Jump to content

RTX 4090 Video Card Error in watercooled PC

DeadlyDeath

Hey together,

happy to be here! But sadly first post with a big not so happy issue

I have a problem with the watercooled computer I just set together again. I made a reddit post with picture and all details of the specs. Here is a link to that post: https://www.reddit.com/r/watercooling/comments/13fxnr7/new_tubes_for_first_watercooled_pc/?utm_source=share&utm_medium=web2x&context=3

Now, after the first start after reassembly and about 10 minutes of running totally fine as it always did, all monitors went off saying "no connection" while all the Corsair RGBs froze.. When I then manually restarted the computer, the error light of the mainboard kept enlighted near the VGA sign. So I added the Mainboard-Speaker and the error sound (long-short-short) stands for Video Card Error.

I thought it is weird as it worked totally fine in the beginning. The fittings of the watercooling didn't leak (in front and in the back) so neither the header in the PSU nor the Card itself or PCIE Header were wet.

As the CPU (7950X) has an APU, I then took away the power from the card and removed it out of the slot (just holding it in the hand without letting it touch the slot to not have to disassemble the cooling circle) before putting the HDMI into the slot of the motherboard. When turning it back on, there still was the sound and light of the Video Card Error.

I don't understand why neither the GPU nor the APU gives me a picture. The APU also not working gives me hope of not having destroyed my RTX 4090 (I when assembling the block, I changed thermal paste and pads as I have read that the original one can be difficult - but I took very high quality and thus non conductiv material, to make sure those things cant 't destroy the card from the inside).

 

Does anyone have an idea what to do or what the problem is? I would be very thankful for answers and advice, thank you very much in advance!

Link to comment
Share on other sites

Link to post
Share on other sites

Absolutely no leaks visible anywhere? Sounds like this might be a board related error, is there any way you can flash a new bios or flash back to a older bios version? See if that does anything. Not sure why the APU is failing to work as well, probably not a solution but you could try booting with a single stick of ram. See if that makes a difference. Do you have backup hardware you can test? Hardline tubing is going to be a pain to take apart and redo....

Like watching Anime? Consider joining the unofficial LTT Anime Club Heaven Society~ ^.^

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Quote
27 minutes ago, Fred Castellum said:

Absolutely no leaks visible anywhere? Sounds like this might be a board related error, is there any way you can flash a new bios or flash back to a older bios version? See if that does anything. Not sure why the APU is failing to work as well, probably not a solution but you could try booting with a single stick of ram. See if that makes a difference. Do you have backup hardware you can test? Hardline tubing is going to be a pain to take apart and redo....

 

The white pass-throughs you can see seem to have a density problem. Meaning when water flows through it with high preassure, it leaks, but just very very little water. First thing I did there was to put the clothes beneath it. This will be solved when new fully through it going pass-throughs arrive. But nothing dripped down from there and the clothes aren't even wet. 

The APU failing is the thing that seems very weird to me too.

I tried it with just one ram and it sadly still didn't work. I can understand that the APU might not work while the GPU is in the PCIE Slot, but I don't get the reason for which it didn't work when I hold it in my hands to have it out of the slot... (And when I assembled that computer for the first time, I had to install the Nvidia drivers to make the card work, and that time the APU worked while the GPU was installed..)

Yeah, I got an air-cooled SFF-Build with a 2070S in it. For just testing the GPU for some minutes it doesn't need to be cooled (meaning I can just take it out of this system and put it into the other one), does it? The testing system is in my old house like 120km away, but I guess I will go and get it tomorrow. Just as the APU doesn't work, I somehow have the feeling that something else isn't right.

Redoing the tubing will be a very special pain as this is the result of redoing it within the last days lol

I have absolutely no idea what to do.

Link to comment
Share on other sites

Link to post
Share on other sites

1 hour ago, DeadlyDeath said:

Redoing the tubing will be a very special pain as this is the result of redoing it within the last days lol

I have absolutely no idea what to do.

That's why I have only ever done a custom loop once lol.

Like watching Anime? Consider joining the unofficial LTT Anime Club Heaven Society~ ^.^

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

11 hours ago, Fred Castellum said:

That's why I have only ever done a custom loop once lol.

Hopefully this experience won't turn out that bad that the disappointment is this big and I will never do a watercooled pc again lol. In general I really liked the process of planning and building. I liked it even more than building an aircooled PC (also as there is more to do I guess haha)

I'm gonna be able to try the GPU in another system in about two or three hours, hopefully it works there

Link to comment
Share on other sites

Link to post
Share on other sites

Does anybody know if too many thermal pads can be an issue?
Although the XG7 was like all over taped with the pads and I just exchanged them, I mean the card has VRAM everywhere and a big chip which is a reason for needing that many pads. And I searched the internet for which ones to use and took either 1mm and 1,5mm, so theoretically it should be fine.. But although I took real good odyssey 12W/mK pads which shouldn't 'melt' like the cheap ones - especially not after some minutes..
Link to comment
Share on other sites

Link to post
Share on other sites

4 hours ago, DeadlyDeath said:
Does anybody know if too many thermal pads can be an issue?
Although the XG7 was like all over taped with the pads and I just exchanged them, I mean the card has VRAM everywhere and a big chip which is a reason for needing that many pads. And I searched the internet for which ones to use and took either 1mm and 1,5mm, so theoretically it should be fine.. But although I took real good odyssey 12W/mK pads which shouldn't 'melt' like the cheap ones - especially not after some minutes..

As long as your not stacking thermal pads on top of thermal pads you should be fine. I do know thermal pads have to be sized correctly for optimal contact with whatever your cooling, so you need to know how thick the pads have to be for your heatsink and gpu board.

Like watching Anime? Consider joining the unofficial LTT Anime Club Heaven Society~ ^.^

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

  • 4 weeks later...
On 5/13/2023 at 7:09 PM, Fred Castellum said:

As long as your not stacking thermal pads on top of thermal pads you should be fine. I do know thermal pads have to be sized correctly for optimal contact with whatever your cooling, so you need to know how thick the pads have to be for your heatsink and gpu board.

Thanks for your answer! Now, I'm a bit late with mine due to sickness and other technical issues, but hearing this is calming. 

When looking at the posts, I made, I also saw that some people said the 1.5mm pads being too thick, so I might change them for 1.0mm now. But at least the GPU in general seems to be working again (it was a mixture of three failing cables - the Strimer, which didn't give enough power, and two HDMI/DP Cables somehow leading to the error - that nobody could ever have guessed)... Now, I just seem to have high temps again and thus will exchange the pads..

Link to comment
Share on other sites

Link to post
Share on other sites

2 hours ago, DeadlyDeath said:

Thanks for your answer! Now, I'm a bit late with mine due to sickness and other technical issues, but hearing this is calming. 

When looking at the posts, I made, I also saw that some people said the 1.5mm pads being too thick, so I might change them for 1.0mm now. But at least the GPU in general seems to be working again (it was a mixture of three failing cables - the Strimer, which didn't give enough power, and two HDMI/DP Cables somehow leading to the error - that nobody could ever have guessed)... Now, I just seem to have high temps again and thus will exchange the pads..

No problem, hopefully getting properly sized pads help with with those temps. What sensor are you looking at for the temps?

Like watching Anime? Consider joining the unofficial LTT Anime Club Heaven Society~ ^.^

 

 

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×