• Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!

Q-code 62 and vga led after installing EK waterblock on evga 3080 ftw3

Overclockers is supported by our readers. When you click a link to make a purchase, we may earn a commission. Learn More.

Filip.K

Registered
Joined
Mar 4, 2021
So I finally finished my watercooling loop with evga 3080 ftw3 with EK water block and 9900k with corsair xc7 water block.

But after trying to boot first time I got debug error code 62 with no signal to monitor and vga led light up on motherboard (z390 gigabyte aorus master)

Fixes I already tried:
1. CMOS reset
2. reseated in other pcie lanes (was using riser cable, so I tried with and without)
3. checked and reseated all the power conectors , PSU side too (not using daisy chained cables, psu is corsair hx1000i)
4. I can boot into windows with integrated gpu
5.reinstalled gpu stock cooler (fans are spinning, rgb strips work too)

Also, don't know if its important but the power connectors on the gpu have leds over it. They will light up red for a split second when powering the pc up and turn right off and stay like that whole time.

Gonna try to run the gpu in friends pc.

Any suggestions?
 
Blinking LEDs on the PCIe power leads? I don't believe that is normal. Check the GPU manual and see what the flashing lights mean.
 
If possible - Try another GPU in your PC.... Right now you have only 3 options :-(


1) -> You borked something while installing the Water Block and you card is OK, just needs proper cleaning
2) Your PC no longer likes your GPU and will not continue to POST. -> Hope for this <-
3) You TOTALY BORKED the card while installing the Water Block and it will not work - Period - From your comments - This is your problem - :rain: :cry:

^ IIRC - EVGA covers their cards even if you remove and reinstall the HSF. IIRC - as long as the ORIGINAL BIOS is still there so is their warranty :clap: Have you Registered the card yet?? Did you at least get to drive it around the block before you installed the Water Block??

I'm sorry for the BAD NEWS... :chair:
 
Blinking LEDs on the PCIe power leads? I don't believe that is normal. Check the GPU manual and see what the flashing lights mean.

The leds are there to indicate power failure. However in all the cases I saw on the internet, if there is a problem, they stay lit. My just flash once and they turn off. Manual doesn’t mention them.

- - - Auto-Merged Double Post - - -

I really dont believe I hurt the gpu in any way. I inspected the pcb and die many times now and they look same as new. No scratches, no scorching, no cracks, no blown fuses, nothing...

anyways, how can my pc just not like it? I actually had this card since October.

On other forums people suggest that it may be being shorted by something, I am not sure id I installed thermal pads correctly back under the stock cooler. I will try reseat the cooler one more time and try original PSU cables since I used ones from cable mode..
 
I'd maybe try a different PSU for giggles. But, sounds like a short on something... :(

Did you inspect the card after you took the block off? If not, maybe take a close(r) look (again) to see if you can find damage.


EDIT:
and try original PSU cables since I used ones from cable mode..
Mmhmm............... check that..........now that we knows this, swapping back or using a different PSU to test makes complete sense....:thup:
 
The psu should be fine, I can fet into windows when using igpu and have no problems.

I inspected the card and there is no damage I suspect the short too, but there is no real way of knowing, than keep messing with the card which I am now afraid off.

- - - Auto-Merged Double Post - - -

I was using cable mode before the waterblock installation and the card workd without problems, so thats why I am not suspecting them now. But I will try anyway
 
I think I made a mistake in the q code assumtion, it actually is B2 and not 62, I dont know if it changes anything, I used the orginal powesupply cables, the result ist still same, I also reamounted gpu cooler, when the pc turned on, the gpu actually warm to touch, I dont know if it means anything.
 
I had a look at the pics but couldn't see anything obvious.
Have you tried the original cooler on the card to verify it still works?
Code B2 from what I can find refers to CSM or compatibility support. This is UEFI or legacy depending upon the parts in your PC and your OS. This is an option in BIOS. Have you reset the BIOS during this?
 
Yes, I reinstalled the original cooler now two times, still nothing, the gpu lights up, fan spins, card is warm to touch after some time, but no signal.

I updated MB bios on the Z390 AORUS MASTER from F10 to F11n, which came out in the end of january. I have reseted bios many times now. CSM is enabled. I tried UEFI and LEGACY settings too.

I running window 10, 32gb corsair vengance memory - 4 sticks, 3x nvme ssds, all in mb

I was running all of this before, but the problems came after dissasambaling the gpu and installing the water block...

I have no idea what to try next...
 
I assume that you have used the iGPU from the CPU so you can get to BIOS and Windows. Just verify that CSM is enabled and also that the graphics is set to auto in BIOS so it will pick up the card when installed. Also that the cable is installed into the card and not the motherboard
 
That does change the what can be wrong.. Q62 is a straight up Gpu problem. Most of the time it’s just something simple BUT other times bad..
Did you install different pads of the memory or vrm??
Take some IPA (70% - 90 % is better) and Q-tips, then clean all the area of the PCB, making sure to clean the vrm area good..
 
Yes i can get to BIOS with integrated gpu.

The CSM is enabled for sure.
graphics is set to AUTO now, but i tried disabling the igpu and force the graphic input on PCEI slot 1 - no difference.

I read on some forums taht disabling the CSM may help in this situation so thats, what I am gonna try next.

- - - Auto-Merged Double Post - - -

I did that today, you can see the pictures I posted. I am using thermalpad that EK provided to the block because i trew away the original ones that got teared, also I dont have a thermal puddy on hand. Could bad contact of thermal pads with vrm or other componet on the pcb do this?
 
You'll need some type of thermal compound for the GPU die. If things weren't done correctly it is possible that some components may have overheated but typically they would just shut down not burn the component.
 
I am using thermal paste for the gpu die and thermal pads for the memory, mosfets and vrm. Pretty much how the instructions told me to do. Also, I saw some post that overtighting the screws on gpu can cause similar issues. I loosedned the screws a little bit but still there is no change.
 
Okey, so I managed to do something. I disabled CSM in the bios. After that when I tried to post with th gpu installed, it gave me debug 97 (conole output device connect) Am I getting somewhere?
 
It is sounding like there may be something wrong with the card.
not sure at what point you updated the BIOS but just for S&G try using the second BIOS on that board and see what happens.
 
Yes, the problem is certainly caused by gpu after I applied waterblock on it. The question is, how to fix it? Actually I found out there is maybe dozen more people that have this exact problem with 3000 series cards and EK waterblocks. Its becoming a wide spread issue. I dont believe that so many of us screwed up the waterblock installation so much that we destroyed our cards. Or maybe the waterblock caused some short that actually killed the card. But how can we know?

I change in motherboard bios didnt fix the issue
 
Back