Is the pump running? what components are you using on your water loop?

If you can verify that you pump is running and have good flow, then try bleeding the system.

 

did u remove the plastic covering from the cpu block?? make sure its off and that the cpu block is correctly fitted to the chip.

 

Wow that sounds really bad.

the chip is not dellidded? I had the same issue and it turned out that the IHS was not making good contact with the die.

But back to your problem. I usually bleed the loop by gently tiling the PC side to side and back to front with the pump running. My first PC build had the same problem (air) and checked this vid for some advice: Youtube: Bleed a Waterloop properly

 

Quote:

Originally Posted by LaBestiaHumana

Is the pump running? what components are you using on your water loop?

If you can verify that you pump is running and have good flow, then try bleeding the system.

Yeah the pump is running and I assume it has good flow. I'll check the video that flv posted and give bleeding another try.

 

Quote:

Originally Posted by moksh4u2

did u remove the plastic covering from the cpu block?? make sure its off and that the cpu block is correctly fitted to the chip.

I've had the waterblock for some months and there's nothing interfering with the CPU & block itself in terms of contact.

 

Quote:

Originally Posted by flv1333

Wow that sounds really bad.

the chip is not dellidded? I had the same issue and it turned out that the IHS was not making good contact with the die.

But back to your problem. I usually bleed the loop by gently tiling the PC side to side and back to front with the pump running. My first PC build had the same problem (air) and checked this vid for some advice: Youtube: Bleed a Waterloop properly

Nah dude it's not. I'll check out the video and see what I can do. Cheers.

 

With those temps, it's safe to assume the CPU isn't making contact with the waterblock.

If it was making contact, there's no way it would get that hot, so quickly.

 

Okay, a quick update. I think the pump may not be running, in which case I'm pretty impressed that I can actually use the computer with the fans cooling itself. I unplugged the pump from the PSU and it seems that nothing actually seems to start up and that temperature remains that same. This is really odd. Could this be caused by too much coolant?

 

Did you make sure your IN and Out for the CPU block was correctly installed? I actually had mine in reverse the first loop I made, though my temps were only about 5~10C higher than when I had it on correctly.

If bleeding doesn't work, try to apply more pressure on the CPU block by tightening up the screws on each corner. If all else fails, you should try to lower your voltage as most Ivy Bridge chips hit 4.5ghz at around 1.15~1.25v.

My chip is different, but it doesn't hit 90~100C unless I try to go for a Prime95 test for 5ghz @ 1.5V+

What kind of pump do you use? Do you feel any vibrations when it's plugged in?

 

Quote:

Originally Posted by COMBO2

Okay, a quick update. I think the pump may not be running, in which case I'm pretty impressed that I can actually use the computer with the fans cooling itself. I unplugged the pump from the PSU and it seems that nothing actually seems to start up and that temperature remains that same. This is really odd. Could this be caused by too much coolant?

Highly unlikely. I'd say a faulty pump. If it's a MOLEX connector, ensure the pins haven't been pushed back, or bent.

 

Quote:

Originally Posted by Chrit

With those temps, it's safe to assume the CPU isn't making contact with the waterblock.

If it was making contact, there's no way it would get that hot, so quickly.

It it. I promise you it is. I am using the PC as we speak, it's just that idle & load temps are ridiculously high, which is leading me to believe that it's the pump. Think about it. Water is still in the loop right? If it's just cooling by using the water sitting in the rad then it would be still able to keep somewhat cool. What I think is happening is that no new water is being pumped and the water heats up in the rad really quickly. That would explain the huge increase in like 3 seconds.

 

Quote:

Originally Posted by Kokin

Did you make sure your IN and Out for the CPU block was correctly installed? I actually had mine in reverse the first loop I made, though my temps were only about 5~10C higher than when I had it on correctly.

If bleeding doesn't work, try to apply more pressure on the CPU block by tightening up the screws on each corner. If all else fails, you should try to lower your voltage as most Ivy Bridge chips hit 4.5ghz at around 1.15~1.25v.

My chip is different, but it doesn't hit 90~100C unless I try to go for a Prime95 test for 5ghz @ 1.5V+

What kind of pump do you use? Do you feel any vibrations when it's plugged in?

It's the XSPC X20 750 res/pumo combo. I feel vibration but it's not spinning as far as I can see...

 

Quote:

Originally Posted by Chrit

Highly unlikely. I'd say a faulty pump. If it's a MOLEX connector, ensure the pins haven't been pushed back, or bent.

I'll try another molex plug and see what happens. Cheers.

 

Alright well changing molex connectors didn't help... Faulty pump or problem from bad install? What would you guys say? It's going from res > pump > rad > waterblock > res.

 

Even without delidding it - it shouldn't get anywhere near that temperature with that low an OC... that would definitely cause problems with the stock cooler (were it to be used) at stock clocks!

If it's a bleeding problem, it should be resolved fairly quickly without requiring the torture test. Just let it idle and keep and eye on the temps... if you're hearing air noises in the loop... they should stop fairly quickly since you've already run the system for awhile. For reference, my 3580K at 4.7GHz & 1.42V vCore had idle temps around 38C and peaked with one core at 87C after 10 minutes of IBT running (all cores were under 75C with Prime95 running).

It was horrible prior to delidding (the differences between core temps under load was nearly 15C!) - but it still never got hot like that.

It seems like either the mounting is bad (so try remounting the CPU again and check to make sure that you didn't leave the plastic protective label on the block (it happens apparently). When you pull it to remount clean and reapply the TIM - but before that make sure there's a good impression in the TIM from the last mount. Also make sure when you tighten the block down that you tighten the bolts a little bit in a circle... i.e. do 1 turn of one bolt, then 1 turn of the next bolt and so on, going around and around until it's fully tightened. Doing one side all the way at first can lead to uneven pressure on the chip which can cause bad temps.

If all of that fails to make a difference - which I think it probably will - then take the block apart and confirm that the cooling lanes aren't blocked by something or poorly milled.

And while you're at it... take a picture of your loop so we can look at it... it's possible that we can spot something that's not apparently from just your description (i.e. if you're using a directional block like the XSPC Raystorm - the water must flow in a particular direction for the block to work optimally). In any case, we need pics and more information.

 

Quote:

Originally Posted by DiGiCiDAL

Even without delidding it - it shouldn't get anywhere near that temperature with that low an OC... that would definitely cause problems with the stock cooler (were it to be used) at stock clocks!

If it's a bleeding problem, it should be resolved fairly quickly without requiring the torture test. Just let it idle and keep and eye on the temps... if you're hearing air noises in the loop... they should stop fairly quickly since you've already run the system for awhile. For reference, my 3580K at 4.7GHz & 1.42V vCore had idle temps around 38C and peaked with one core at 87C after 10 minutes of IBT running (all cores were under 75C with Prime95 running).

It was horrible prior to delidding (the differences between core temps under load was nearly 15C!) - but it still never got hot like that.

It seems like either the mounting is bad (so try remounting the CPU again and check to make sure that you didn't leave the plastic protective label on the block (it happens apparently). When you pull it to remount clean and reapply the TIM - but before that make sure there's a good impression in the TIM from the last mount. Also make sure when you tighten the block down that you tighten the bolts a little bit in a circle... i.e. do 1 turn of one bolt, then 1 turn of the next bolt and so on, going around and around until it's fully tightened. Doing one side all the way at first can lead to uneven pressure on the chip which can cause bad temps.

If all of that fails to make a difference - which I think it probably will - then take the block apart and confirm that the cooling lanes aren't blocked by something or poorly milled.

And while you're at it... take a picture of your loop so we can look at it... it's possible that we can spot something that's not apparently from just your description (i.e. if you're using a directional block like the XSPC Raystorm - the water must flow in a particular direction for the block to work optimally). In any case, we need pics and more information.

Thanks for the info mate. I'm using an XSPC Rasa waterblock, AlphaCool NexXxos ST30 240 (bought yesterday) & a XSPC X20 750 res/pump combo. There are pictures on my page under PC maintenance album. Along with those I've got some 6x XSPC compression fittings and 7/16" ID, 5/8" OD tubing. at the front of the res, I actually can't see the pump spinning. How long are the warranties on these things? Only had it for about 16 months...

 

Quote:

Originally Posted by COMBO2

Thanks for the info mate. I'm using an XSPC Rasa waterblock, AlphaCool NexXxos ST30 240 (bought yesterday) & a XSPC X20 750 res/pump combo. There are pictures on my page under PC maintenance album. Along with those I've got some 6x XSPC compression fittings and 7/16" ID, 5/8" OD tubing. at the front of the res, I actually can't see the pump spinning. How long are the warranties on these things? Only had it for about 16 months...

On the Rasa block there is an in and out so assumedly it's a directional design - however even if you had that backwards, it shouldn't be THIS bad... for reference if the XSPC logo is at the bottom the left port should be the inlet and the right port should be the outlet. (Of course, in theory it should actually say 'in' and 'out' but if it's like my Raystorm... this wears off quickly). Unfortunately it doesn't say what the warranty is on that pump on the XSPC site - or if it does, I sure couldn't find it. Seems unlikely it died however - and the ST30 should be great for getting the heat away on just a CPU... so I'm still thinking it's either a problem with the block mount, or the block is physically clogged (if it's used it's quite possible, if it was new this shouldn't be the case).

Since you can't really physically see if there's turbulence in the reservoir since it's in your system you would have to drain the system and pull it to test in a sink, etc... and if you're going to all that trouble - you could just put one of these handy things in your system.

One of the other things to check if your tubing is running behind things or around a tight corner is to verify that you don't have a kink in your tubing - that can cause starvation to the block or prevent proper bleeding of the loop... which would cause problems with temp.

Really there's several possibilities - and a dead pump is definitely one of them - but not necessarily the most likely reason.

 

Quote:

Originally Posted by COMBO2

Alright well changing molex connectors didn't help... Faulty pump or problem from bad install? What would you guys say? It's going from res > pump > rad > waterblock > res.

is the pump running and liquid just isn't moving, or is the pump dead and not making any noise when you turn it on? You either have a bad blockage somewhere from the coolant or a dead pump. Bummer

 

Quote:

Originally Posted by DiGiCiDAL

On the Rasa block there is an in and out so assumedly it's a directional design - however even if you had that backwards, it shouldn't be THIS bad... for reference if the XSPC logo is at the bottom the left port should be the inlet and the right port should be the outlet. (Of course, in theory it should actually say 'in' and 'out' but if it's like my Raystorm... this wears off quickly). Unfortunately it doesn't say what the warranty is on that pump on the XSPC site - or if it does, I sure couldn't find it. Seems unlikely it died however - and the ST30 should be great for getting the heat away on just a CPU... so I'm still thinking it's either a problem with the block mount, or the block is physically clogged (if it's used it's quite possible, if it was new this shouldn't be the case).

Since you can't really physically see if there's turbulence in the reservoir since it's in your system you would have to drain the system and pull it to test in a sink, etc... and if you're going to all that trouble - you could just put one of these handy things in your system.

One of the other things to check if your tubing is running behind things or around a tight corner is to verify that you don't have a kink in your tubing - that can cause starvation to the block or prevent proper bleeding of the loop... which would cause problems with temp.

Really there's several possibilities - and a dead pump is definitely one of them - but not necessarily the most likely reason.

With the reservoir though, it's got the pump inside it, and the opening of the pump for the water I can see. There would usually be a turbine spinning. I see a turbine, but it's not spinning. There isn't any kinking and as far as I know air wouldn't create this issue and I can't see any. I tested the thing outside of the case and it seems to work fine so I'm really frustrated right now... How long is the warranty period on pumps and stuff because nor can I & especially not my parents pay for another res/pump...

 

Quote:

Originally Posted by DaXxJaPxX

is the pump running and liquid just isn't moving, or is the pump dead and not making any noise when you turn it on? You either have a bad blockage somewhere from the coolant or a dead pump. Bummer

When I plug it in to the 4 pin molex, it starts vibrating but nothing spins and there are no bubbles formed anywhere in the res. I'm almost certain it's not a blocked part of the WC loop. Everything is set perfect... so confused...

 

ive just got a raystorm kit (RS750)

to check if water is flowing i use a torch and shine through the pump/res window, the water moves about a lot.

also pinch the tube, you should feel it vibrate/flow

just a question, when the temps raise very quickly, what do the temps fall like?

usually a poor mount the temps raise quickly and drop slowly! how long does the temps take to fall.

 

Quote:

Originally Posted by d1nky

just a question, when the temps raise very quickly, what do the temps fall like?

usually a poor mount the temps raise quickly and drop slowly! how long does the temps take to fall.

This will determine if you have a bad mount. If they fall just as quickly (1~2 seconds) from 100C to your 45C idle, it means it is not a mounting problem.

 

Quote:

Originally Posted by Kokin

This will determine if you have a bad mount. If they fall just as quickly (1~2 seconds) from 100C to your 45C idle, it means it is not a mounting problem.

if they do that...... raise and fall quickly at and after full load. the problem may be more serious. i dont have a clue what, first guess voltage regulator.

 

Quote:

Originally Posted by d1nky

ive just got a raystorm kit (RS750)

to check if water is flowing i use a torch and shine through the pump/res window, the water moves about a lot.

also pinch the tube, you should feel it vibrate/flow

just a question, when the temps raise very quickly, what do the temps fall like?

usually a poor mount the temps raise quickly and drop slowly! how long does the temps take to fall.

Yeah it drops as quickly as it raises. I'm still thinking it's the pump. Nothing is really moving or flowing in the res and theres no pressure in the tubes and it doesn't feel like anything is flowing. Damn though, I'm impressed by the rad. It's acting like a heatsink at the moment!

Another thing I should mention is that on 4.2GHz (OC Genie's 1.275 Vcore) it jumps to 85 instead of 102 but it still slowly increases.

 

water will be moving if it drops as quick as it raises!

no way could stagnant/still water raise in temps then drop like that, there would have to be some kind of flow.

maybe the flow is just slow, but it wouldnt drop that quick.

definitely reseat the cpu and make sure theres enough pressure, and not too much!