PC problem, trying to figure out which component is bad.

Sato_

Reputable
Feb 27, 2014
6
0
4,510
Alright so I think I just broke my pc, trying to figure out what exactly is failing. I will try to be detailed here, also hope my English is enough to get this across.

My build

Case: Fractal Define r4
Mobo: ASRock Z77 Pro4
RAM: Corsair Vengeance Blue @1600mhz
CPU: I5 3570K @4.5ghz
Cooler: Corsair H100i
GPU: Asus GTX 1070
PSU: OCZ ModXStream Pro 600w
SSD: Intel 330 60Gb (OS)
SSD: Samsung Evo 840 250gb (storage)


As you can see pretty much everything in my build except the GPU is ancient, and has been running since 2011, and it still performs like a beast.

What happened was I noticed that my CPU temps under load were gradually becoming higher than usual. When playing new games like Kingdom Come Deliverance, the cpu would be at 78-83c. This game pretty much keeps my CPU at full load the whole time, so you would think that these temps while a bit high - aren't unheard of for a 4,5ghz OC. But the weird thing is when playing something like Escape from Tarkov which is very poorly optimized and uses no more than 30-50% CPU, the temps were still in lower-mid 80's.

So naturally my first thought was to change the thermal paste. Because it hasn't been touched since around 2013 when I slapped on the h100i (using the stock compound pre-applied on the heatsink). I should definetly also blow out the rad as its pretty dusty, but I dont have acess to an air compressor at home.

Anyway when I unbolted the h100i heatsink from the mobo, I realised that the thing is totally fused to the CPU, it was stuck so firmly there wasnt a slightest chance to just pull it off. Watching some guides I saw people using some alcohol on the sides of the cpu and using a teeth floss to unstick it, but they all had standard heatsinks and could remove them (together with the glued on CPU) from the case to have better acess to it. For me it wasnt so easy, because of the 2 hoses coming out of the side of the cooler there was no way to unlock the CPU retaining latch. The hoses blocked the movement of the pin. So I thought about it for a minute, and what I came up with was to turn the PC back on, and run Aida 64 to heat it up a little bit. So I ran the test, left it for a couple minutes to let it warm up and then started gently twisting the cooler side to side. And eventually it just popped off! It required some force, but nothing too crazy. But as soon as it popped off the pc crashed, showing a very beautiful and colourful artifact pattern on the screen lol. At that point I instantly shut the PC down by holding the power button for a couple seconds.

Something to note: at the time the cooler popped off, I had already stopped the Aida 64 test a moment before, and the temps were already down to like 40c. Also the CPU stayed firmly secured in its seat, no pins were bent or anything. And my GPU was not in the case at the time, I ran it from the IGPU.

I was a little worried about the crash, maybe the mobo twisted too much and something just broke? Anyway I went on cleaning the old thermal paste with isopropyl alchohol 99%. And it was so hardened I literally rubbed the heatsink non stop for like 15minutes to completely get it off (took only a few swipes for the CPU though). Then I applied some arctic MX-4 and put everything back together. Still without GPU.

The PC powered on fine, in fact im typing this on it right now. The problem is that once I put my GPU back in, there is no signal to the monitor! The power light and RGB's on the GPU light up fine, the fans are not spinning but they are not supposed to until 60c. If I put in my old and still working gtx 770 they do spin btw. Still no signal though.

what i've tried so far: plugging into different ports with different cables, DVI and HDMI. Tried completely taking out and reseating the card to make sure its in properly. Tried another GPU which still didnt work (hopefully this means my 1070 is still ok). Tried to apply some pressure on the ram just in case it popped out of socket during the cleaning. I did this without turning my PC off and just with the slightest pressure applied on the ram sticks - the pc instantly shut down. So either the mobo is broken, or I shorted it when pressing down. The standoffs are pretty long though. A few reboots later now my audio is gone too. The usb ports are still working tho.

Also the PSU seems to be acting weird, when I used the on/off switch (the one on the psu itself) it made a quite loud clicking noise, as if there was a spark inside or something.

What do you guys think is happening here? Do you think the mobo is on its last legs? I knew it was time for a new build soon... I just wasnt ready for this tho. Its a sad day because my 3570k is still an absolute champ :( If I was to get a new one now It wont even be that much of an improvement I think. Hell I havent even been keeping up with the cpu market, hopefully there is something nice and affordable.

edit: I have a brand new 1000w EVGA psu lying around waiting for the new build. I could try to swap it for a test but then again i think if there was something wrong with the PSU i wouldnt be typing right now lol.

 
Solution

westeind

Prominent
Jul 16, 2017
67
0
640


Have you tried another PCI-E slot on the mobo?
 
Solution

Sato_

Reputable
Feb 27, 2014
6
0
4,510
Have you tried another PCI-E slot on the mobo?

I just did that and it works. Also had my sound back for a few seconds but its gone again. So i guess ill swap it back to the main slot again now to confirm its really dead... If so I believe the mobo is done for right? What are the disadvantages of running it from the lower slot?

edit: actually the sound is fine, the playback sources were reset and mixed up for some reason.
thats what my mobo manual says about the slots:
- 1 x PCI Express 3.0 x16 slot (PCIE2: x16 mode)
- 1 x PCI Express 2.0 x16 slot (PCIE3: x4 mode)
- 1 x PCI Express 2.0 x1 slot

So now im in the 2.0 slot, x4 mode. Not sure what that means but im pretty sure its shit :(
 

westeind

Prominent
Jul 16, 2017
67
0
640


I think buying a new mobo would solve all of your problems but i dont know if you think thats worth it, i dont know what else you are able to do tbh
 

Sato_

Reputable
Feb 27, 2014
6
0
4,510


dont think i can buy a new mobo for a 2011 cpu lol, and i really dont feel like buying used parts... already putting together a new build
 
sure sounds like your got as short somewhere on that board, and sounds like it slowly taking components (chips) out of the work force.
My bet is you need to change the motherboard for another one.
taking in consideration on eBay or Newegg the average price for a new (yes they exist) Z77 board runs about 130-140$, maybe you need to consider an upgrade of sorts.

link in case you rather just buy a board https://www.newegg.com/Product/ProductList.aspx?Submit=ENE&DEPA=0&Order=BESTMATCH&Description=z77&ignorear=0&N=-1&isNodeId=1
 

Sato_

Reputable
Feb 27, 2014
6
0
4,510


Well I am already ordering parts for a new pc, I already have an almost brand new 1070 and a brand new EVGA supernova p2 1000w psu, so 2 quite expensive parts are already crossed off the list. And it was time for a new system anyway. This one will be of good use for my parents since its snappy as hell and they are using a dual core from 2007 with no ssd... it will be a dream for them!

Another problem is that I live in an eastern european shithole where I have to buy locally available parts (from a very limited selection too sadly) because shipping anything from abroad is very impractical here. Newegg doesnt even ship to my country at all.

I literally just paid 50 euros for shipping of Be Quiet Dark Rock Pro 3 cooler off ebay, because I only had absolute trash options locally. Shame I will have to wait like 3 weeks for it.... Luckily I put together a pretty good system with what I could find here other than the cooler.

Could pick up some AIO for about the same price but I just dont see myself needing to overclock an 8700k for a long while, so I wanted to avoid all the risks that come with closed loops for now.