• Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
    Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.

Pwer Supply Going?

M

Monk0101

Weaksauce
Joined
Oct 1, 2005
Messages
93
Hi,
I have a Seasonic X-1250 power supply in my rig. It is about 3 years old. In the last month I have been having cold boot issues. When I press the start button on the case (either right after shutting down or after the computer has been shut down over night), absolutely nothing happens, no fans, no case lights, nothing. If I either unplug the line cord or flip the switch on the power supply and wait a while (half hour or more), then plug in the line cord or flip the power switch on, then press the case start button, the computer starts normally.

My first guess is Power Supply (second would be Motherboard). Any ideas?

If I assume the problem is the Power Supply, I wonder how much of a problem I am going to have convincing Seasonic that the Power Supply is bad. After all, it works if I leave the computer unplugged or manually switched off for a while.

Thanks for any input.

Bob
 
Last edited:
I was pondering this for a while... have you tried unplugging/removing everything from the MoBo that isn't required to run (like your SSDs, HDDs, GPU, ODD, 7 out of the 8 RAM modules, etc) and see if that makes any difference?

May help determine if the PSU is bad with what you are normally loading it with vs a lighter "barebones" load.
 
Last edited:
What voltages are indicated on the 5V and 12V rails at idle and under heavy load?
If you have a multimeter to check them, even better.
 
I had a similar issue. Put a new PSU in and it worked fine.
 
Monk0101 said:
Hi,
I have a Seasonic X-1250 power supply in my rig. It is about 3 years old. In the last month I have been having cold boot issues. When I press the start button on the case (either right after shutting down or after the computer has been shut down over night),
Click to expand...

Turning things on and off is stressful on your components. A lot of things (PSUs, HDDs, etc.) will last years if left on and then one reboot and they are dead. Seasonic is a pretty good brand, but if you are shutting it down every night - it could possibly shorten its life. But as with all things YMMV.
 
Monk0101 said:
Hi,
I have a Seasonic X-1250 power supply in my rig. It is about 3 years old. In the last month I have been having cold boot issues. When I press the start button on the case (either right after shutting down or after the computer has been shut down over night), absolutely nothing happens, no fans, no case lights, nothing. If I either unplug the line cord or flip the switch on the power supply and wait a while (half hour or more), then plug in the line cord or flip the power switch on, then press the case start button, the computer starts normally.

My first guess is Power Supply (second would be Motherboard). Any ideas?

If I assume the problem is the Power Supply, I wonder how much of a problem I am going to have convincing Seasonic that the Power Supply is bad. After all, it works if I leave the computer unplugged or manually switched off for a while.

Thanks for any input.

Bob
Click to expand...

I would jump the switch with a screwdriver (on the mobo) to eliminate the possibility of a bad case power switch.
 
MySongRanHills said:
Turning things on and off is stressful on your components. A lot of things (PSUs, HDDs, etc.) will last years if left on and then one reboot and they are dead. Seasonic is a pretty good brand, but if you are shutting it down every night - it could possibly shorten its life. But as with all things YMMV.
Click to expand...

I have never heard or seen any evidence that shutting down a power supply shortens its life. I may be wrong here, but I just have never heard it.
 
I have the same issue on one of my rigs right now. Though it's not the PSU. I've tried multiple high power, high quality gold PSUs, different power switches, ram, new bios, etc and it still happens. Doing a hard reset on the mobo is the only thing that makes it boot again. For my situation it appears to be a BIOS setting or a mobo issue. All settings being the same for a while, the issue is new. Haven't taken the time to fool around with it further to figure it out, since I just leave it on all the time.
 
Rav3n said:
I have never heard or seen any evidence that shutting down a power supply shortens its life. I may be wrong here, but I just have never heard it.
Click to expand...
Tbh I haven't seen any testing proving this to be true, but it stands to reason that cycling on and off is the most stressful time for the PSU as it is a sudden surge vs steady usage.

I could be wrong, but with HDDs i have observed this first hand - drives working perfectly for a very long time and die upon reboot
 
Rav3n said:
I have never heard or seen any evidence that shutting down a power supply shortens its life. I may be wrong here, but I just have never heard it.
Click to expand...

The theory is that metal expands and contracts with temperature fluctuations. The more those components move (expand and contract) the more likely something is to come loose or break. Since shutting it down will mean the opponents will go to room temperature and powering it on will heat it up it's believed that leaving it running will keep the overall temperate closer to the average temperature thus eliminating the extreme fluctuations allowing your components to last longer.

I do not think there has ever been any real testing to prove this however it does seem very plausible.
 
MySongRanHills said:
Tbh I haven't seen any testing proving this to be true, but it stands to reason that cycling on and off is the most stressful time for the PSU as it is a sudden surge vs steady usage.

I could be wrong, but with HDDs i have observed this first hand - drives working perfectly for a very long time and die upon reboot
Click to expand...

The only reason that happens is because the parts that move have worn to be way out of tolerance limits.

So when the drive shuts down, it parks the head and also stops spinning.

Sometimes the head(s) will not make it back onto the platter properly, and sometimes the drive is not able to spin up anymore.

In either case, a drive like that is a ticking timebomb.
 
MySongRanHills said:
Tbh I haven't seen any testing proving this to be true, but it stands to reason that cycling on and off is the most stressful time for the PSU as it is a sudden surge vs steady usage.

I could be wrong, but with HDDs i have observed this first hand - drives working perfectly for a very long time and die upon reboot
Click to expand...

By that same token, not having to filter and regulate power for lengths of time may in fact be better for the unit, and outweigh the risks of power cycling. Like I said in the 20+ years I have built and repaired computers, I have never heard of power cycling being an issue.
 
Skillz said:
The theory is that metal expands and contracts with temperature fluctuations. The more those components move (expand and contract) the more likely something is to come loose or break. Since shutting it down will mean the components will go to room temperature and powering it on will heat it up it's believed that leaving it running will keep the overall temperate closer to the average temperature thus eliminating the extreme fluctuations allowing your components to last longer.

I do not think there has ever been any real testing to prove this however it does seem very plausible.
Click to expand...

FTFY. Like I said above, not having to filter and regulate a naturally dirty power source for lengths of time most likely benefit the unit. Now if you are talking a strictly controlled pre cleaned power source in a temperature controlled environment, I could see the point.

Even labs don't have power that is that clean. Dirty power and brown power kills PSU's, not random breakage of internal components due to age. I can't say I have ever seen something inside a PSU that was broken do to suspected age or power cycling. And I have seen hundreds if not thousands of PSU's over the years. Just food for thought.
 
Nenu said:
What voltages are indicated on the 5V and 12V rails at idle and under heavy load?
If you have a multimeter to check them, even better.
Click to expand...

Thanks for the response,

I checked with a DVM and the Idle Voltages on the 24 pin power connector to the motherboard are 12.13 and 5.04. The voltages while running Prime95 and the CPU at 100% are 12.11 and 5.04.

One thing I did notice this afternoon when I was messing around with the DVM, was that on the 12v rail, it takes forever to reach 0.0 volts after shutting down. After shutting down and unplugging from the AC outlet and disconnecting anything outside the case (monitors, G19 Keyboard/Mouse) it still took over an hour to reach 0.1 mv. Granted that isn't much, but it still surprised me. Even then the computer would not cold boot. Still no fans, not leds, nothing. The next time I tried to boot was about 4 hours later and the system booted with no issues. But, I'll bet that if I shut down right now, I won't be able to boot until sometime in the wee hours of the morning.

There is a power switch on the MB which mimics the case switch in case you want to start the system while you are "under the hood" - it does not work any better than the case switch. I understand that both switches are a momentary contact switch that when press tells the MB to tell the PS to go to work. If that is right, then I suppose I could have a MB issue and the MB is not telling the PS to start up.

Also, I do not experience any crashes or other abnormalities. Only this warm Cold Boot issue.

Any more ideas?

Thanks,
Bob
 
DejaWiz said:
I was pondering this for a while... have you tried unplugging/removing everything from the MoBo that isn't required to run (like your SSDs, HDDs, GPU, ODD, 7 out of the 8 RAM modules, etc) and see if that makes any difference?

May help determine if the PSU is bad with what you are normally loading it with vs a lighter "barebones" load.
Click to expand...

Thanks for the idea, I'll give that a try over the weekend.
 
Monk0101 said:
Thanks for the response,

I checked with a DVM and the Idle Voltages on the 24 pin power connector to the motherboard are 12.13 and 5.04. The voltages while running Prime95 and the CPU at 100% are 12.11 and 5.04.
Click to expand...
Thats very good.

One thing I did notice this afternoon when I was messing around with the DVM, was that on the 12v rail, it takes forever to reach 0.0 volts after shutting down. After shutting down and unplugging from the AC outlet and disconnecting anything outside the case (monitors, G19 Keyboard/Mouse) it still took over an hour to reach 0.1 mv. Granted that isn't much, but it still surprised me. Even then the computer would not cold boot. Still no fans, not leds, nothing. The next time I tried to boot was about 4 hours later and the system booted with no issues. But, I'll bet that if I shut down right now, I won't be able to boot until sometime in the wee hours of the morning.
Click to expand...
This is expected behaviour of a good PSU.
The power rails are supported by capacitors to maintain capacity and cope with transients.
These take time to discharge when mains power is removed.

There is a power switch on the MB which mimics the case switch in case you want to start the system while you are "under the hood" - it does not work any better than the case switch. I understand that both switches are a momentary contact switch that when press tells the MB to tell the PS to go to work. If that is right, then I suppose I could have a MB issue and the MB is not telling the PS to start up.

Also, I do not experience any crashes or other abnormalities. Only this warm Cold Boot issue.

Any more ideas?

Thanks,
Bob
Click to expand...
Given the above, this suggests there is a bad setting or incompatibility, rather than an outright fault.
although it doesnt rule out a fault.

Can you reset your cmos?
 
Monk0101 said:
I had not thought of a simple cmos reset. I'll give it a shot.

Thanks
Click to expand...

Unfortunately, reseting the cmos did not help. I did the reset with the jumper on the MB and after that did not help, I reset with the jumper and removing the cmos battery. Again no luck.

It does not seem to matter if the system is on for hours or a few minutes. If I do the shut down thru Windows, the system will not boot back up until it has been setting unplugged from the ac line for a while (1+ hour or so).

I have not made any hardware or software (other than normal windows updates) leading up to this issue starting.

Any other thoughts?

Thanks for the assistance.
 
Check your memory is properly set up in the BIOS according to manufacturer specs.
ie voltage, speed.
Try different memory profiles.
 
It could be one of those weird things where your only choice would to replace the suspected faulty components.
 
Nenu
The memory settings look right.

Tsumi:
You may be right (unfortunately).

I think I'll strip everything down over the weekend and see what happens with a bare bones system. Then it looks like a tossup between motherboard and power supply. Guess I'll flip a coin and see what happens.

Thanks everyone for the help!

Bob
 
Good luck, Bob!

If I still had a spare PSU, I'd let you borrow it so you could at least eliminate that variable.
 
I need so advice from someone a lot smarter that I am.

After doing some reading (not one comment Mick), I have an idea. Since the issue I am having is the PS not turning on when not completely drained, I should be able to isolate the issue with either the PS or MB by doing the following, with the PS plugged in and switched on and the MB still connected to the PS 24 pin power cable.

1. Measure the voltage with a DVM across the PSON# pin on the power connector to ground. Then press the case or MB power on button. The voltage should drop to 0.0. If not, the issue is with the MB.

2. I should be able to jumper the PSON# pin on the power connector to ground and the PS should start up the MB. If not, the issue is with the PS.

Does this make sense to anyone?

Thanks,
Bob
 
Monk0101 said:
I need so advice from someone a lot smarter that I am.

After doing some reading (not one comment Mick), I have an idea. Since the issue I am having is the PS not turning on when not completely drained, I should be able to isolate the issue with either the PS or MB by doing the following, with the PS plugged in and switched on and the MB still connected to the PS 24 pin power cable.

1. Measure the voltage with a DVM across the PSON# pin on the power connector to ground. Then press the case or MB power on button. The voltage should drop to 0.0. If not, the issue is with the MB.

2. I should be able to jumper the PSON# pin on the power connector to ground and the PS should start up the MB. If not, the issue is with the PS.

Does this make sense to anyone?

Thanks,
Bob
Click to expand...

Since your MoBo has on-board power and reset, have you tried pulling the chassis switch leads off their headers and just tried the on-board switched? Just another variable to eliminate...
 
DejaWiz said:
Since your MoBo has on-board power and reset, have you tried pulling the chassis switch leads off their headers and just tried the on-board switched? Just another variable to eliminate...
Click to expand...

Yup, tried that this morning. No luck. Thanks though!
 
Good news / Bad new thing.

I tore the system down and cleaned everything up. I was still having the same symptoms. So I went to the closet and drug out an old PC Power and Cooling 510 watt Power Supply.

I just hooked up the 24 pin connector to the Motherboard and a 4 pin to the EATX12V connector on the Motherboard. When I turn on the "new" Power Supply, the Q-Code LED started counting up like normal. I was able to turn off the Power Supply for a couple seconds and turn it on again and get the system to try and boot. I was able to do this repeatedly. So, I am assuming I have narrowed the problem to the Power Supply.

I know the 510 watt Power Supply is not enough to really run the entire system. So, I'll have to wait until tomorrow when the Mailman (Letter Carrier if you want to be politically correct) bring an EVGA Supernova 1000 G2 that I ordered yesterday (just in case).

Now all I have to do is convince Seasonic that the Power Supply is bad. After all, it does start up once. It is getting it to re-start 5 minutes to 8 hours later that I have a issue with. Since I don't know if Seasonic will honor the warranty or how long the turn around will be, I at least will have a new Power Supply to keep me going.

Thanks all for your help! I'll post how I get by with Seasonic.

Bob
 
Monk0101 said:
Mailman (Letter Carrier if you want to be politically correct).
Click to expand...

"Packaged Correspondence Transport and Delivery Technician" :p


...glad you got her figured out, Bob. Post back when your system goes Supernova to let us know how everything is holding up.
 
I received the EVGA Supernova 1000 G2 and installed it last night. Everything is back to normal. Since both cold boot issue is gone with both the 510 watt PC Power & Cooling or the EVGA power supplies, I am certain the Seasonic power supply was bad (or at the very least going bad).

Now, I'll wait and see if Seasonic reaches the same conclusion and honors their warranty.
 
If your sig is correct, that PCP&C unit should have been fine with your load.
 
Jorona said:
If your sig is correct, that PCP&C unit should have been fine with your load.
Click to expand...

My sig is correct as of this moment. I have a habit of switching things around and went overkill on the PSU so that I don't have to be concerned about it when I add/change other hardware. ;)

While I am at it, I did notice one other thing while swapping PSU's. On my MB there are several LEDs that light up when the system is powered up. With the Seasonic running things, these LEDs would take a long time (several minutes or more) to go dark after switching off or unplugging the AC cord on the PSU. With the other two PSU's, they went dark after a couple seconds. In my simple mind this would indicate that the Seasonic had some serious hefty capacitors which held a larger charge longer, or there was some error in the PSU which was allowing voltage stored in the capacitors to discharge to the MB after AC was shut off.

I am going back to the dark ages when I used to mess with radio transmission PSU's. If I recall correctly, the PSU had bleeder resistors to provide a minimum load to the PSU and also bleed off any residual charge in the capacitors when the input voltage was removed. I suppose if this were still the case and these resistors went bad, it would take longer for the voltage to bleed off. During this time, the logic on the MB or PSU would see even a couple of MV of voltage and could think the system was still on and therefore ignore any attempt to start the system, thus causing the very issue I was seeing. Keep in mind that my experience was a long time ago (just as vacuum tubes were phasing out) and also I may not be recalling things correctly. :eek:
 
Monk0101 said:
Good news / Bad new thing.

I tore the system down and cleaned everything up. I was still having the same symptoms. So I went to the closet and drug out an old PC Power and Cooling 510 watt Power Supply.

I just hooked up the 24 pin connector to the Motherboard and a 4 pin to the EATX12V connector on the Motherboard. When I turn on the "new" Power Supply, the Q-Code LED started counting up like normal. I was able to turn off the Power Supply for a couple seconds and turn it on again and get the system to try and boot. I was able to do this repeatedly. So, I am assuming I have narrowed the problem to the Power Supply.

I know the 510 watt Power Supply is not enough to really run the entire system. So, I'll have to wait until tomorrow when the Mailman (Letter Carrier if you want to be politically correct) bring an EVGA Supernova 1000 G2 that I ordered yesterday (just in case).

Now all I have to do is convince Seasonic that the Power Supply is bad. After all, it does start up once. It is getting it to re-start 5 minutes to 8 hours later that I have a issue with. Since I don't know if Seasonic will honor the warranty or how long the turn around will be, I at least will have a new Power Supply to keep me going.

Thanks all for your help! I'll post how I get by with Seasonic.

Bob
Click to expand...

Update:
The EVGA Supernova 1000 G2 has been working fine for a few days now. So, I take that as confirmation that the Seasonic was indeed faulty.

Regarding Seasonic, I sent my RMA Seasonic X-1250 last week. They got it on Thursday 1/22/2015. I received a FedEx Shipment Notification on Friday 1/23/2015 saying they would be bringing a package to me from the Shipping Dept in Azusa CA on 1/28/2015(Azusa is where I sent the RMA to) . I have no idea at this point if FedEx is bringing back my old PSU because Seasonic could not find anything wrong or if FedEx is bringing a replacement. I have received no email or communication of any kind from Seasonic. The package should be here on Wednesday. I guess Wednesday will be like Christmas morning all over again. You open the package and see what you got! :)

Bob
 
Glad to hear your system is back to it's old self. Hope there is some good news from Seasonic when it gets delivered. :)
 
You must log in or register to reply here.
Back
Top