I blew up my Zalman 300w power supply!
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee, Devonavar
-
- Patron of SPCR
- Posts: 946
- Joined: Sat Sep 06, 2003 9:57 pm
- Location: Berkeley, CA, USA
- Contact:
I blew up my Zalman 300w power supply!
Reader's Digest condensed version:
- Zalman 300w power supply
- Modified with 80mm Panaflo L fan
- Running 24/7 for 4-5 months
- Dual undervolted athlon XP (barton) 1.7ghz server measured at ~150w draw with Kill-a-watt (dual folding)
Ran fine for about 4-5 months, then last night, POW! When it went, the UPS it was attached to actually lost power briefly, which turned off my main PC (it's a large UPS).
The power supply didn't run that hot.. it was warm but not uncomfortable to touch for extended perions. But I guess the constant load eventually proved too much for it. And it is dead.. I hooked that PS up to a 2nd system and it won't even power on at all.
The system itself was fine with a new power supply (did a quick swap), so there was no damage. Just the bizarre UPS reaction.
I guess there is some wisdom in Mike's suggestion to use the Panaflo "M" (MEDIUM AIRFLOW) instead of the "L" (LOW AIRFLOW) for these power supply mods...
- Zalman 300w power supply
- Modified with 80mm Panaflo L fan
- Running 24/7 for 4-5 months
- Dual undervolted athlon XP (barton) 1.7ghz server measured at ~150w draw with Kill-a-watt (dual folding)
Ran fine for about 4-5 months, then last night, POW! When it went, the UPS it was attached to actually lost power briefly, which turned off my main PC (it's a large UPS).
The power supply didn't run that hot.. it was warm but not uncomfortable to touch for extended perions. But I guess the constant load eventually proved too much for it. And it is dead.. I hooked that PS up to a 2nd system and it won't even power on at all.
The system itself was fine with a new power supply (did a quick swap), so there was no damage. Just the bizarre UPS reaction.
I guess there is some wisdom in Mike's suggestion to use the Panaflo "M" (MEDIUM AIRFLOW) instead of the "L" (LOW AIRFLOW) for these power supply mods...
Re: I blew up my Zalman 300w power supply!
My sympathies. Glad the rest of the system didn't give up their magic smoke as well.wumpus wrote:When it went, the UPS it was attached to actually lost power briefly, which turned off my main PC (it's a large UPS).
From the symptoms you described, sounds like in its last moments there was a surge in power which caused your UPS to shut down to prevent damage.
Are you going to open it up, or will you replace it under warranty? (Warranty is void, of course, since you broke the seal to mod the fan, but we're not discussing ethics here).
If you decide to do a post mortem, please post pictures. If there's a loud pop, chances are an electrolytic capacitor gave up its ghost. Remember to take proper precautions.
-
- Patron of SPCR
- Posts: 946
- Joined: Sat Sep 06, 2003 9:57 pm
- Location: Berkeley, CA, USA
- Contact:
I hooked the L1 directly into the wires where the old fan went, so it was being controlled by the PSU, not at fixed voltage.Did you run the Panaflo from the stock fan contoller or fixed voltage.
No, just quickie 1-to-1 fan swap.Any mods to fan controller ckt such as thermistor relocation?
Well, I was overdramatizing a bit when I said "pow". There wasn't actually a significant noise when the PSU died, at least not that I could hear over my main system powering off, and the UPS cycling. Definitely no explosions or anything, and no burning smells. It went out with a whimper instead of a bang.If you decide to do a post mortem, please post pictures. If there's a loud pop, chances are an electrolytic capacitor gave up its ghost. Remember to take proper precautions.
Sucks that it cost me $60 to find out that a L1A isn't enough In the future, I'd definitely recommend the M1A to anyone planning to run a server 24/7 with ~150w load!
alas.. i have a friend who just did this exact same mod, i'll warn him.wumpus wrote:I hooked the L1 directly into the wires where the old fan went, so it was being controlled by the PSU, not at fixed voltage.Did you run the Panaflo from the stock fan contoller or fixed voltage.
but are you certain the L1A swap is to blame?
i'd have thought that with this approach, if the psu was too hot the fan controller would just volt-up the L1A until it was pushing more air. i.e. so you'd end up with a faster spinning L1A instead of a somewhat slower stock fan, but similar cooling/pushinng similar quantity of air..
-
- Site Admin
- Posts: 12285
- Joined: Sun Aug 11, 2002 3:26 pm
- Location: Vancouver, BC, Canada
- Contact:
To share my PSU failure experiences -- outside of cataclysmic testbench failures...
SH (SuperPower?) 350W. Modded for my first p4-1.6 system with Panaflo 80L and thermistor reposition mod. Ran (hot) about 1 year before starting to buzz/whine. Pulled from system before failure. A couple minutes on the PSU loaderat ~150W (DC) load killed it.
Enermax 350. Don't recall exact model. Ran (hot) with thermistor move mod and no bottom fan for ~1.5 years. Buzzed/whined for a few days before dying on power up in system. No damage to other components.
Seasonic 300FS NPFC. Modded with Panaflo 80L and cut grill. Much abused, ran 2 years in variety of systems, died same way as SH.
None of the above drew more than ~160W AC for any length of time (even w/folding 24/7). It's hard to know exactly what were the exact causes for the failures. Chances are, overheating was the #1 culprit; even running the PSU over the long term at near the top of the acceptable thermal range probably shortens its life. I think there's also a chance that these were samples that on the lower end of the acceptable production variance.
I also have other PSUs (Seasonic, Fortron, Nexus) that have run 2 years or more 24/7 on folding and show no signs of any deterioration (ie -- whine/buzz). However, most are on open bench systems and thus have ambient operating temps at least 10 degrees lower than in-case PSUs.
SH (SuperPower?) 350W. Modded for my first p4-1.6 system with Panaflo 80L and thermistor reposition mod. Ran (hot) about 1 year before starting to buzz/whine. Pulled from system before failure. A couple minutes on the PSU loaderat ~150W (DC) load killed it.
Enermax 350. Don't recall exact model. Ran (hot) with thermistor move mod and no bottom fan for ~1.5 years. Buzzed/whined for a few days before dying on power up in system. No damage to other components.
Seasonic 300FS NPFC. Modded with Panaflo 80L and cut grill. Much abused, ran 2 years in variety of systems, died same way as SH.
None of the above drew more than ~160W AC for any length of time (even w/folding 24/7). It's hard to know exactly what were the exact causes for the failures. Chances are, overheating was the #1 culprit; even running the PSU over the long term at near the top of the acceptable thermal range probably shortens its life. I think there's also a chance that these were samples that on the lower end of the acceptable production variance.
I also have other PSUs (Seasonic, Fortron, Nexus) that have run 2 years or more 24/7 on folding and show no signs of any deterioration (ie -- whine/buzz). However, most are on open bench systems and thus have ambient operating temps at least 10 degrees lower than in-case PSUs.
-
- *Lifetime Patron*
- Posts: 1464
- Joined: Mon Dec 29, 2003 2:51 am
- Location: Elk Grove, CA
- Contact:
I am part of that small fraction of the world's population that pays that much for a UPS. *points at his APC Back-UPS CS. And it has saved me a few times in terms of saving work and getting shut down when the power goes out. Gotta love California rolling blackouts during the summer.Gholam wrote: I'm a PC technician in a computer store. And, as you can guess, the number of people here who invest $80-$100 in a UPS is nearly nonexistant.
-
- Patron of SPCR
- Posts: 946
- Joined: Sat Sep 06, 2003 9:57 pm
- Location: Berkeley, CA, USA
- Contact:
Interesting, those sound like almost exactly what happened to mine. I'm starting to reconsider my "running kind of hot doesn't matter" philosophy when it comes to PSUs.To share my PSU failure experiences -- outside of cataclysmic testbench failures...
At least no system components were lost when they died.
Also, could possibly explain why a lot of prior fanless PSUs mysteriously disappeared from the market-- excessive failure rate due to heat.
-
- Patron of SPCR
- Posts: 946
- Joined: Sat Sep 06, 2003 9:57 pm
- Location: Berkeley, CA, USA
- Contact:
I opened up my dead Zalman 300w PS and I don't see anything at all unusual-- no swollen capacitors, nothing burned or blackened. I am by no means a PSU expert but it looks absolutely functional with the casing off. But unfortunately it isn't.. still won't power up at all on my bench PC. Ah well, at least I can salvage the panaflo L I had in there.
Given Mike's track record plus this experience, I am definitely going to be more conservative when it comes to PSU silencing mods from this point forward. Panaflo "M" at minimum..
Given Mike's track record plus this experience, I am definitely going to be more conservative when it comes to PSU silencing mods from this point forward. Panaflo "M" at minimum..
-
- Posts: 127
- Joined: Sat Apr 17, 2004 7:49 pm
- Location: Australia
I'm just one of the unlucky ones thenGholam wrote:Oh come on sthayashi, PSUs frying systems aren't that common - even bad, cheap, chinese PSUs. I have a rather large pile of dead PSUs at work, of which I dispose regularly, but there has only been a handful of times when it resulted in damage to system.
I've had two whole systems get up and go due to faulty power supplies. And the power supplies were unmodified.
-
- Posts: 1283
- Joined: Wed Sep 03, 2003 1:35 am
- Location: Sweden, Linkoping
Every PSU brand is bound to have at least a small fraction of the PSU's die after a fairly short period of time.
The gracefull death of this Zalman 300W PSU is quite promising. No loud POFF, no smoke, didn't kill anything else etc is rather nice.
Defenatly a PSU that I could consider using in one of my computers.
Since the PSU could have died due to heat caused by low airflow the chance that one of the MOSFETs died is quite high. Although time to check and repair might not be worth it comparing with the cost of a new PSU.
The gracefull death of this Zalman 300W PSU is quite promising. No loud POFF, no smoke, didn't kill anything else etc is rather nice.
Defenatly a PSU that I could consider using in one of my computers.
Since the PSU could have died due to heat caused by low airflow the chance that one of the MOSFETs died is quite high. Although time to check and repair might not be worth it comparing with the cost of a new PSU.
I'm happy that these quality PSU's don't make more damage when they blow up...
...unfortunately I can't say the same about several no-name PSUs I've had for years (before I met the now nearly extinct SPC ML). I may have blown up aprox 4-5 "El-cheapo" PSUs, all of them unmodified. It seems that heat is enemy #1 for them, and you know, we have plenty of that here in Summer. When I opened 2 of those PSUs everything was blown up, except for the fuse.
But for now, my modified Enermax still runs fine, although it whines sometimes, so based in MikeC's experience I'll have an eye (or ear, whatever) on it.
And my unmodified Seasonic, Zalman and my new Fortron (20$ new) I have in other systems also run great with quite low noise (ordered from lower to higher noise).
...unfortunately I can't say the same about several no-name PSUs I've had for years (before I met the now nearly extinct SPC ML). I may have blown up aprox 4-5 "El-cheapo" PSUs, all of them unmodified. It seems that heat is enemy #1 for them, and you know, we have plenty of that here in Summer. When I opened 2 of those PSUs everything was blown up, except for the fuse.
But for now, my modified Enermax still runs fine, although it whines sometimes, so based in MikeC's experience I'll have an eye (or ear, whatever) on it.
And my unmodified Seasonic, Zalman and my new Fortron (20$ new) I have in other systems also run great with quite low noise (ordered from lower to higher noise).
Uhh..
What UPS do you have and have you run a diagnostic on it lately?
APC Smart UPS have an anoying habit of taking out systems when their batteries get old. I have lost two power supplies and one complete system when the batteries on APC UPS's of mine have failed. Work lost a dozen systems before they stopped buying APC. I know dozens of people who have the same experience.
Run the diagnostic weekly and replace batteries or discontinue use of the UPS when they say they need replacing or you are risking your equipment.
What UPS do you have and have you run a diagnostic on it lately?
APC Smart UPS have an anoying habit of taking out systems when their batteries get old. I have lost two power supplies and one complete system when the batteries on APC UPS's of mine have failed. Work lost a dozen systems before they stopped buying APC. I know dozens of people who have the same experience.
Run the diagnostic weekly and replace batteries or discontinue use of the UPS when they say they need replacing or you are risking your equipment.