12V Rails: is high bad, too?

ikjadoon

Limp Gawd
Joined
Jun 21, 2006
Messages
308
Quick question. I know a low 12V is bad, but what about high?

My friend has a questionable power supply in his relatively high-end system (E8200, eVGA 680i, 8800GT, 4GB RAM, 2 DVD burners, hard drive powered via a Apevia 500W). Checking the 12V rail, it runs about 12.3V just in the BIOS.

The issue is that he is having some random freezes and some graphic tearing/freeze (not enough power to GPU?). Sound like PSU? RAM is MemTest OK and temps are fabulous. The only other thing that we think is the motherboard, as it is refurbished.

Just checking, thanks. :)

~Ibrahim~
 
The difference between 12V and 12.3V is 2.5%
Things may run a tiny bit warmer but it is well within the ATX spec.

Also, the readings themselves might not be very accurate, its best to measure with a good multimeter.
 
My friend has a questionable power supply in his relatively high-end system (E8200, eVGA 680i SLI 8800GT, 4GB RAM, 2 DVD burners, hard drive powered via a Apevia 500W). Checking the 12V rail, it runs about 12.3V just in the BIOS.

Even if the PSU wasn't the problem, I'd still recommend getting a new PSU nonetheless. Apevia makes absolutely shitty PSUs:
Apevia ATX-AS500W-BL
Apevia Warlock Power 900W
XQPack's Included 420W PSU
Aspire Beast 680W Power Supply

So make sure your friend gets a new PSU ASAP.

Anyway, Nenu is right: Software reading of voltages isn't very accurate. Test with a multimeter. Here's a FAQ on how to do that:
http://www.bfgpower.com/troubleshooting.html#DMM
 
I might be able to get a multi-meter; I guess I can find them at Radio Shack, possibly?

Yeah, I've definitely getting him a new one, but just wanted to check if it was PSU. He really wants one with blue LEDs like this one, so I'm looking at an OCZ 700W. Currently, this one is is using 20-pin adapter to use in the motherboard and only has the 4-pin for CPU. The latter isn't that much of a big deal (I think), but the former does freak me about a bit. I mean, it could be the motherboard, but the guy he bought this PSU from actually had BSOD issues in his last computer.

BTW, I edited my first post. It is a 680i SLI motherboard, not a 680i with SLI 8800GTs.

Wow, thanks for that review link, mate. I didn't know [H] did a review on his very one! I'll ask him tonight to see if he has any melting. Very freaky.

I guess he'll need a new one, anyways, as he would like to overclock and is thinking about SLI.

~Ibrahim~
 
Yeah, I've definitely getting him a new one, but just wanted to check if it was PSU. He really wants one with blue LEDs like this one, so I'm looking at an OCZ 700W.

I recommend ditching that OCZ GameXStream PSU. It's based off the FSP Episilon PSU design which has out of spec ripple, or voltage fluctuations, at high loads that can damage or kill your system. Granted, you may not reach such high loads but why pay for a PSU that can kill your system?

Oh and radioshack does have multi-meters IIRC. And since that PSU already has a past history of hurting hardware, get rid of it.. Especially since this happened in the above review:
It should be noted that I stopped testing on the first Apevia unit we had after Test #3 because the unit began to melt. I could not completely identify where the unit was melting from nor photograph it very well because the unit had to remain in tact in order to get an RMA as these units were purchased at retail. However, the second unit we received also began to melt during Test #3 but not nearly as badly and from what appeared to be a different location. This time the melting was obviously from some of the caulk used to secure the components on the PCB. It would seem that the temperature rating by Apevia may have been for a good reason, however it should be noted that even at these temperatures the manufacturer should not be using caulk or plastics that should melt. This unit is supposedly UL certified and if so I would be interested to find out how that happened.
 
Here are the voltages....

NAME: SETTING: CURRENT:
cpu core 1.225V 1.18V
cpu fsb auto 1.2V
memory 2.100V 2.100V
SPP auto 1.25V
mcp auto 1.500V
HT<->MCP auto 1.20V

Any clues? i took my gpu out, wasnt that.. put in ram that my buddy owns that works, still freezing/restarting.. .....

I'm going to flash my bios, i had the latest ones, maybe they're not workin with me, so ill get ones that are a little older.

I'm freezing/restarting for a good 15-20 minutes or so, then after that i dont freeze for a while.. I thought maybe the PSU wasnt delivering as much power to each capacitor or something, because i either freeze up, screen tear, or restart (possibly 3 different things).. Anyone have a clue? Ib's getting a mulimeter, and in the meantime Im going to be looking at power supplies to get, as this one i feel is the core of the problem.. Should i RMA the board?
 
Dont try and do everything immediately.
Get the new PSU and see how it goes.
 
I've looked around Newegg endlessly and I can't find a reputable manufacturer (minus a modular Raidmax.....) that has LED fans.

I think, Ross, it might be easier to get one without the lights and then grab a cathode light for like $8.

So, that leads me to my ultimate question: if the OCZ 700W is no good, which to buy? I know a fair bit about power supplies to stay away from no-name brands, but are there any stand-out awesome PSUs available for about $70-80? He is thinking about SLI down the road, so SLI-certified/ready might be nice. How does the FSB Blue Storm II 500W look? It has the beautiful blue paint job to kick. This should be able to handle SLI and a overclocked E8200, right? I was just hoping for some more wattage for future-proofing like an 8-pin PCIe. This OCZ ModXStream 600W looks pretty good, too, with modular cabling and an 6+2 PCIe connector plus a regular 6-pin.

~Ibrahim~
 
So, that leads me to my ultimate question: if the OCZ 700W is no good, which to buy? I know a fair bit about power supplies to stay away from no-name brands, but are there any stand-out awesome PSUs available for about $70-80? He is thinking about SLI down the road, so SLI-certified/ready might be nice. How does the FSB Blue Storm II 500W look? It has the beautiful blue paint job to kick. This should be able to handle SLI and a overclocked E8200, right? I was just hoping for some more wattage for future-proofing like an 8-pin PCIe. This OCZ ModXStream 600W looks pretty good, too, with modular cabling and an 6+2 PCIe connector plus a regular 6-pin.

~Ibrahim~

I don't usually quote myself, but I think a previous post I made is pertinent in this case:
Some suggestions for good modular PSUs are the Corsair HX520 and HX620, Antec NeoPower 500W, 550W, and 650W, BFG MX-550 and MX-680, Silverstone ST60F and ST70F, and Seasonic M12 500W and 600W. Be prepared to pay a premium if you want modular cables though.
 
I mean, he doesn't *need* modular cables, but cables aren't too easy to hide in the Antec 900, so I thought it might be nice.

Well, I've settled on the BFG MX-550W or the Silverstone ST60F. Both are about $80 after MIR. I'm inclined to like the Silverstone a bit better, but his GPU is BFG and they have a pretty nice RMA process.

~Ibrahim~
 
Really? Interesting...I think, ultimately, it will depend on how much money he has. For about $110 (minus MIR of $25), there is the BFG MX-550W. It has modular cabling and the 8-pin PCIe. For about $70 (with free shipping, too) is the FSP Blue Storm II. It doesn't have modular cabling nor 8-pin PCIe. Wattage difference: 500W 12V+ combined on the BFG; 480W 12V+ combined on the FSP. Pretty close...

I think we might have to forgo the modular cabling for a cheaper price. And blue cables won't look half-bad, really. The FSP is good, though, right? I mean, I've heard they are quite a good maker, but the comment on their OCZ PSU's make me a bit doubtful.

~Ibrahim~
 
Really? Interesting...I think, ultimately, it will depend on how much money he has. For about $110 (minus MIR of $25), there is the BFG MX-550W. It has modular cabling and the 8-pin PCIe. For about $70 (with free shipping, too) is the FSP Blue Storm II. It doesn't have modular cabling nor 8-pin PCIe. Wattage difference: 500W 12V+ combined on the BFG; 480W 12V+ combined on the FSP. Pretty close...

I think we might have to forgo the modular cabling for a cheaper price. And blue cables won't look half-bad, really. The FSP is good, though, right? I mean, I've heard they are quite a good maker, but the comment on their OCZ PSU's make me a bit doubtful.

~Ibrahim~

The Blue Storm II is mediocre: http://www.jonnyguru.com/modules.php?name=NDReviews&op=Story&reid=84

If your friend can't go with a modular PSU, then here are some recommendations for other fairly inexpensive units that are quality and will be able to power the PC fine:

http://www.newegg.com/Product/Product.aspx?Item=N82E16817702008
http://www.newegg.com/Product/Product.aspx?Item=N82E16817151069
http://www.newegg.com/Product/Product.aspx?Item=N82E16817371006
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139003
http://www.newegg.com/Product/Product.aspx?Item=N82E16817703005
http://www.newegg.com/Product/Product.aspx?Item=N82E16817151034
http://www.newegg.com/Product/Product.aspx?Item=N82E16817702009
http://www.newegg.com/Product/Product.aspx?Item=N82E16817103939
http://www.newegg.com/Product/Product.aspx?Item=N82E16817139005
http://www.newegg.com/Product/Product.aspx?Item=N82E16817341011
http://www.newegg.com/Product/Product.aspx?Item=N82E16817103940
 

I already linked to the TX650, and those two PCP&C units in my post above. The other ones you suggest are good, with the exception of that Rosewill.
 
Thanks for the links, guys. I really appreciate the help. :)

I had it down to the four PSU's that had a 8-pin CPU connector and at least one 8-pin PCIe:

SeaSonic 500W (392W on 12V+)
BFG 550W (500W on 12V+)
Corsair 650W (624W on 12V+)
BFG 680W (620W on 12V+)

I think it is down to the Corsair 650W and the BFG 680W. I dropped the SeaSonic model because I felt it should include some more watts on the 12V+ (392W). It definitely looked good, though. I know I may be overshooting his needs by a bit, but he is planning to overclock. We thought we had fixed the issue one day and tried overclocking. Beauty hit 3.9GHz, 55C load. Don't know how safe that was on that power supply, however, lol, but we will definitely be doing it again once this new PSU comes in. Or am I just being ridiculous and the BFG 550W will be fine?

Between the 650W and the 680W, they seem pretty similar. Any recommendations of one over the other? I mean, wattage is similar (624W on 12V+ on Corsair; 620W on 12V+ on BFG), sleeving is good on both, BFG's fan is a wee bit bigger (135mm vs 120mm)?

Only bad thing I've heard about the Corsair is that it gets a bit loud (38dB) on load.

~Ibrahim~
 
hey Zero82z next time you post if you could provide the format of links like I did it, helps a lot, because many of us know the specs of a lot equipment and it saves time instead of having to click on 7 different links to get through your post. Not picking on you but I'm pretty sure its part of the guidelines for the forum. I definitely think ikjadoon is on the right track here now.
 
I recommend ditching that OCZ GameXStream PSU. It's based off the FSP Episilon PSU design which has out of spec ripple, or voltage fluctuations, at high loads that can damage or kill your system. Granted, you may not reach such high loads but why pay for a PSU that can kill your system?

Oh and radioshack does have multi-meters IIRC. And since that PSU already has a past history of hurting hardware, get rid of it.. Especially since this happened in the above review:

If that ever happends to my system with my stealthx, ill send you 100$ bucks
 
Or am I just being ridiculous and the BFG 550W will be fine?
Take a look at the rig in my sig and consider that it's running on a 520W PSU. The 550W will be fine. They're both great PSUs though, so he may want to go with the 680W version anyway for the headroom.
Between the 650W and the 680W, they seem pretty similar. Any recommendations of one over the other? I mean, wattage is similar (624W on 12V+ on Corsair; 620W on 12V+ on BFG), sleeving is good on both, BFG's fan is a wee bit bigger (135mm vs 120mm)?

Only bad thing I've heard about the Corsair is that it gets a bit loud (38dB) on load.
They're both about equal quality. The BFG is a ForePoint while the Corsair is manufactured by Seasonic. Just get whichever is cheaper.
 
Well, this upcoming friday im goin to have around 160-200 dollars comin in, so its not going to be a big deal, im hoping fora 3 day shipping, but regardless i think i'm fine. I'm going to get the two led blue fans and i guess the psu that Ib picks out.. Black is what im prefering, but now i dont care, just as long as my problem is fixed.. I'm just a little scared it's not power supply, maybe my mobo or soemthing else that we're just not seeing. (my luck).. But everything looks and seems like it revolves around psu.. Ive had 3 different 8800gts, and i have a 9600gt i havent tried yet, but i think it'll be pointless. Are you guys sure it is psu?, i looked at the picture of the melting psu, and on mine it looks like there is some glue or liquid on a part of my psu. It looks as if it is gluing two pieces of it either in the middle ( its a yellow/greenish color) Both pieces it is connecting seem to be touching the bottom of the psu, so im not for sure if it is supposed to be there or not.. Overall, since the psu was used before i put it in, it is very dusty inside, and is kinda raggedy inside. I'm going to post a few pictures of it onto here, but i can only get a few shots of it.. Anyone have a clue??, im hoping badly its not mobo or anything else, i just hope its psu.
 
Are you guys sure it is psu?

No, but we are sure that the PSU you're using now is a ticking time bomb. Even if it doesn't turn out to be the problem, it should be replaced anyway for the safety of your other components.
 
No, but we are sure that the PSU you're using now is a ticking time bomb. Even if it doesn't turn out to be the problem, it should be replaced anyway for the safety of your other components.

Exactly what I've said twice in this thread: Replace that PSU!
 
Well, could it be the bios? To where id have to flash it back.. i havent restarted/froze since last night (12 hours) so its very random.. It can happen sometimes when i first boot the computer, and then it restarts again, so i figured its no software/virus..

BIOS Vendor Phoenix Technologies, LTD
BIOS Version 6.00 PG
BIOS Date 09/04/2008
 
Well, could it be the bios? To where id have to flash it back.. i havent restarted/froze since last night (12 hours) so its very random.. It can happen sometimes when i first boot the computer, and then it restarts again, so i figured its no software/virus..

BIOS Vendor Phoenix Technologies, LTD
BIOS Version 6.00 PG
BIOS Date 09/04/2008

I doubt it. BIOS issues rarely cause that kind of freezing and crashing.
 
That's what I was thinking...

So, we've got those three PSUs. I think we're going for the BFG 680W. I'm going to pick out some nice blue LED fans. :)

~Ibrahim~
 
btw, when i was playing wow the other day, i ran speedfans, and it showed me my 12V rail, and when i get into my bios, it says around 12.15, and when im paying somethin, or doing a lot of stuff, like playing wow, it was at 11.14.. I was at a friends house, and my comp lasted for like a day, and then i went to work.. It might have gone off by itself,but it lasted another 9 hours whenever i came back and turned it on, then i had to leave.. It seems like when i move it, it reset/crashes frequently, then gets used to the power or something. It sounds to me like psu, but im still flippin my sh*t because its taking so long
 
Software voltage monitoring isn't remotely accurate. You need to use a digital multimeter to properly examine your voltages. In any case, the low quality and poor reliability of your current PSU has already been well established.
 
Yep yer PSU is struggling, thats a large 9% drop.
The final load voltage is only just within the ATX spec but that isnt what is of significance, the drop is too large (even accounting for inaccuracy).
If further load is put on your PC or if the PSU weakens more, the drop will increase in size, that will push the PSU towards death.
You risk losing other components in your PC if the PSU dies.
 
The final load voltage is only just within the ATX spec

The ATX spec is +/- 5%, so it actually isn't within spec. However, like I said earlier, software voltage monitoring is not accurate so the actual voltages could conceivably be well within safe levels.
 
The ATX spec is +/- 5%, so it actually isn't within spec. However, like I said earlier, software voltage monitoring is not accurate so the actual voltages could conceivably be well within safe levels.

Oops yer right, its the -12V and -5V that are 10%.
So yes OP, your PSU is already way outside of its comfort zone, you should not use it any more.
 
yeah, basically. im just hoping its psu, because it seems like one thing is losing power first, then reacting. like the screen tear could be the gpu not getting the power, and restart/freeze be something else. I'm going to reformat the day i get my psu as well, just to know for sure that its not drivers.
 
I got the psu, as well, i reformatted, dled the newest (newer than the one i was using) nvidia gpu drivers, and the nforce (chipset) drivers.. still does it.. still the wierd looking crash/restart.. idk what it is now, has to be motherboard..
 
To be honest, it sounds like more of a GPU issue than a motherboard issue. However, either component could be at fault. If you have access to some other components that you could test with, it would be very helpful in narrowing down the problem.
 
Well, ive changed gpus, ive gotten 2 other 8800gts, and a9600gt that ive tried. all done the same thing.. So, i just now took out a stick or ram (2x2gigs, my windows only holds 2 anyways), and i put the one stick in another socket (black, instead of blue) and also hooked up another cord from the psu to the mobo (One didnt have one in it) and hopefully this wiill work, but my luck, says no.. So, im going to either take it to someone if this keeps happening, or RMA my hdd/mobo/cpu, because they're the only things i havent necassarily tested
 
We've tried it; ran the whole night with no errors. It was an older MemTest I had laying around, but still no errors.

I mean, we've RMA'd the GPU three times now and the last time was a totally different card. 12V+ is a more stable now, at 12.03V, though.

Is it likely to have three bad GPUs (four including the original)? The motherboard was refurbished, but the crashes/artifacts all sound GPU-like.

~Ibrahim~
 
It's highly unlikely to have three faulty video cards. It seems like it would be the motherboard in this case, although this is highly odd behaviour for a motherboard issue.
 
I'm not for sure, but i took oen of the 2g ram sticks out, and put the single one in a different slot.. Also, i hooked up another cable from the psu to the mobo. and its gone for a few hours, and hasnt done anything. I'll just have to see, but most likely im going to rma this mobo.
 
Back
Top