• Welcome to TechPowerUp Forums, Guest! Please check out our forum guidelines for info related to our community.

weird PC behaviour

With the exact same "weird" symptoms? No. The odds of that happening is extremely rare. The odds it happening at the same point in time (when the PSU works fine otherwise) is near impossibly rare. I am not saying two supplies can't fail completely at the same time, but that is not what's happening here. The supplies work fine once the boot finally happens.

The power-on behavior after a power outage is set in the BIOS Setup menu. These are not PSU options. So if this is a setting issue, it will be in the BIOS. If it is a fault somewhere, it did not suddenly develop in two different PSUs.
I didn't say the PSUs failed. Failed and not doing something other PSUs can isn't a failure :).

The point I tried to make was that cheap crappy PSUs don't have all the protections quality PSUs have. Because of this, from an outage, it is possible that a spike or something went through and keeps resetting the BIOS. By default, I believe a PC will NOT turn on after power off. I wouldn't bet my life on it , but have seen this before. If its a quality PSU that isn't nearly as likely for it to wonk up a board... hence why I asked about the brand and model of the second one.

I do believe it is the board at this point though.

Really just waiting to try and see if the slot is wonky, perhaps flashing the latest BIOS again is in order, or even breadboarding. There are multiple, efficient ways to skin this cat. But hey, let's just overwhelm the dude and bicker.

GL OP, you are well taken care of here. :)
 
That seems like a reasonable troubleshooting approach to me as this would test both the graphics solution and motherboard.

only way to test without a FI PCI/PCIE Card lol
 
Tell me does this happens if you put gpu in white pci-e 16x slot?
my video card is in white pci-express slot

Have you tried to use the other pciex slot just to check
yes,same result

For giggles, what is the brand and model of the second PSU you tried?
Chieftec 750w,same result

yes maybe I wrong when i choose this psu but is 80+ gold gaming and I thaught is good.Full review here but is in another lang.
https://www.mygarage.ro/teste-si-re...-inter-tech-argus-rgb-750cm-80-gold-750w.html
 
Last edited:
Chieftec 750w,same result
Chieftec is also known for starting fires more so than powering up a PC......... :(

Again, 80 Plus means little when it comes to quality. Whoever reviewed that thing didn't test it thoroughly. They tested for voltage drop under a gaming load using software/OCCT. Cracker Jack box testing. No ripple tests, noise, transient response...oscilliscope, hot box, etc.

Anyway, if you haven't flashed the BIOS again (for giggles) try that and see (along with breadbaording the system). Otherwise, I would look into purchasing a new motherboard to solve the issue and a new power supply at the same time.
 
Last edited:
motherboard can be repaired?is worth it?
 
The point I tried to make was that cheap crappy PSUs don't have all the protections quality PSUs have. Because of this, from an outage, it is possible that a spike or something went through and keeps resetting the BIOS. By default, I believe a PC will NOT turn on after power off. I wouldn't bet my life on it , but have seen this before. If its a quality PSU that isn't nearly as likely for it to wonk up a board... hence why I asked about the brand and model of the second one.
And that is a valid point. But even with two cheap-quality PSUs, it would be extremely rare for them to both cause the computer to exhibit the same very odd behavior (and that's an important distinction in this case) the OP described. That's the point I was making.

It is right to question the PSU - no disagreement there. But the OP tried two PSUs already and got the exact same very odd behavior. That rules out a "PSU" problem, IMO.

But does it rule out a "power" problem? NO!!! There could still be a problem with how power is being regulated, divided, or distributed by the mobo (or graphics card).
motherboard can be repaired?is worth it?
Usually not unless the faulty component is clearly visible. Otherwise, a tech has to troubleshoot and isolate the problem down to the component level and that is typically the most difficult (and time consuming) part of troubleshooting. It is easy to replace a component. Finding which component to replace can be a real challenge - and that assumes a replacement part is available.

Once you start looking at a couple hours of labor (plus parts) it does not take long for it to no longer be cost-effective to repair, even if repair is possible.
 
It is right to question the PSU - no disagreement there. But the OP tried two PSUs already and got the exact same very odd behavior. That rules out a "PSU" problem, IMO.
Before you posted this, did you see what the other PSU brand was he tested with? Chieftec...my kids could build a cheiftec unit :p. Being serious, it is crap too, and could still be a problem in the same light I spoke of... or as you have with regulation, ripple, etc... I do think it's the board, but regardless either PSU shouldn't really be used and replaced much sooner than later, regardless if it is the source of the problem and especially if he gets a new board.

I wouldn't be surprised if his original one (or God only knows what dud he was using before this) was sending dirty power to the board for however long he's had it and borked it.
motherboard can be repaired?is worth it?
That old?? I would think a new board is cheaper and less timing consuming to order than trying to repair something we aren't 100% sure (but likely) is at fault.
 
What was the last change you made before you started having issues?
did you install the new battery correctly? did you try another battery yet?
did you check the clear CMOS jumper? sometimes it can fall off if bumped
 
Before you posted this, did you see what the other PSU brand was he tested with?
Yes I saw and it does not matter. :( You are still suggest two totally different PSUs from different makers are causing the exact same very odd behavior. :kookoo:

And just because Chieftec is not a reputable brand, that IN NO WAY suggests it cannot do its job. The world is full of PCs that have been chugging along for many many years with cheap PSUs from Deer, Chieftec, and no-name generics.
I do think it's the board, but regardless either PSU shouldn't really be used and replaced much sooner than later, regardless if it is the source of the problem and especially if he gets a new board.
Okay, but that's not helping the OP with this problem, is it?
 
I would flash the bios, sounds like it's the cause.

PSU is junk too, just not the issue here. Look to upgrade it before it becomes one.
 
He said in his opening post the battery is new. That resets the BIOS too.

Not always, sometimes the battery has to be inserted then jumper cap shorted
 
Not always, sometimes the battery has to be inserted then jumper cap shorted
Its not a "cap". You use a jumper to momentarily short the two pins which dumps the battery's voltage to ground - thus removing that voltage from the CMOS device. If he was really really fast and took less than a couple seconds to replace the battery, then, maybe, the data may be retained. But that's not likely. Motherboard engineers and designers purposely selected CMOS memory devices for that data so user-settings in the BIOS could easily and quickly be reset.

CMOS memory devices are unique among memory devices in that they need a "holding" voltage to maintain the data they are holding - hence the battery. Remove the voltage and the data within is almost instantly (within a couple clock cycles) dumped. And there are no storage capacitors (or caps) in that circuit - again that is on purpose because the engineers and designers want those settings to be easily reset. So either way, pull the battery, or short the pins to dump the voltage to ground, the data is almost instantly lost when the voltage is removed. There is no need to do both, pull the battery and move the jumper.

It is a common mistake to think it is necessary. Not sure where that "myth" started but that is exactly what it is. I have even seen some take it steps further and say the batteries have to be left out for hours or even over night. Or that once you remove the battery you have to press and hold down the front panel power button on the PC for 2 minutes. All myths.

If the designers wanted it to be hard to reset those user settings, they would have used a different data storage memory device, like an EEPROM.
 
Its not a "cap". You use a jumper to momentarily short the two pins which dumps the battery's voltage to ground - thus removing that voltage from the CMOS device. If he was really really fast and took less than a couple seconds to replace the battery, then, maybe, the data may be retained. But that's not likely. Motherboard engineers and designers purposely selected CMOS memory devices for that data so user-settings in the BIOS could easily and quickly be reset.

CMOS memory devices are unique among memory devices in that they need a "holding" voltage to maintain the data they are holding - hence the battery. Remove the voltage and the data within is almost instantly (within a couple clock cycles) dumped. And there are no storage capacitors (or caps) in that circuit - again that is on purpose because the engineers and designers want those settings to be easily reset. So either way, pull the battery, or short the pins to dump the voltage to ground, the data is almost instantly lost when the voltage is removed. There is no need to do both, pull the battery and move the jumper.

It is a common mistake to think it is necessary. Not sure where that "myth" started but that is exactly what it is. I have even seen some take it steps further and say the batteries have to be left out for hours or even over night. Or that once you remove the battery you have to press and hold down the front panel power button on the PC for 2 minutes. All myths.

If the designers wanted it to be hard to reset those user settings, they would have used a different data storage memory device, like an EEPROM.

Ive always called them caps because besides shorting 2 pins they cover them
 
He said in his opening post the battery is new. That resets the BIOS too.

Yes, but that doesn't fix the issue. I said flash not clear.

I think the bios has some kind of corruption and needs reflashing.

I think the bios is shitting the bed and enabling igp, ignoring the addin card and causing the issue.
 
gordon93 when you try to power on, do any leds lit up or is it just dead like no power at all?
Did you try weaker PSU?
 
gordon93 when you try to power on, do any leds lit up or is it just dead like no power at all?
Did you try weaker PSU?
the funny things is, led lit on motherboard is green so its powered, but not start when I press power buton.Only with vga card removed first...
I searched carefully for damaged vrm and capacitors on entire motherboard,but all capacitors are fine...very strange
 
If you haven't breadboard it yet, do so... otherwise buy a new mobo or sell all other parts and upgrade. :)
 
The OP would be well advised to disconnect the power before clearing the CMOS or changing the battery. At least, that's what I always do.
 
I don t want to open another thread,so I decide to ask you all here.
I decided to buy UPS because in the area where I live take the current very often. I'm interested in a UPS that keeps me a back-up of 20 minutes in gaming to a source of 750w + 62w screen. The system I have is: 2600k i7 processor
-video r9 fury x
-3 hdd + 1 ssd
I suspect I need ups from 1500VA ? Wait for your suggestions. Thank you
 
Ive always called them caps because besides shorting 2 pins they cover them
Ah, I've never heard that but I see how you came to that. I think they actually are called "jumper blocks" because they look like little toy blocks. But I note most do not actually cover the pins - that is, they leave the tops of the pins exposed so you can put a meter probe on them, as seen here.
the funny things is, led lit on motherboard is green so its powered, but not start when I press power buton.Only with vga card removed first...
I searched carefully for damaged vrm and capacitors on entire motherboard,but all capacitors are fine...very strange
Remember, there are 3 main voltages supplied by the power supply, then the motherboard has divider circuits to create even more. The green LED may just be to alert you to the presence of 1 of those voltages - my guess would be the +5Vsb. Or it may be to give you a status report, or it may be to remind you power is present and to, as Splinterdog correctly suggests, remove all power before doing maintenance inside (like changing the CMOS battery or removing RAM). Your motherboard manual should tell you what that green LED means.

While a visual inspection of the capacitors to see if any are leaking or bulging is good, not seeing anything abnormal does not mean all is good.

I agree with Earthdog. It may be time to do a bench test (or breadboard test). That is, removal all from inside the computer case and assemble it on a breadboard to see if it works fine there.

@gordon93 - you said in your opening post that the BIOS was current. Did you recently flash it to this current version? If so, were you having these problems before flashing?
 
I don t want to open another thread,so I decide to ask you all here.
I decided to buy UPS because in the area where I live take the current very often. I'm interested in a UPS that keeps me a back-up of 20 minutes in gaming to a source of 750w + 62w screen. The system I have is: 2600k i7 processor
-video r9 fury x
-3 hdd + 1 ssd
I suspect I need ups from 1500VA ? Wait for your suggestions. Thank you
 
Back
Top