liteon looks almost like it could be a generic power supply so it may lack proper or working short circuit protection. u previously mentioned the system had a no power issue. it is possible some kind of a short is preventing the psu from starting or the mobo from starting due to too low 12v voltage.

which also brings me to point two. if it is indeed a generic kind of psu, the power and current ratings on the label cant be trusted. according to the label, it claims to be a 300w psu. but if its generic build quality, the actual wattage would be more like half that. 150w is too little power to run a quad core system. of course to be sure, we need either photos of the rectifiers on the secondary side or u pull the rectifiers and tell us and post the part numbers here.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

ok, after u said that, there could be a few more possibilities too. one, there could be a partial short on the 12v line somewhere on the board or video card drawing a lot of current thus pulling it down. the liteon looks almost like it could be a generic power supply so it may lack proper or working short circuit protection. u previously mentioned the system had a no power issue. it is possible some kind of a short is preventing the psu from starting or the mobo from starting due to too low 12v voltage.

which also brings me to point two. if it is indeed a generic kind of psu, the power and current ratings on the label cant be trusted. according to the label, it claims to be a 300w psu. but if its generic build quality, the actual wattage would be more like half that. 150w is too little power to run a quad core system. of course to be sure, we need either photos of the rectifiers on the secondary side or u pull the rectifiers and tell us and post the part numbers here.

LiteOn is like BesTec, which is also good but LiteOn is better build with better components. Also the output rectifiers are all STMicroelectronics chips, I don't like ST that much as they seem to make chips that are equal to others, but always with some small trade off.

Now, as for LiteOn, it can definitely do it's wattage. I've done some research and I'm just going to get some ultra fast recovery rectifiers for the output, really nice ones to replace this, as this PSU is a 24 pin with a lot of connectors on it, and I'd like to be able to reuse it in other computer systems after this.

Now, as for the parts, I'm probably going to go with some Fairchild components, as they are the best right along side with SanKen (IMO).

[Wester547]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Maybe the PSU simply needs a full recap except for the lone Rubycon and the input electrolytics?

I think it is somewhat risky using 23-year-old NOS, especially when it's from a series which is prone to leaking from the bung over time. The PR, PL, and PF series from Nichicon all did this eventually due to the usage of quaternary ammonium salts in the electrolytic solution, which rubber seals quickly degrade under due to the pH balance changing from acidic to basic, or alkaline. To be more specific, those particular electrolytes were used because their conductivity was very good for the time, and they were fully liquid even at room temperature. They are (or were) called "quaternary" because "quaternary" is the fourth bond that neutral nitrogen can form if a positive charge is allowed. They also have excellent low-temperature characteristics, which was remarkable for the time because the conductivity of the salts would certainly decline at low temperatures as the free ionic mobility decreased, also increasing the surface tension between the electrolyte and the dielectric. In the end, though, those old caps would irrevocably leak their insides over the PCB, eat away live tracks, and then ...

That said, I suppose anything is better than a blown capacitor, especially crappy LTEC. It seems that it's more ripple current than ambient heat that kills crap brands, which doesn't surprise me since they probably lack the neutralizers, oxidizers, hydrogen absorbers, and depolarizers to deal with any hydrogen gas that occurs as the direct result of the electrolyte conducting ripple current (which is probably why most LTEC datasheets, as well as many bad cap brand low ESR cap datasheets, do not include the full ripple current in their "load" tests).

But yes, both Lite-on and Bestec generally make good PSUs. This PSU is definitely worth a repair.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Maybe the PSU simply needs a full recap except for the lone Rubycon and the input electrolytics?

I think it is somewhat risky using 23-year-old NOS, especially when it's from a series which is prone to leaking from the bung over time. The PR, PL, and PF series from Nichicon all did this eventually due to the usage of quaternary ammonium salts in the electrolytic solution, which rubber seals quickly degrade under due to the pH balance changing from acidic to basic, or alkaline. To be more specific, those particular electrolytes were used because their conductivity was very good for the time, and they were fully liquid even at room temperature. They are (or were) called "quaternary" because "quaternary" is the fourth bond that neutral nitrogen can form if a positive charge is allowed. They also have excellent low-temperature characteristics, which was remarkable for the time because the conductivity of the salts would certainly decline at low temperatures as the free ionic mobility decreased, also increasing the surface tension between the electrolyte and the dielectric. In the end, though, those old caps would irrevocably leak their insides over the PCB, eat away live tracks, and then ...

That said, I suppose anything is better than a blown capacitor, especially crappy LTEC. It seems that it's more ripple current than ambient heat that kills crap brands, which doesn't surprise me since they probably lack the neutralizers, oxidizers, hydrogen absorbers, and depolarizers to deal with any hydrogen gas that occurs as the direct result of the electrolyte conducting ripple current (which is probably why most LTEC datasheets, as well as many bad cap brand low ESR cap datasheets, do not include the full ripple current in their "load" tests).

But yes, both Lite-on and Bestec generally make good PSUs. This PSU is definitely worth a repair.

Yeah, I'm gonna fully recap it later in time with some more suiting capacitors.

I'll throw something else other than that PL in there, also remember this likely won't leak from the bung as even if the rubber wears away, the cap is upside down so it's not gonna leak from there likely. However, I'll throw something else in there. If I do recap it though, it's gonna be all Nichicon.

[petehall347]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

when was the last time your meter was calibrated ?

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

when was the last time your meter was calibrated ?

Which one? Multi-meter or cap meter or my transistor tester.

If you mean multi-meter, it hasn't been calibrated ever, however I have more than one meter and they all read about the same.

[petehall347]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

ok it was just a thought with the voltages being pretty close .

[Longbow]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Alright so after letting the pc run for awhile:
It runs fine, no instability and an almost perfect dc sine wave on the output.

What is meant by "perfect dc sine wave"?

If I understand correctly, the previously faulty power supply runs fine after it is warmed up for awhile? If so, you have just described the most common behavior of equipment with aging capacitors. Typically, gear that is run 24/7 will keep running until it is shut off for a while, then won't start up again. Near-dead electrolytics seem to come back a little when they are warmed up.

I would get the supply warmed up and working. Take a can of freeze spray and carefully cool down individual caps. I think you will get some information using that method.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

What is meant by "perfect dc sine wave"?

If I understand correctly, the previously faulty power supply runs fine after it is warmed up for awhile? If so, you have just described the most common behavior of equipment with aging capacitors. Typically, gear that is run 24/7 will keep running until it is shut off for a while, then won't start up again. Near-dead electrolytics seem to come back a little when they are warmed up.

I would get the supply warmed up and working. Take a can of freeze spray and carefully cool down individual caps. I think you will get some information using that method.

I didn't mean sine wave, I was tired when I was posting and I didn't have full brain power. I meant the ripple coming out was low as it should be. As for the heat, I just threw it into the PSU and at first it will go down to around 11.4, and go up from there over time but not by much. 6A of load brings it to its knees. I'm gonna recap it with nichicon and see how it fares.

[momaka]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

When under load, I can bring the voltages down to around 11.5V, sometimes touching high 11.4's yet the system acts as if this is totally normal.

It's running the 8500GT I recapped a few weeks ago, with an intel Core 2 Quad q8300. I just can't get this thing to fail.

Anybody know what's the deal here???

Nothing.

C2Q with a GeForce 8500 and those voltage seem okay.

I have an older gen 250 Watt HiPro PSU in a PC with a Pentium Dual Core E2160 and Radeon 2400 video card, and my 12V rail easily drops to 11.75V under full load.

Not sure why, but it seems like there is something with socket 775 boards bringing the 12V rail on older PSUs a bit lower than usual. But it's fine.

the atx specs for the 12v rail say it can have a ±5% variance so anything from 11.6v to 12.4v is within spec.

Your math is a little off
+/-5% of 12 is 11.4 and 12.6.

So jazzie's PSU is still in spec.

fortunately and unfortunately (why i'll explain later), u appear to be using it on a "modern" lga775 board so the board is probably well designed to handle out of tolerance voltages so it continues to run normally. however, this can shorten the lifespan of the switching regulators on the mobo by forcing them to work harder and produce more heat. therefore, i recommend u stop using the pc until u iron out the undervoltage issue on the psu. improperly functioning psus can kill mobos.

It's really only the CPU VRM that has the highest load, so it is most likely to get affected by the drop in the 12V rail. That said, the board probably won't have any issues until the voltage on the 12V rail drops way lower than normal... like 8-10V or so.

from the power label on the psu, it appears that this psu is actually designed to work in a pentium 3 or older system when the load was primarily on the 5v rail. "modern" lga775 systems are primarily heavy on the 12v rail, therefore this psu is not designed to handle that

Actually, that PSU is from the "transition" period, when CPUs stopped using the 5V rail for power and started using the 12V rail. So that's why the 5V rail has such a high rating.

As such, PSUs from that era were actually designed to handle both 5V and 12V heavy systems. They are not good at either, but they are fine. All of my HiPros and Bestecs cross-load in a similar manner. But I have them recapped, and thus the only thing that can kill them is a sledge hammer. Otherwise, they are workhorses.

the liteon looks almost like it could be a generic power supply so it may lack proper or working short circuit protection.

Nah, that's a true Lite-ON PSU inside there. As far as I know, they are the only ones to use those inwards-curved heatsinks.

And no, it doesn't lack protections. If anything, it probably has pretty good short-circuit protection, being a single-transistor forward-converter design with a UC384x current-mode PWM IC. All of the big OEM PSUs from that era have excellent SC protection. I've tested more than a few shorted video cards and boards, so I'm pretty sure of that. It's the older generation of half-bridge PSUs with TL494/DBL494/KA7500 that people should be beware of, as many lack UV/SC lock-out and will try to power up again.

As for OCP: if you see a thick jumper near the 12V wires, then that PSU has OCP on the 12V rail as well. Many OEM PSUs from that era do, including LiteON, so I wouldn't be worried about overloading the 12V rail.

Now, as for LiteOn, it can definitely do it's wattage. I've done some research and I'm just going to get some ultra fast recovery rectifiers for the output, really nice ones to replace this, as this PSU is a 24 pin with a lot of connectors on it, and I'd like to be able to reuse it in other computer systems after this.

No, don't get Ultra Fast rectifiers!

If you want to increase the voltage output on the 12V rail, then use a schottky rectifier, as it will have a lower Vf drop. I recommend something like a STPS40H100 or two MBR30100's in parallel. With that, you will still probably be limited to 19A on the 12V rail due to other components, but at least the voltage sag on the 12V rail will be much smaller.

23 year old Nichicon PL series caps in tape-and-reel? Where'd you find those?

Anywhere but eBay and AliExpress, I hope.
Otherwise, he might be in for a surprise after a few months.

....
By the way, jazzie366, can you *please* use the forum image attachment tool to upload your pictures. Off-site hosting make the page load times terrible.

[jazzie366]

Re: LiteOn PS-5301-08HA Blown Ltec cap & won't hold good voltage

Nothing.

C2Q with a GeForce 8500 and those voltage seem okay.

I have an older gen 250 Watt HiPro PSU in a PC with a Pentium Dual Core E2160 and Radeon 2400 video card, and my 12V rail easily drops to 11.75V under full load.

Not sure why, but it seems like there is something with socket 775 boards bringing the 12V rail on older PSUs a bit lower than usual. But it's fine.


Your math is a little off
+/-5% of 12 is 11.4 and 12.6.

So jazzie's PSU is still in spec.


It's really only the CPU VRM that has the highest load, so it is most likely to get affected by the drop in the 12V rail. That said, the board probably won't have any issues until the voltage on the 12V rail drops way lower than normal... like 8-10V or so.


Actually, that PSU is from the "transition" period, when CPUs stopped using the 5V rail for power and started using the 12V rail. So that's why the 5V rail has such a high rating.

As such, PSUs from that era were actually designed to handle both 5V and 12V heavy systems. They are not good at either, but they are fine. All of my HiPros and Bestecs cross-load in a similar manner. But I have them recapped, and thus the only thing that can kill them is a sledge hammer. Otherwise, they are workhorses.


Nah, that's a true Lite-ON PSU inside there. As far as I know, they are the only ones to use those inwards-curved heatsinks.

And no, it doesn't lack protections. If anything, it probably has pretty good short-circuit protection, being a single-transistor forward-converter design with a UC384x current-mode PWM IC. All of the big OEM PSUs from that era have excellent SC protection. I've tested more than a few shorted video cards and boards, so I'm pretty sure of that. It's the older generation of half-bridge PSUs with TL494/DBL494/KA7500 that people should be beware of, as many lack UV/SC lock-out and will try to power up again.

As for OCP: if you see a thick jumper near the 12V wires, then that PSU has OCP on the 12V rail as well. Many OEM PSUs from that era do, including LiteON, so I wouldn't be worried about overloading the 12V rail.


No, don't get Ultra Fast rectifiers!

If you want to increase the voltage output on the 12V rail, then use a schottky rectifier, as it will have a lower Vf drop. I recommend something like a STPS40H100 or two MBR30100's in parallel. With that, you will still probably be limited to 19A on the 12V rail due to other components, but at least the voltage sag on the 12V rail will be much smaller.


Anywhere but eBay and AliExpress, I hope.
Otherwise, he might be in for a surprise after a few months.

....
By the way, jazzie366, can you *please* use the forum image attachment tool to upload your pictures. Off-site hosting make the page load times terrible.

I can't figure out how to use the on site image attachments, it only gives me the option to use a link.

Also, I'll get a STPS40H100 and see how it works.