Fuzion CPU middle to Top Block gap
FYI
Doing some CPU block testing so, I've had my fuzion sitting here on my work bench. I started looking more closely at the match between the middle and top derlin blocks to try a more agressive bow and I think I see a big part of why bowing helps so much on the Fuzion (It stops an internal leak).
There is a rather large gap between the middle and top block. Try putting the two blocks together, block the outlet of the middle block center tube and try to blow through the block, mine leaks like a sieve.
My first bow try was after creating a nozzle and I made about a 1/32" plastic washer. To my suprise if I remove the outer gasket to check clearances that thin washer barely even closes the gap that's there. I ended up gaining almost 3C improvement simply by using this very thin gasket which didn't bow it very much, I think alot of that gain is simply by sealing and preventing the inlet water from bypassing the base:up:
This is how the mid block looks with a straightedge across the top (almost a 1mm gap).
http://www.overclock.net/gallery/dat...pMidBlocks.jpg
And this is the thin washer I made, but any sort of thin o-ring or gasket material cut to fit would help alot with this.
http://www.overclock.net/gallery/dat...nBowWasher.jpg
Update: The proof is in the "Play-Doh"...:D Right at 1.0mm of gap remaining after assembling the block:
http://www.overclock.net/gallery/dat...ksPlay-Doh.jpg
Update2: 1/16" O-Ring and the bow that you get:
http://www.overclock.net/gallery/dat...wingO-Ring.jpg
Straightedge with O-ring:
http://www.overclock.net/gallery/dat...ingResults.jpg
Testing with the gap filled
Hey Martinm,
I can see your headed towards full testing of the Fuzion with your washer and nozzle in place. I have not had a chance to publish these results here in a more current Showdown at High Noon article yet (I just finished grad school and have spent the last two weeks moving a Penske truck across the country to California), but my results (with the gap filled) were slightly different:
Q6600 (at 3.2 GHz) with MicroRes/MCR 220/Fuzion/MCW60/MCP 655 and 1/2 inch tubing as follows:
Load (with the gap unfilled - stock) - 42.3 deg C with 1.83 gal/min flow
Load (with the gap filled - 0.5 mm washer) - 40.8 deg C with 1.93 gal/min flow
Load (with gap filled and custom nozzle) - 39.7 deg C with 1.48 gal/min flow
Hope this helps out Martinm,
Jay