Question PC Power cycles itself, replaced CPU & PSU

[Markerater]

I recently built a computer for my SO and I was originally having a probpem where the CPU would overheat. After reseating and then switching to water cooling the issue still wasn't resolved so I replaced the CPU and it no longer overheated, however the PC would still shut off and turn back on at random intervals if under load. I verified it's not overheating, all cores sitting around ~70 under full load for a few minutes until it resets. I swapped out the PSU and all was well for a little bit.. but it just started back up again. I've clean installed Windows, disabled Intel boost in bios, changed power settings, reinstalled all drivers. This is driving me crazy, anybody know anything about this???

Specs:
Mobo: Gigabyte q270m-D3H
CPU: i7 7700k
RAM: Corsair Vengeance LPX 16 GB ddr4 2400mhz
PSU: Corsair 750 watt 80+ bronze modular
Cooler: Corsair AIO
GPU: GeForce RTX 2070 Super

 

[Darkbreeze]

That board has a VERY weak VRM and power delivery configuration. It should not even be used with that CPU. I'd highly recommend replacing it with an H270 or Z270 board that is at least relatively middle of the road. I'd suspect that your problems are very likely to stem from that.

That board has a horrible 3+2 power phase design and while it does technically support that CPU, there were plenty of motherboards that technically showed support for CPUs that they should never have been used with. And that's assuming that nothing is wrong with the motherboard, which there well could be, due to there being something obviously wrong with the previous CPU that was being used in it. Bad CPU, motherboard or memory can often mean one or more of the other two fails as well OR was the culprit to begin with.

To compound the problem, you have a high end CPU AND a high end graphics card, in a low end business chipset motherboard. It's really not surprising that you have problems, at all.

 

[Markerater]

Okay, I'll try one of the boards you suggested. Thanks for the speedy reply, I'll post am update when it arrives

 

[Darkbreeze]

Getting difficult to find 7th gen motherboards, and they are more expensive than they are worth because they haven't been manufactured in several years. It's unfortunate that you've already bought into this platform since it is already several years outdated.

It would be money better spent, for the purchases you already made and for another motherboard, had it gone towards something much newer.

What country are you in?

I'd probably recommend going with one of these boards at this late stage of the game, as any of them SHOULD be good enough for that CPU. The Gaming K4 would obviously be the better, but more expensive choice, of these three boards, but you need to be sure your case does not REQUIRE a micro ATX board like what you have now and can support a standard size ATX motherboard since the Gaming K4 IS a full size standard ATX motherboard.

Motherboard Comparison

pcpartpicker.com

I don't think any motherboards that are more expensive are really worth spending the money on unless you plan to overclock the CPU and then none of these boards I've recommended would be a good choice and you will want something a little higher quality than the K4.

 

[Markerater]

Worst case scenario I just buy a different case for 30 bucks so it's not that big a deal for motherboard size. I'd rather spend a little on a case than have to settle for lower quality if it's micro or mini. I'll check those cases but it's gonna be a bit before I can afford one of them because refund period is well past due on the micro :/

 

[mdd1963]

Actual CPU temps as observed in HWMonitor might help us at least determine if we are even dealing with an overheating CPU....or rule it out, vice simply guessing as much. (Certainly, not all AIOs can handle even the 7700K, as I've seen some 7700K's w/ off-brand pumps allowing 85C temps at even stock clocks)

I'd be a bit surprised if a brand name Q/H/Z 270 board's VRM could not handle a stock clocked 7700K without throttling....but, even then, I'd not expect the rig to hard power off/reset, etc...

 

[Darkbreeze]

What is the exact model of your Corsair PSU and how long has it been in service?

As far as that Q270m motherboard goes, yes, there are absolutely some that can't handle a stock 7700k. And one with only a 3+2 power phase, and almost certainly very low quality components making up that VRM, undoubtedly fits the bill. It's like running an FX-8350 on a 760G chipset board.

 

[Markerater]

I'm positive it's not the psu as I've already replaced it, I don't have the exact model number of it I'll search when I get home. I'm buying a new motherboard today and we will see how that goes. We will have replaced almost everything at this point

 

[Markerater]

Actual CPU temps as observed in HWMonitor might help us at least determine if we are even dealing with an overheating CPU....or rule it out, vice simply guessing as much. (Certainly, not all AIOs can handle even the 7700K, as I've seen some 7700K's w/ off-brand pumps allowing 85C temps at even stock clocks)

I'd be a bit surprised if a brand name Q/H/Z 270 board's VRM could not handle a stock clocked 7700K without throttling....but, even then, I'd not expect the rig to hard power off/reset, etc...

When I was running the monitor I would watch it climb to 99 then the moment it hit 100 it'd shut off instantly. After replacing the CPU the temps wouldn't go above 65-75 under Max load for 10 minutes. But if I'm playing a game or doing anything GPU intensive then it hard resets.

 

[Darkbreeze]

Black screens and hard resets are OFTEN power supply related. Resets MIGHT be memory related in some cases and can be driver related as well. If a new board doesn't fix the problem and you will want to be sure to go to the product page for your motherboard and download/install the network adapter and audio drivers, then go to the Intel website and download and install the latest chipset driver for your chipset family, ie, Z270, H270, etc., then we can look at some other potential issues but honestly at that point it will almost be down to PSU and storage devices aside from potentially needing to just do a clean install of Windows.

Also, be sure to update the BIOS to the latest version right off the bat if it is not already on the latest version.

 

[Markerater]

Black screens and hard resets are OFTEN power supply related. Resets MIGHT be memory related in some cases and can be driver related as well. If a new board doesn't fix the problem and you will want to be sure to go to the product page for your motherboard and download/install the network adapter and audio drivers, then go to the Intel website and download and install the latest chipset driver for your chipset family, ie, Z270, H270, etc., then we can look at some other potential issues but honestly at that point it will almost be down to PSU and storage devices aside from potentially needing to just do a clean install of Windows.

Also, be sure to update the BIOS to the latest version right off the bat if it is not already on the latest version.

The bios update was one I saw I needed to do, and the PSU is what I thought to be a nice one. It has an M.2 for storage and I thought it could be memory related but it didn't explain why it only occured under graphical load. Never resetting when under the loads from HWMonitor. The parts arrive this week so I'll give an update once that happens

Edit: Corsair PSU Model CP-9020061-NA
Brand new out of packaging and already replaced once

 

[Darkbreeze]

 

[Markerater]

Did not fix. New motherboard and new tower to fit the new motherboard. Rip 200+ dollars. I have no idea what else to do. Using brand new m.2 already tried wiping that. All that's left is ram really :/ but can ram be causing a hard reset like this?

 

[Markerater]

Did not fix. New motherboard and new tower to fit the new motherboard. Rip 200+ dollars. I have no idea what else to do. Using brand new m.2 already tried wiping that. All that's left is ram really :/ but can ram be causing a hard reset like this?

Not a ram issue, tried a spare stack that I had. Upon further research I did not replace the GPU yet... I guess that's obviously the next step

 

[Markerater]

Not a ram issue, tried a spare stack that I had. Upon further research I did not replace the GPU yet... I guess that's obviously the next step

Sorry for the spam, but after testing more with HWmonitor everything is working just fine. I still reset when playing games and everything and I see that in getting Kernel 41 errors so I believe it to still be hardware related. I am able to return the PSU and I will do so and try a gold standard. Most forums point towards PSU. I'm ready to toss this thing out of the window I'm so frustrated.

 

[Darkbreeze]

Understandable.

That CX power supply you have. It is a black and gray label or a black and green label?

 

[Markerater]

Understandable.

That CX power supply you have. It is a black and gray label or a black and green label?

Was a green. Just bought her a full modular 750 watt gold. RX or RM I forget and it's been fine for a coue days now. I'm confused how a same watt power supply could be fine just different versions? Part quality or something? Because we replaced the power supply two times now and the third one is the charm I guess. Just changing from cx to rm (or rx)

 

[Darkbreeze]

Watts are not relevant. Watts on the label makes absolutely no difference UNLESS the unit is actually able to supply that wattage AND the unit doesn't have any other internal problems. It doesn't much matter if a unit can supply 1000w even if you only need 500w, if the unit is putting out twice the ripple that your system can handle or the specifications allow for, or if the voltage is fluctuating wildly between 11 and 13v when it should be staying more steadily between 11.8 and 12.2-ish. There is, obviously, a lot more TO this, and a lot more involved, but that is a quick burst of reasoning.

The green label CX units like what you had before were always trash. They were one of the more problematic units we commonly saw around here over the last six or seven years or so.

What you have now is a MUCH better power supply.

Be kind to yourself, for future endeavors, and read this:

Discussion - PSU recommendations and power supply discussion thread - Tom's hardware

Below you will find MY standard list of recommended power supplies and beyond that this thread is intended as a landing place for questions or discussions regarding specific units, platforms or related PSU tech, all of which are all welcome to be discussed here. If it's related to power supplies...

forums.tomshardware.com

Then read this:

https://www.corsair.com/us/en/blog/why-does-a-better-power-supply-mean-a-better-computer-experience


And this:

https://www.tomshardware.com/reviews/psu-buying-guide,2916.html


Educating yourself about power supplies is probably one of the single most important things you can do to make yourself "smarter" about desktop computers in general. When your power supply doesn't work right, NOTHING works right, because EVERYTHING relies on the power supply.