First attempt at water cooling...FAIL!

[intel igent]

using 1" tubing is very impractical for computer cooling and would prolly net worse performance. personally i think 7/16" is the way to go and will be what i use next time around

 

[imperialreign]

using 1" tubing is very impractical for computer cooling and would prolly net worse performance. personally i think 7/16" is the way to go and will be what i use next time around

Im just curious here - but why do you think going slightly smaller than 1/2" would be better?

 

[intel igent]

i didnt say it was better i said i PREFFERED it to 1/2" tubing. i use 1/2" tubing now and i think you get a cleaner overall look with the slightly smaller tubes and flowrates are still optimal. best of both worlds IMO.

neways ne updates?

 

[Wile E]

I'm happy with 3/8", personally.

Are those the new DD TDX blocks, Dr.S? If so, I can't wait till you get running, I want to see how good they are.

 

[sneekypeet]

Something else I noticed in those pics....the screws are supposed to be tightened until the springs are in full compression as per the DD video instructions. Prolly wont hurt it for booting purposes but it will deff. affect your temps!!!!


http://www.dangerden.com/index.php?option=com_content&task=view&id=55&Itemid=42

 

[intel igent]

I'm happy with 3/8", personally.

Are those the new DD TDX blocks, Dr.S? If so, I can't wait till you get running, I want to see how good they are.

pretty much no performance difference between 3/8" - 1/2" tubing. all comes down to preference really.

looks like the new DD blocks.

Dr.S does it look like they left room for adding the nozzles like on the old TDX's?

Something else I noticed in those pics....the screws are supposed to be tightened until the springs are in full compression as per the DD video instructions. Prolly wont hurt it for booting purposes but it will deff. affect your temps!!!!


http://www.dangerden.com/index.php?o...55&Itemi d=42

good eye

 

[Dr. Spankenstein]

*UPDATE*
Bonehead Alert! Bonehead Alert!

One MAJOR screw-up was the fact that I didn't check the contact between the CPU and the block. There was about 1/4" of contact along the right edge and that was it! Come to find that the reason for the bad contact was the copper block was resting on one of the aluminum caps around the socket! Double !
I haven't had a chance to reseat and check because I'm at work now. Looks like my layout is about to change a little since the block has be be aligned vertically now.

I spent the good part of this afternoon with my bottle of lab-grade acetone and a bazillion q-tips cleaning up my video card. Think I discovered a couple chunks of Arctic Silver adhesive (that I had mixed 50/50 with AS5 for the ramsinks) that had wedged themselves under the ram chips and up against capacitors on the card.

@ Wile E & intel igent,

Yes, these new DD blocks are a sight to behold. I don't think there is room for another nozzle (ala RBX, did you mean?). They refer to the CPU block as the MC-TDX (MC for multi-core) though I'm not sure I'm clear on how these are "optimised" for multi-core. The chipset is the new design applied to the 975 Maze4. The thing I like most about them is the non-tarnishing finish they have applied. Man, my Tyee block develops tarnish if I even look at it!

Does anyone know if the CPU block touching a cap would actually short something to the point of it not booting? I mean, I know it can't be a positive thing!

I am going to take the advice of having my ram cooling plugged into the CPU header when I put it back together.

(Boy, am I glad I pre-ordered Crysis and picked it up the evening of the 15th! It's been sitting here, mocking me....constantly!)

Thanks again to everyone for their helpful suggestions and comments! Don't stop giving me ideas, cuz this probably won't work....

 

[sneekypeet]

SHIT....never even looked that close when I checked your spring tension....the block IS 180* out ...the pointy end iof the block goes to the right, not in a vertical alignment....Expect an edit I will get a pic of mine as its running now!

See how the DD is on the left of the CPU.

 

[intel igent]

*UPDATE*
Bonehead Alert! Bonehead Alert!

One MAJOR screw-up was the fact that I didn't check the contact between the CPU and the block. There was about 1/4" of contact along the right edge and that was it! Come to find that the reason for the bad contact was the copper block was resting on one of the aluminum caps around the socket! Double !
I haven't had a chance to reseat and check because I'm at work now. Looks like my layout is about to change a little since the block has be be aligned vertically now.

I spent the good part of this afternoon with my bottle of lab-grade acetone and a bazillion q-tips cleaning up my video card. Think I discovered a couple chunks of Arctic Silver adhesive (that I had mixed 50/50 with AS5 for the ramsinks) that had wedged themselves under the ram chips and up against capacitors on the card.

@ Wile E & intel igent,

Yes, these new DD blocks are a sight to behold. I don't think there is room for another nozzle (ala RBX, did you mean?). They refer to the CPU block as the MC-TDX (MC for multi-core) though I'm not sure I'm clear on how these are "optimised" for multi-core. The chipset is the new design applied to the 975 Maze4. The thing I like most about them is the non-tarnishing finish they have applied. Man, my Tyee block develops tarnish if I even look at it!

Does anyone know if the CPU block touching a cap would actually short something to the point of it not booting? I mean, I know it can't be a positive thing!

I am going to take the advice of having my ram cooling plugged into the CPU header when I put it back together.

(Boy, am I glad I pre-ordered Crysis and picked it up the evening of the 15th! It's been sitting here, mocking me....constantly!)

Thanks again to everyone for their helpful suggestions and comments! Don't stop giving me ideas, cuz this probably won't work....

that couldve done it! hopefully after the re-seat she'll be good to go.

within regards to the nozzles; no, not like the RBX. the old RBX was 3 BARBS, the old TDX was 2 BARBS but had "accelerator nozzles" that could be purchased separately that went in the inlet port of the block. same idea as what D-tek is now using except the DD models were like a plate vs the D-tek versions which appear to be cylindrical in shape. layout doesnt necessarily have to change you just need to rotate the block

 

[sneekypeet]

that couldve done it! hopefully after the re-seat she'll be good to go.

within regards to the nozzles; no, not like the RBX. the old RBX was 3 BARBS, the old TDX was 2 BARBS but had "accelerator nozzles" that could be purchased separately that went in the inlet port of the block. same idea as what D-tek is now using except the DD models were like a plate vs the D-tek versions which appear to be cylindrical in shape

The issue with his block accepting the accelerator nozzels is this...mine is an older AMD conversion with a 775 top screwed on it. It internally has a square hole to accept the nozzels, which by looking at the new design with all the pins and no "holder" for the accelerator, I'd have to lean towards a no on if they will accept them!

http://forums.techpowerup.com/showthread.php?t=44253&highlight=danger+Den

Look in picture #2 see the difference in design....no holder just lowered pins.

 

[nflesher87]

glad to hear you most likely figured this out Dr. S!

I must say one thing though, and correct me if I'm wrong anyone, but I believe Q-Tips aren't lint free, making them a horrible choice for cleaning up TIM
my recommendation:
- Coffee Filters
- anything Microfiber

I hope the reseat straightens everything out!

 

[sneekypeet]

I wont correct you , because you are correct....I use them to remove the bulk as they work great and are "cleaning pads on a stick" (well designed for using). Then I use filters myself as a final clean b4 reapplying TIM.

 

[nflesher87]

I wont correct you , because you are correct....I use them to remove the bulk as they work great and are "cleaning pads on a stick" (well designed for using). Then I use filters myself as a final clean b4 reapplying TIM.

fair enough

 

[intel igent]

@ sneekypeet : from what i remember (its been a long time since i looked at a TDX) the holder for the nozzle/plate (whatever they called it) was in the top itself (under the inlet barb) not on the inside of the block.

am i understanding ur post correctly?

ive always used qtips. just made very shure everything was extremely clean and dry before re-assembly

 

[Dr. Spankenstein]

SHIT....never even looked that close when I checked your spring tension....the block IS 180* out ...the pointy end iof the block goes to the right, not in a vertical alignment....Expect an edit I will get a pic of mine as its running now!

See how the DD is on the left of the CPU.

Hrrmmm...so is your inlet on the right or left? I would have though you wanted the inlet to be centered over the pins (and subsequently the CPU) to get the turbulence and pick up the most heat.

i r counfuzed!

@ intel igent,

Oh, I got you now! I don't believe this model utilizes the accelerator nozzles, but I'm not sure.

More research...

 

[sneekypeet]

@ sneekypeet : from what i remember (its been a long time since i looked at a TDX) the holder for the nozzle/plate (whatever they called it) was in the top itself (under the inlet barb) not on the inside of the block.

am i understanding ur post correctly?

ive always used qtips. just made very shure everything was extremely clean and dry before re-assembly

I think we are on the same page here and I just pulled my old 939 top out of the box that used to be on the block Im using in my 775 setup in the pic above....its just flat clear plastic with 2 holes for fittings and 2 holes for mounting, no dents or anything to place the nozel.


EDIT #2 this is my block...note the square hole for the accelerators.....
http://www.madshrimps.be/?action=getarticle&articID=260

EDIT @ Dr. Spank....I have an older TDX block, just the shape is the same...im gonna google some pics to solve both intel's and your questions!

 

[intel igent]

doesnt look like there will be provisions for accelerator plates/nozzles. man its been sooooo long since ive actually looked at one of those blocks

maybe next time around?

im off for happy hour

 

[Dr. Spankenstein]

Toss a couple back for me, will ya?!?

 

[Dr. Spankenstein]

*2nd UPDATE*

I'm up and running again. It was nothing more than reseating the block (without it shorting a cap!)

Again, I do appreciate how everyone offered their help!

Now, I'll open up the floor to questions or benches you would like results from....

HERE WE GO!!

 

[intel igent]

Toss a couple back for me, will ya?!?

done. And done

*2nd UPDATE*

I'm up and running again. It was nothing more than reseating the block (without it shorting a cap!)

Again, I do appreciate how everyone offered their help!

Now, I'll open up the floor to questions or benches you would like results from....

HERE WE GO!!

Nice!

what kinda temps ru u seeing? b4 L/C + after?

 

[Dr. Spankenstein]

Here's the quick and dirty:

Ambient temp. 69F (20.6C)-from thermostat (I know, not the most scientific. I'm just jazzed this thing is running, so cut me a little slack!)

TAT running 100% load for 10 min. (3000Mhz @ 1.30Vcore)
On my Tuniq 120: 21/19 idle 58/56 load
On water: 25/22 idle 45/44 load

ORTHOS small FFT's @ 6 priority for 10 min. (3000Mhz @ 1.30Vcore)
On Tuniq 120: 21/19 idle 52/50 load
On water: 25/22 idle 39/39 load

ATI Tray Tools Artifact Scan for 5 min. (722/999 @ 1.60 core 2.1025 mem)
On HR-03 w/Noctua 120: GPU 58-60 Env. 44
On water: GPU 40 Env. 37

I was also able to make it through an entire 3DMark 06 run with the card @ 735/999. This was impossible before. Can't wait to push it!

In short, I am amazed at how it keeps the temps down. they shoot up fast but seem to hit a "ceiling" that it is hard-pressed to break though. On air it was as slow and steady rise to much higher temps.

I love it!!

 

[intel igent]

nice to see its all working well for ya.



edit: in case i missed it what does your whole loop consist of? thnx

 

[Dr. Spankenstein]

Loop parts list:

DD MC-TDX, DD 975 Maze4 chipset, DD Tyee, Liang D5, BIX 120.3 and Primoflex UV blue tubing and tube clamps. T-line, no res. Distilled w/PT Nuke.

I have no idea what temps to expect, but I am happy for now. I more than likely put to much MX-2 on the CPU and I need to shorten some tubing to relieve some pressure on fittings but pleased in general. The Primoflex is extremely easy to work with. It's hard to even make it kink!

Please don't dog me for the cables! It's still a WIP!

 

[sneekypeet]

hey Dr. your CU block is still on incorrectly, as I posted previously it should still line up as mine does in the pic I posted to ya. Also look at the link to Danger Den so you can watch the install video.

Just my 2 pennies.

 

[Dr. Spankenstein]

Yeah, I'm just not going to be able to stand having my tubes cris-cross over the block (meaning the flow from the rad will have to go to the left side of the block then the outlet on the right will have to veer left to meet either the GPU or NB block). I'm still of the belief that you should be having the inlet over the pins, so unless I had the rad on the back (or on top) of the case, I kind of hosed (pun) myself with my layout.