Fuzion V2 Nozzles Compared - Core 2 Quad

[[XC] riptide]

http://www.xtremesystems.org/forums/...d.php?t=202394

Added to the Info/Guides & Review/Tests master thread. It takes up a lot of diff. factors. I had to think about where to put the entry.

[Vapor]

Thanks Rip

Hopefully when I get all the nozzles done (and more blocks added) I can put it in the CPU block section

p.s., guys....4.5mm nozzle done, doing the 6.5mm next (takes 2-3 days), then I'll repost graphs (I do graphs as I work, so there shouldn't be any wait). Then onto the last two (would consider forgoing the 3.6mm one if 4.5mm is 'worse' than 5.5mm, but I want to verify functionality of my setup at very, very low flow)

[Sadasius]

Thanks for the testing Vapor. Well done!

[[XC] riptide]

Moar?!

[Vapor]

Updating with three more nozzles now (6.5mm, 5.5mm, 4.5mm) but am removing the Quad Insert data.

Wanted to reinvestigate after getting appreciably higher flowrates than expected (I really really don't think it matches the stock insert for flowrates at a given pumping power)...and also noticed that the nozzle doesn't have the o-ring in it (though I honestly can't recall if it did or not during my tests). I've installed the o-ring and am running again. Getting lower flowrates (good thing, hints that the o-ring was not installed previously) but will have no temp data for a couple days. Will also do the 3.5mm nozzle after this (maybe--might just save it for when I go to i7).

EDIT: okay, new graphs are up Again, make what you will of the data...seems to me the single-jet nozzles just don't do much on a dual-die CPU The thermal data is basically a wash...

[gOtVoltage]

Updating with three more nozzles now (6.5mm, 5.5mm, 4.5mm) but am removing the Quad Insert data.

Wanted to reinvestigate after getting appreciably higher flowrates than expected (I really really don't think it matches the stock insert for flowrates at a given pumping power)...and also noticed that the nozzle doesn't have the o-ring in it (though I honestly can't recall if it did or not during my tests). I've installed the o-ring and am running again. Getting lower flowrates (good thing, hints that the o-ring was not installed previously) but will have no temp data for a couple days. Will also do the 3.5mm nozzle after this (maybe--might just save it for when I go to i7).

EDIT: okay, new graphs are up Again, make what you will of the data...seems to me the single-jet nozzles just don't do much on a dual-die CPU The thermal data is basically a wash...

Wow great results,, Youve been busy


Hmm Single jet Nozel for CORE 2Duo or Phenom!! should show a nice differance though? Remeber the Quad Nozel aims at the two seperate cores and low from Center of IHS. A single Nozeel will Aim Center or High on Intel Quad and be uneven coverage no matter what.

..Maybe test the Single Nozzels with a Dual Core Intel or even better a Phenom!! This would have all Cores in the middle(Close to center).
Could be fun

[skinnee]

Good work Vapor. Don't you just hate it when you stumble upon something and have to retest a huge chunk? Drives me nuts, always happens when I try to work faster than I should. Excellent work though,

One small request, could we get a data table to go along with the charts?

[Vapor]

Alright....a compilation table is up under the Big Chart and individual tables are in their own new section now

Thanks for the feedback everyone

(it'll pay off and be more interesting when there's more blocks being tested, I swear! )

[skinnee]

Thanks Vapor!

The tables help find the intersects of the data points once the charts get populated, or maybe I'm crazy and need a guide for following charts. The more I think about it, the more I think its the crazy.

[Vapor]

Another way of representing the same data, so might as well

EDIT: my one big grudge against tables is this: people like to split hairs. For instance, the data says that with 1 pump, the 6.5mm nozzle has .21C lower temps than the 4.5mm nozzle--but outside of the change in flowrate, they're really a tie. Considering the DTS sensors, that's within margin of error on my testbed.

[xeon_1]

I have a question.
Would the quad core nozzels also improve or atleast not hurt the temps on a dual core cpu?

[lazy]

nice work Vapor

[AndreaBZ]

A really good work!
About the fuzion v2 stocks (single WB which I have data for comparison)
The accuracy is very high!

I place some graphs where temperatures have been brought to the level of MLL (Martin liquid labs)
(difference temperature core-temperature water).

D-Tek Fuzion V2 Stock VAP (Vapor) (blue)
D-Tek Fuzion V2 Stock reported in spreadsheet (green)



D-Tek Fuzion V2 Stock VAP ** = while neglecting the data to 2.12 GPM (blue)
D-Tek Fuzion V2 Stock reported in spreadsheet (green)
Almost indistinguishable !



Overview

D-Tek Fuzion V2 Stock reported in spreadsheet (blue)
http://www.xtremesystems.org/forums/...d.php?t=206542
D-Tek Fuzion V2 Stock VAP (red)
D-Tek Fuzion V2 Stock VAP ** (green)
V2 Stock MLL The Martin data on 2.26 GPM.




v2 + nozzle :

[SpuTnicK]

Is quad nozzle and this baby :
http://www.performance-pcs.com/catal...ducts_id=23987
are the same? If not is there any data on how well it performs?

[Martinm210]

Is quad nozzle and this baby :
http://www.performance-pcs.com/catal...ducts_id=23987
are the same? If not is there any data on how well it performs?

I'm not sure if the Quad Noz Vapor tested is actually the red nozzle insert that you buy in the V1 kit or the Quad insert that you linked. I suspect they would perform similarly...but I don't know. When I played around with the quad insert on my Q6600, I was finding the stock setup without any nozzle was working best, but my testing wasn't nearly as precise or good as Vapor did here. My mounting inconsistency made it hard to find anything conclusive.

[Vapor]

I tested both but left the black insert's results out of my write-up for a few reasons. Mainly, it performed horribly.

The red nozzle performed really well, easily being the best nozzle on my QX6700 (though the difference was small). But the black insert was just way behind all of them. At first I thought it was user error, but I retested and rebuilt numerous times and just couldn't get comparable performance out of it (considering all the nozzles were basically within a degree of each other at any given flowrate or pumping power).

IIRC, it was 1-1.5C worse than the rest of the bunch. It really stuck out as being inferior, IMO. Maybe it's better on a Yorkfield.