Strange power related issue

**selldoor** · 06-16-2012, 02:33 PM

Re: Strange power related issue

Hi I am no expert, they will be along later but as you say the lower power psus will not run your rig, have you tried it with less stuff connected, perhaps the initial load is too high.

Another thing you could try is after it has been off for the 3 hours pre heat the psu with a hairdrier and see if it comes straight on.
Then try the same thing on the motherboard.
I am thinking cap on its way out - the initial on gives it enough heat to work second time or a cold solder joint.
Is the 3 hours relative the time it is on, that is if you have it on to full power up then switch it off is the wait the same as having it on for say 4 hours then off.

Do your case mods involve power from the psu for say ccfls

**popeye** · 06-16-2012, 02:34 PM

Re: Strange power related issue

check the voltage on the cmos battery should be at least 3v then if ok check the power supply voltages are correct and also check power supply for ripple if supply ok check that everything is seated properly memory etc:.

**slamscaper** · 06-16-2012, 11:50 PM

Re: Strange power related issue

Thanks for the suggestions guys.

Regarding the CMOS battery... I was thinking about this earlier. Any time I've had a CMOS battery die, I lost my BIOS settings after disconnecting my system from the wall outlet. This hasn't happened with my current setup (yet).

Could a bad CMOS battery actually cause this? If so, it'd be the first time I've ever experienced it. I guess I could pick up another one, as they're pretty cheap.

Anyway, I'm considering buying a cheap power supply tester. They are not useful for testing a PSU under load, but the issue I'm having doesn't crop up when my machine is under load anyway. I just want to let the PSU sit a while, then check all the voltages. I could do this with a DMM (I do own one), but the PSU tester will be easier. I've been meaning to pick one up for a while, so there's no time like the present.

Anyway, keep the suggestions coming.

**selldoor** · 06-17-2012, 05:19 AM

Re: Strange power related issue

"but the issue I'm having doesn't crop up when my machine is under load anyway"

As I said I am no expert but I thought the maximum load on a psu
was when it was first turned on?

**tom66** · 06-17-2012, 05:29 AM

Re: Strange power related issue

Originally posted by selldoor

"but the issue I'm having doesn't crop up when my machine is under load anyway"

As I said I am no expert but I thought the maximum load on a psu
was when it was first turned on?

Not always... PSU load can be high on startup as disks spin up and caps charge up but it's much higher e.g. when using a graphics intensive game.

**slamscaper** · 06-17-2012, 11:45 AM

Re: Strange power related issue

Yes, tom66 is correct. I've measured the power consumption from the wall socket in the past and it's highest (by far) when I'm running GPU intensive applications like Heaven 3.0.

**slamscaper** · 06-17-2012, 02:10 PM

Re: Strange power related issue

Thankfully, I'm pretty sure I finally pin-pointed the problem.

I tore apart my PC again yesterday, just to check over EVERYTHING thoroughly. I inspected every single cap and trace on the MOBO, plus I took the PSU apart again and inspected the components (and the connectors). Nothing looked out of the ordinary. I also took some readings with a DMM and they were all within tolerance (used the paperclip trick).

Considering that, plus the fact that I don't have any issues at all when I'm up and running, I started to consider that some other outside factor may be causing this issue (one that's not directly related to my PC).

That's when I decided to take a good hard look at my surge protector, which is a cheap 7 outlet unit I bought a long while back. I disconnected everything from the surge protector and swapped it for another one I had laying around. I powered down my PC and left it for a few hours.

It fired right up with no problems, which leads me to believe it is in fact the surge protector crapping out when the PC starts up initially. It all makes sense now really.

I'm going to continue to monitor it closely of course, but I'm hoping I nipped this issue in the bud because replacing the surge protector is an easy fix.

**ben7** · 06-17-2012, 02:24 PM

Re: Strange power related issue

Good job, those cheap surge-protectors can be dangerous (undersized wiring, poor solder joints)!

**tom66** · 06-17-2012, 04:59 PM

Re: Strange power related issue

Could be a failed NTC in the surge suppressor. That's a fire hazard... I've never liked surge suppressors, they present more of a risk than worth it in my opinion. Seen one blown up quite badly from a surge, to be honest I'd rather it just killed the PSU and didn't start a fire... And I've never had a problem not using one.

**slamscaper** · 06-17-2012, 07:20 PM

Re: Strange power related issue

I spoke too soon, lol. I have been trying to reproduce the problem all day and I finally did a couple more times (using a different surge protector).

I can't explain why the system fired right up that first time, because it hasn't done that in a good while.

Anyway, there goes that theory... I was really confident I found the culprit too, but now I'm back to square one.

I'm starting to consider that the MOBO is the problem. If the PSU was on its way out, I would expect to have issues when my system is under load, which clearly I do not.

It's like the motherboard is simply failing to transfer power to everything when the button is pressed, although its PLED is lit like everything is fine and dandy.

I really do hate vague issues like this...

**severach** · 06-17-2012, 09:45 PM

Re: Strange power related issue

Originally posted by tom66

Could be a failed NTC in the surge suppressor. That's a fire hazard... I've never liked surge suppressors, they present more of a risk than worth it in my opinion.

Not only a fire hazard but they also fail to do what you think you want them to: stop lightning strikes. I only buy power taps and when I get a free surge strip I like I cut out the surge circuitry. My lightning protection is a long ways upstream of the power taps.

**slamscaper** · 06-18-2012, 10:22 AM

Re: Strange power related issue

Figured I'd post some pics of my PSU. Maybe you guys will see something I don't. I'm really perplexed by this issue I'm having.

I should note that my system will start up fine if I cut the power before pressing the button. This is true no matter how long the system is powered down. Very strange.

**budm** · 06-18-2012, 11:11 AM

Re: Strange power related issue

So it is not the surge protector outlet strip that is causing the problem?

**selldoor** · 06-18-2012, 12:20 PM

Re: Strange power related issue

Hi well despite you deciding that Tom was correct in saying greatest load was during playing a high speed graphic sequence game even though you are not playing said game but are in fact starting up when the problem occurs, I will chip in a further comment.
Did you try the preheating the caps in the PSU as I suggested? - Well my reasoning was that when the caps are hot they work ok but when they are cool the esr will be higher. Add that to the fact that at least one of the big caps is a FUHJYYU (Rubbish) (Probably both are but can only see it on one) and that they are 85deg caps in a cap cooking situation they have probably simply degraded to a point where they dont work properly. I dont know why the switching off and back on makes a difference - perhaps they get a bit more of a surge that
kick starts them. Type FUHJYYU into the search above and read about them.

**slamscaper** · 06-18-2012, 01:31 PM

Re: Strange power related issue

Originally posted by budm

So it is not the surge protector outlet strip that is causing the problem?

Sadly no. I spoke too soon earlier, because the problem reared its ugly head again, even after swapping the surge protector for another one. I can't explain why it worked the the first go round, but I chalk that up to a coincidence.

Originally posted by selldoor

Hi well despite you deciding that Tom was correct in saying greatest load was during playing a high speed graphic sequence game even though you are not playing said game but are in fact starting up when the problem occurs, I will chip in a further comment.
Did you try the preheating the caps in the PSU as I suggested? - Well my reasoning was that when the caps are hot they work ok but when they are cool the esr will be higher. Add that to the fact that at least one of the big caps is a FUHJYYU (Rubbish) (Probably both are but can only see it on one) and that they are 85deg caps in a cap cooking situation they have probably simply degraded to a point where they dont work properly. I dont know why the switching off and back on makes a difference - perhaps they get a bit more of a surge that
kick starts them. Type FUHJYYU into the search above and read about them.

Are you talking about the big caps? Those are HP3 (Hitachi) made I believe, or is that the same difference? I would assume Hitachi caps (made in Japan obviously) should be of a decent quality. This is a $330 PSU, so Thermaltake should be crucified if they used low quality caps in it. I researched this PSU before dropping the cash on it (read multiple reviews), but honestly I did not pay attention to the caps at all. I assumed it used high-quality components all around. I'll be upset to hear otherwise...

I'll try heating up the PSU before firing up my system and see how that goes.

**selldoor** · 06-18-2012, 03:30 PM

Re: Strange power related issue

Blast! i am wrong again- still an honest mistake- I attach a pic showing the logo on your cap which as you say is hitachi and the logo for FU*
I think it is the same but in negative Ill have to get some new specs!!
Sorry thought I had something there.
Anyway the Cooked caps bit may still prove correct.

Attached Files

**tom66** · 06-18-2012, 04:04 PM

Re: Strange power related issue

Definitely a Hitachi cap. That "S" shaped vent is unique to them.

Hitachi make very high quality primary side capacitors. Similar caps are used in industrial three phase motor inverters. So they should last. Besides, the most common failure when that big cap fails is it kills the PFC circuitry because the energy can't be dumped quick enough from the inductor into the cap, fries the MOSFET and sometimes the diode too.

The PSU looks pretty well made (good caps and good component choices), might be a CWT from the green transformers? Not sure...

I would start suspecting the motherboard.

**slamscaper** · 06-18-2012, 07:14 PM

Re: Strange power related issue

Originally posted by tom66

Definitely a Hitachi cap. That "S" shaped vent is unique to them.

Hitachi make very high quality primary side capacitors. Similar caps are used in industrial three phase motor inverters. So they should last. Besides, the most common failure when that big cap fails is it kills the PFC circuitry because the energy can't be dumped quick enough from the inductor into the cap, fries the MOSFET and sometimes the diode too.

The PSU looks pretty well made (good caps and good component choices), might be a CWT from the green transformers? Not sure...

I would start suspecting the motherboard.

It seems you know what you're talking about.

Up until now I've been scrutinizing the PSU, but today I turned my attention to the motherboard. As I said earlier, I've been suspecting the MOBO ever since I inspected the PSU thoroughly.

I decided to rule out the PSU by installing it into another PC so I could test it with another MOBO. My other PC is housed in a CM storm scout (it's just a basic PC), so it was pretty funny seeing this huge PSU in such a little case. Anyway, the system ran fine and didn't have any issues powering on. I left the system on standby for a good while, but it started right up every time. So... The PSU seems to be functioning perfectly.

It actually does make sense to me. This MOBO isn't built to the same standard that the PSU is. It's EVGA branded, but FOXCONN made.

Not sure how I'm going to handle this. I've been wanting to upgrade to X79 for a while, however I'm unemployed now so I have to watch how much I spend. I planned on building a system around this board and selling it to a friend in the future, so maybe I'll send it in to EVGA for replacement. That way I'll earn a little dough towards my new X79 build.

**ben7** · 06-19-2012, 05:02 AM

Re: Strange power related issue

You cant switch this PSU out for another one, but you can put it in a lower-end PC, just to see if it has the same startup problem.

Strange power related issue

Strange power related issue

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Comment

Related Topics