-
Welcome to Overclockers Forums! Join us to reply in threads, receive reduced ads, and to customize your site experience!
Just installed watercooling... HELP
- Thread starter whozyodaddy
- Start date
OP
- Joined
- Mar 6, 2005
- Location
- California
- Thread Starter
- #102
He just doesn't think they work. He wants me to return the 6002 because he thinks it is defective, same with the maze 4. He also wants to buy a heat generator, and test the cpu waterblock, I have such bad luck with this.
Last edited:
Theoretically, you can't get your temps any cooler cooling with water than with air because in the end you are still cooling with air (blowing over the radiator).
It's my opinion that the reason for water cooling is to remove the heat to another location (radiator) by water so that air cooling can be more effective. This allows you to have a larger surface area (radiator) than any heatsink you can fit onto a cpu to disipate the heat. You can only hang so large of a heat sink onto the cpu and put so many fans into a computer case blowing on it. Water cooling allows you to move the heat to a location where you can mount a larger heatsink (radiator) which can be cooled with larger and more numerous fans (maybe even quieter ones).
A bonus for me is noise reduction. The OEM fan and heat sink that Intel provided was just too loud and after awhile I was willing to pay just about any price to reduce the noise. It will also allow me to overclock and keep my temps under control without making my computer room sound like it is a helicopter pad.
It's my opinion that the reason for water cooling is to remove the heat to another location (radiator) by water so that air cooling can be more effective. This allows you to have a larger surface area (radiator) than any heatsink you can fit onto a cpu to disipate the heat. You can only hang so large of a heat sink onto the cpu and put so many fans into a computer case blowing on it. Water cooling allows you to move the heat to a location where you can mount a larger heatsink (radiator) which can be cooled with larger and more numerous fans (maybe even quieter ones).
A bonus for me is noise reduction. The OEM fan and heat sink that Intel provided was just too loud and after awhile I was willing to pay just about any price to reduce the noise. It will also allow me to overclock and keep my temps under control without making my computer room sound like it is a helicopter pad.
I see nothing wrong with this. I think there are a lot of people on this forum who would like to see an independent test run on watercooling gadgets.
Maybe you can get him interested in doing this. A great hobby for you and your dad to do together. Be sure and post your results.
Keep in mind when talking with your father that he is an engineer and they are trained to think about cost vs. effectiveness. In other words if a $5 item will do just as good or nearly as good a job as a $450 item then in his mind it's not worth the extra cost for increased performance and you should use the $5 item.
Overclocking is not a business, it is a hobby and no it's not cost effective. People pay money for entertainment and education. Computers are entertainment with a side benifit of being educational. In my opinion, well worth the cost on both accounts.
OP
- Joined
- Mar 6, 2005
- Location
- California
- Thread Starter
- #107
Many good points are being spread here, special thanks to you CJ-5.
Although the problem still exists... 70c idle cpu temps and a possibly dead gpu. I am going to try to bleed the system again, then launch up with just gpu cooled and leave it on for a while.
I guess we have pointed out every possible problem, no?
Although the problem still exists... 70c idle cpu temps and a possibly dead gpu. I am going to try to bleed the system again, then launch up with just gpu cooled and leave it on for a while.
I guess we have pointed out every possible problem, no?
- Joined
- Apr 1, 2002
Since your block aren't getting warm and the processors are roasting I think we should go back to the processor->waterblock thermal transfer problem side of things.
If you:
1) clean off your blocks and processors,
2) coat the blocks (blocks only) with a thin layer of thermal goop,
3) mount the blocks and torque down the mounting hardware to manufacturers specs,
4) let your computer do a 30sec failed boot cycle,
5) remove the blocks,
do you get an even layer of thermal goop on the processors and a nice imprint of the processors in the goop on the bottom of the blocks?
edit: for spelling
If you:
1) clean off your blocks and processors,
2) coat the blocks (blocks only) with a thin layer of thermal goop,
3) mount the blocks and torque down the mounting hardware to manufacturers specs,
4) let your computer do a 30sec failed boot cycle,
5) remove the blocks,
do you get an even layer of thermal goop on the processors and a nice imprint of the processors in the goop on the bottom of the blocks?
Hey! I'm an engineer and I waste time and money on stupid cooling things all the timeCJ-5 said:
edit: for spelling
Last edited:
Ok, so your dad is an engineer.. let him read this, fair amount of numbers:
http://www.icepak.com/prod/icepak/solutions/articles/iceart19.htm
Now onto your pictures:
1) Since the radiator is sitting on the bottom of the case:
a) Do you have a hole cut in the bottom of the case for the fans to push/pull air? Not having a hole will detriment the performance of the radiator.
b) Having the fans blow into the case would provide the best benefit (imho) because they would blow directly onto your video card.
2) It looks like your video card mount is incorrect... Why are you not using all 4 holes on the mounting block? It also looks like you have no screws (threads) above the gold washers on the mount. Did you cut these? If you don't have any, I would suspect you do not have the block screwed down tight enough.
3) Did you remember to plug the 4pin power adapter in to the video card?
4) I assume that is you "holding" the 6002 block up to your cpu... When you do that, is there any pressure put on the corner of the video card due to all of the cabling you have there and then the 3/4" tube stress from the pump-> cpu feed?
5) Lets go over your loop... from the picture it looks as follows:
D5->6002->Maze4->BIX2->Reservior->D5
This correct?
6) Lets discuss the CPU mounting... When you mounted it, did you mount it with the step side over the socket lever? Since this block is mounted with the discharge on the bottom, have to let it run while its on top for about 8-10 hours to bleed air out of the cpu block? Do you have the center barb (inlet) coming from the d5? If you don't, then your tubing is incorrect. Does your assembly match here: http://www.swiftnets.com/products/installationguide_MCW6000_2-775.pdf
4) For a stacker, it seems as though you have little room inside.
Thats all I can think of right now.... Lets see if we cant figure this one out...
http://www.icepak.com/prod/icepak/solutions/articles/iceart19.htm
Now onto your pictures:
1) Since the radiator is sitting on the bottom of the case:
a) Do you have a hole cut in the bottom of the case for the fans to push/pull air? Not having a hole will detriment the performance of the radiator.
b) Having the fans blow into the case would provide the best benefit (imho) because they would blow directly onto your video card.
2) It looks like your video card mount is incorrect... Why are you not using all 4 holes on the mounting block? It also looks like you have no screws (threads) above the gold washers on the mount. Did you cut these? If you don't have any, I would suspect you do not have the block screwed down tight enough.
3) Did you remember to plug the 4pin power adapter in to the video card?
4) I assume that is you "holding" the 6002 block up to your cpu... When you do that, is there any pressure put on the corner of the video card due to all of the cabling you have there and then the 3/4" tube stress from the pump-> cpu feed?
5) Lets go over your loop... from the picture it looks as follows:
D5->6002->Maze4->BIX2->Reservior->D5
This correct?
6) Lets discuss the CPU mounting... When you mounted it, did you mount it with the step side over the socket lever? Since this block is mounted with the discharge on the bottom, have to let it run while its on top for about 8-10 hours to bleed air out of the cpu block? Do you have the center barb (inlet) coming from the d5? If you don't, then your tubing is incorrect. Does your assembly match here: http://www.swiftnets.com/products/installationguide_MCW6000_2-775.pdf
4) For a stacker, it seems as though you have little room inside.
Thats all I can think of right now.... Lets see if we cant figure this one out...
OP
- Joined
- Mar 6, 2005
- Location
- California
- Thread Starter
- #110
@JCLW - Yes, the goop does spread around in a larger circle. The outside of the proccesser is (permanently) imprinted on the cpu block.
a. Yes, the stacker already came with a cut bottom, I just removed the fan grille.
b. I have one pushing and the other one pulling.
I can only use three... the fourth screw conflicts with the mobo heatsink. I don't think there is anything to put above the gold washers... I tightened it pretty good.
Yep.
No, not really.
Bingo, yep that's correct.
I did mount it with the side step over the lever. I didn't let it run for 8-10 hours... what are you talking about? The d5 is connected to the center barb. Assembly does match the guide.[/QUOTE]
It does, doesn't it? It's just crowded with cables. The Powerstream does include quite a bit.
ribuck97 said:
a. Yes, the stacker already came with a cut bottom, I just removed the fan grille.
b. I have one pushing and the other one pulling.
I can only use three... the fourth screw conflicts with the mobo heatsink. I don't think there is anything to put above the gold washers... I tightened it pretty good.
Yep.
No, not really.
Bingo, yep that's correct.
I did mount it with the side step over the lever. I didn't let it run for 8-10 hours... what are you talking about? The d5 is connected to the center barb. Assembly does match the guide.[/QUOTE]
It does, doesn't it? It's just crowded with cables. The Powerstream does include quite a bit.
Well, I don't think I've ever seen a person do a push and a pull on the same side of a radiator... doesn't mean it wont work.
The 8-10 hours... With the swifty6000 series mounted on a 775 you end up with the outlet barb below the inlet barb. If you leave the block in your case mounted vertically and since the cpu block is the highest part of your loop, all of the air bubbles when you bleed the system will collect in the top portion of your block. I wish I could draw a diagram but I'll exaggerate a little and tell you that it is possible that all of the surface area above the inlet is nothing but air. I would think that a waterblock that has air trapped within it is worse than a copper heatsink (similar to a thermalright) with no fan attached. Understand?
It also looks like your reservoir is only 2/3rds full? I don't use a reservoir, maybe someone else can comment but I thought it should be filled with little to no air...
The 8-10 hours... With the swifty6000 series mounted on a 775 you end up with the outlet barb below the inlet barb. If you leave the block in your case mounted vertically and since the cpu block is the highest part of your loop, all of the air bubbles when you bleed the system will collect in the top portion of your block. I wish I could draw a diagram but I'll exaggerate a little and tell you that it is possible that all of the surface area above the inlet is nothing but air. I would think that a waterblock that has air trapped within it is worse than a copper heatsink (similar to a thermalright) with no fan attached. Understand?
It also looks like your reservoir is only 2/3rds full? I don't use a reservoir, maybe someone else can comment but I thought it should be filled with little to no air...
If you are using only three mounting screws on the Maze 4 then you don't have even pressure over the entire GPU. The mounting hardware for the 6600 and under GPU uses only two screws. I would think in your case that using two mounting screws in opposite corners of the block would be a better situation than using only three of the four mounting screws which tends to tip the water block to one side. The best of course would be to figure out a way to use all four mounting screws. What ever you end up doing be sure to tighten down the screws evenly so the the pressure from the spring on each screw is even.
I agree with CJ on the mounting of the video block, three screws mounting the block is a bad idea.
What you could try is to find a nut that is thinner than the provided gold mounting hardware and try that... You also might have to dremel the mounting screw to make it shorter.
What you could try is to find a nut that is thinner than the provided gold mounting hardware and try that... You also might have to dremel the mounting screw to make it shorter.
Also, depending on how your video card is made, using 3 screws leads to a potentially hazardous situation of putting too much pressure on one corner of the GPU, thus damaging your chip.
Kinda like cracking one of the corners off of the old style chips without the heat spreader.
That <<could>> explain why your GPU is acting very strange.
Kinda like cracking one of the corners off of the old style chips without the heat spreader.
That <<could>> explain why your GPU is acting very strange.
- Joined
- Dec 5, 2001
- Location
- Canada/was England
whozyodaddy said:
Ask him why do they watercool cars and motorbikes ?
maybe ask him how does a bumble bees fly,
the body is too big for the wing area
Engineers cant work out how they fly but they do LOL
I dont work or ever worked for any watercooling company nor do any of my friends
but its funny how my set up works tho eh
all this coming from a British Tool and Die making Engineer LOL
Sounds like the Graphics card is the problem
you could have damaged it on your first install (dont ya just hate that )
nice birds nest of wires too
I hope you get your system worked out
Borgy
Last edited:
OP
- Joined
- Mar 6, 2005
- Location
- California
- Thread Starter
- #118
My reservoir is full, when the picture was taken it was not.
I am only using two screws. The only way for me to use all four would be to remove the mobo heatsink. I am gonna bleed the system again.
I am only using two screws. The only way for me to use all four would be to remove the mobo heatsink. I am gonna bleed the system again.
OP
- Joined
- Mar 6, 2005
- Location
- California
- Thread Starter
- #119
Just bled it again, same problems. Darn.
One thing to test, can you move the rheostat on your pump while its on changing the settings from 5 to 1 and back to 5? You should (I hope) be able to tell in your reservoir the flow rate change of the pump. If you don't notice a flow rate change, I would say you might have a faulty rheostat on the pump that is causing it to run at its lowest setting, thus causing your extremely high temps.
