Overclock.net banner

1 - 19 of 19 Posts

·
Registered
Joined
·
734 Posts
Discussion Starter #1
I think I fried one of my 980Ti's. I put Heatkiller Waterblocks on both cards, and they didn't have the best instructions, one works fine, but the other wouldn't even boot if it was connected. I finally got it up and running but as soon as I ran Valley Benchmark, I got a hard shutdown. Now if I plug in that one card to the power supply it will not boot.

I ran Valley with the other card and it ran perfect, on the other Heatkiller block, even overclocked it never went above 34C.

I'm not sure what I did wrong. I put all pads on correctly, I put the three rubber bumpers between the backplate and the back of the card. There where some black plastic pieces, that are stil in the bag, that I never used. The instructions mad no mention of these though.

I ordered another card from Amazon today and will try and put the block on the new card. I just don't want to bust another card.
mad.gif


I know on the EK Waterblocks. I've used I had to leave a few screws in the around the GPU portion, and all the rest of the screws where removed and replaced when you put on the backplate. On the Heatkiller instructions you put all new screws on when you put on the backplate, and there are three rubber pumpers between the backplate and card.

Anyone with any Heatkiller experience have any advice on what I might of done wrong?
 

·
Registered
Joined
·
16 Posts
Did you remove the block and test the card with the stock cooler to confirm it really is dead?

You might have shorted something, but it could come back on the stock cooler.
 

·
Registered
Joined
·
734 Posts
Discussion Starter #3
Quote:
Originally Posted by ilal2ielli View Post

Did you remove the block and test the card with the stock cooler to confirm it really is dead?

You might have shorted something, but it could come back on the stock cooler.
I've actually had the card for a while. It was fine air cooled, and also on a ek block for the past year.

I just love the look of the heatkiller and wanted to change to that block.

I swapped 2 blocks at the same time one is,fine while the other..... not so good.
 

·
AMD s939 LVL 99+
Joined
·
2,996 Posts
I put a block on an AMD 5870 YEARS ago, and somehow accidentally broke a SMD resistor / capacitor off the board. Found some who knows what it was SMD, and went for broke. Worked again, lol. Maybe you chipped off a SMD component?

It doesn't take much.
 

·
Registered
Joined
·
734 Posts
Discussion Starter #6
Quote:
Originally Posted by ianthegreater View Post

+1 for trying the old cooler on there. You won't know if it's truly dead unless you try the last known-good state.
If I did short it by a faulty install, is there anyway to fix that?
 

·
Registered
Joined
·
830 Posts
Quote:
Originally Posted by Newtocooling View Post

If I did short it by a faulty install, is there anyway to fix that?
Well, one way is to put the stock air cooler back and see if it works.

Obviously, if it works on air - then there was a problem with previous installation.
 

·
Registered
Joined
·
734 Posts
Discussion Starter #8
Quote:
Originally Posted by Ironsmack View Post

Well, one way is to put the stock air cooler back and see if it works.

Obviously, if it works on air - then there was a problem with previous installation.
It was working prior on an EK Waterblock as well though for the past year.
 

·
Registered
Joined
·
139 Posts
Quote:
Originally Posted by Newtocooling View Post

If I did short it by a faulty install, is there anyway to fix that?
It depends on whether or not there was any physical damage to the card. If you're lucky the installation just kept the card from powering on and didn't fry a cap or anything.

If it did break something, you miiiiight find a charred cap, but chances are slim.

What I was trying to get at is that you've said the card is dead with the other cooler. If it's truly dead, you can't really harm it much more by putting the stock cooler back on and seeing if it works. You never know - you might get lucky.
smile.gif
 

·
Registered
Joined
·
734 Posts
Discussion Starter #10
Quote:
Originally Posted by ianthegreater View Post

It depends on whether or not there was any physical damage to the card. If you're lucky the installation just kept the card from powering on and didn't fry a cap or anything.

If it did break something, you miiiiight find a charred cap, but chances are slim.

What I was trying to get at is that you've said the card is dead with the other cooler. If it's truly dead, you can't really harm it much more by putting the stock cooler back on and seeing if it works. You never know - you might get lucky.
smile.gif
By charred cap would that be noticeable after taking the cooler off? Thanks for all the help by the way I really appreciate it.
thumb.gif
 

·
Registered
Joined
·
139 Posts
Yeeesh - that thing looks nasty. That was definitely a larger 'thermal event.'

If something like that happened you'd get a lot of smoke. Since a lot of the surface mount caps are really small (just a tiny bit bigger than grains of pepper) it'll be difficult to get a good look to see if you notice any damage. If you sniff around you might be able to smell any burned up components - the smell is very noticeable.

To be honest, the likelihood of finding something is slim. Still worth a shot though.
 

·
Newb to Overclock.net
Joined
·
4,185 Posts
What PSU did you use? Are you using the split connectors or 1-1 PCI-e cables? Check for any melted ends.

I had a similar issue on my single GPU by running a split connector and did a benchmark pulling around 400W (with custom BIOS). GPU and PC keeps shutting down whenever the card is stressed, and eventually didn't boot. Turns out, it was a melted connector on my Seasonic X850 and not my GPU. A simple swap of PCI-e cable resolved the issue for me.
 

·
Registered
Joined
·
734 Posts
Discussion Starter #14
Quote:
Originally Posted by mouacyk View Post

What PSU did you use? Are you using the split connectors or 1-1 PCI-e cables? Check for any melted ends.

I had a similar issue on my single GPU by running a split connector and did a benchmark pulling around 400W (with custom BIOS). GPU and PC keeps shutting down whenever the card is stressed, and eventually didn't boot. Turns out, it was a melted connector on my Seasonic X850 and not my GPU. A simple swap of PCI-e cable resolved the issue for me.
Corsair 1500 AXI with Ensourced cables I did swap and try original corsair cables too. Straight 8 pin cables no splits.
 

·
Registered
Joined
·
734 Posts
Discussion Starter #15
Well thanks to Jakob I finally figured out my mistake. When I installed the backplate I misunderstood the directions and missed the part to install the black standoff plastic pieces.
mad.gif


I have one card re-installed properly and working awesome. Overclocked and running GTA V, or stress testing with 3D Mark temps don't go over 31C

GPU Clock Offset: 300Mhz
Mem Clock Offset: 400Mhz

Heatkiller blocks for me from now on!!!!
thumb.gif
 

·
Premium Member
Joined
·
1,126 Posts
Did the other card make it?
 

·
Registered
Joined
·
734 Posts
Discussion Starter #17
Quote:
Originally Posted by d0mini View Post

Did the other card make it?
The second card works, but for some reason It shows on GPUZ and Asus bios as X4 PCIE lane bandwidth. So I've tried another motherboard, and still the same. I'm in the process of looking for a used second card on Ebay now.
rolleyes.gif


This whole build a has been a lesson in frustration. At one point I was worried, since my case is on casters that I was going the kick the D....mn thing out the front door and let it roll out into the street.
biggrin.gif


Live and learn, I just lucked out that the 1080 was released, it allowed me to get 2 cards for the price I originally paid for one.
 
1 - 19 of 19 Posts
Top