who wants to make one of these out of brass???? i might
Printable View
who wants to make one of these out of brass???? i might
Wiki is usually a pretty reliable source of information.
Yes, anyone can edit it, but any changes usually need to be approved before they become permanent.
gosh, those xspc products are kickin!
Good work!
Was a backing plate used in all the tests? Is there any measure of the clamping force?
It would be interesting to see the XSPC epoxied so water only flowed over the hot. The restriction may go up, however, I wonder if the temps would go down.
Have you tried the 3.6mm nozzle?
Update:
Saw the use of the backplate on your site and how you tightened it. Without a spring (where measuring the compression would tell you the load) torquing is the only way to get consistant results. Small differences in sensitivity in your finger tips, minute debris on the threads, etc could have caused differences. What is interesting is that this is a cross flow design beating the impact designs, presubably without bowing. This may require a step on the base plate to get the bowing effect.
Quote:
Originally Posted by vkbms
I did mention the question about alternative top materials. I believe they are considering that and the more people request that, the better.:up:Quote:
Originally Posted by p2501
Yeah, backplate used in all tests, and not I didn't measure mounting pressure in any way other than feel. I should probably use a spring to ensure consistency there, I just like hard mounting by preference and use a good nylon washer and the same brass mounting nut for each block so they all have the same friction and feel. But I agree a spring would probably help with pressure consistency in pressure, but I'm not sure it's anywhere near as significant as the off axis issue with four mounting bolts.Quote:
Originally Posted by DavidNJ
I'm still thinking that until someone comes up with a single screw type mount apply pressure directly over the block a big part of the inconsistency is tightening the four bolts the same distance. There is nothing in the four bolt system to prevent an off axis mount and that's probably the biggest factor in mount variability. Practice is the only thing that helps with that and even with that once bowed, mounting consistency goes down the tubes as every little of change in axis rotation rolls the contact patch around and it's impossible to feel and get spot on every time.
In the end the only way to really minimize mounting inconsistencies is to mount multiple times. This can get you down from a 2 degree error to about half a degree using 5 mounts, 10 mounts might get you to a quarter of a degree. Probably worse yet is the accuracy of the processor diodes. Only way to avoid that is with a separate IHS mounted sensor or better yet...a die simulator. But then you have another set of issues with trying to correlate results to an actual processor.
The real problem lies with the fact that all the gains anymore in block designs are very small, and approaching the limit we can even test the differences with a worthwhile level of confidence.
Martin, I just finished my first official run of the XSPC block, have 4 more to go but, it is so out of whack with your results... I mean, not even a little close in terms of relationship to the Fuzion.
Edit, actually Martin, looking at your graph the D-Tek block looks weird. My nozzled and bowed fuzion has not been touched by anything yet. I think perhaps the EXTRA bow I gave to the fuzion is the difference. Martin, you have to do this now, it's gonna suck, but you must! Can you take an EK barb o-ring and use that between the mid and top plate of the V1 Fuzion. This is how I got best results! It bows the block more.
Yeah, I'm going to try with an o-ring and bowing next, my test was run with the kit washer only, no o-ring for bowing. Actually I'm going to switch over to the new V2 fuzion, the bases are different...not the same, so I suspect V2 will be better. The V2 base is thinner and the pins are shorter. This is why it's showing as more restrictive that a washer only V1. D-Tek knows what they are doing, so I'm sure the changes are for the good in thermal performance.Quote:
Originally Posted by nikhsub1
You'll be a bit different anyway. You're IHS is stock, mine is lapped, so you're results will have a preference towards a bowed base especially if your particular IHS is concaved. I'm also using a better TIM compound. Bowing is an optimization to minize TIM thickness where it counts the most, so the better the TIM material the less bowing will have an impact. I ran several mounts with Artic Ceramique vs the TIM consultants grease 0098 and got a good 3C better results with the TIM consultants grease. Better grease will again not have so much preference on bowing.
I might try milling a step in both blocks and comparing that way too...not sure yet. I was curious to see your test results on the stepping vs bowing. Or I could just bow each block, o-ring on the fuzion and some sort of spacer on the XSPC block, but I would want to test bowed vs bowed. This first round was sort of a stage1 stock vs. stock, then stage 2 mod vs mod, next would be adding in the base bowing.
We also have different pumps, I've already heard people say they've seen as much as 2C with a stronger pump on the more restrictive blocks. The D5 is much weaker than an Iwaki at 18V, so you're results will have a stronger preference towards nozzled and higher restriction blocks than my testing with a D5. I'm also using a normal Hydrx solution as opposed to straight distilled, so I'll have a little more viscosity, slightly lower flow rates, slightly lower heat transfer of the fluid. Not sure the fluid matters that much, but I expect oiur results to be somewhat differet especially with the IHS lapping/TIM used/ and pump differences.
But it's all good, post your results, then people can compare how it might react with a more high end pump and stock IHS..:up:
Also have you noticed any creep in temperature on your processor. I was always curious if over time as the silcone degrades if there is any difference in testing results? Have you gone back to revalidate any tests to see if anything has changed over time?
It also says:Quote:
Originally Posted by C'DaleRider
Quote:
Originally Posted by wikipedia
I never do that. Wikipedia serves its purpose quite well as a quick and dirty and very imperfect source of information about almost everything. I once spent an entire day re-editing an offensively erroneous entry on wikipedia. But that was on a very controversial subject. A popular book had just been published with lots of misinformation. For noncontroversial topics I have found wikipedia to do an excellent job. No, it is absolutely not a valid original source. And could never be used in a research paper. But as long as everyone is aware of the uncertain nature of its data I don't see the problem with citing it in a web forum. On non-controversial topics like the chemical composition of a plastic I would expect better than a 80% ratio of correct to incorrect information. Are you saying that the wiki entry is wrong? I am not a chemist. So I have no way of knowing whether it really does have a similar chemical composition to acrylic. It only seemed plausible because iandh is a machinist I think and I figured he would know better than I about something like that.Quote:
Second, quit taking everything found on Wiki as gospel
Looks interesting but I too have concerns that the block is held down by a piece of acrylic and not metal. I would be worried about over tightening and cracking it.
Martin to your talk of the 4 corner mount I always wondered when the best would be pressure applide to the center of the block why someone does not make a mount that use the 4 screw holes to hold a frame over the block and have one big screw in the center like from a C-clamp pressing down there. You would have to mount the barbs off center to do it but I think you would have better pressure directed to the center of the CPU. I would fly in the face of all blocks mounting convention, but why not? Sounds like something I would try if my set of tool and shop toys was as extensive as yours. Maybe next toy will be a mill if I ever finish my current project.
:rofl:Quote:
Originally Posted by cegras
very true... however i was getting comfortable with micro pin layout. :P
I already have too many toys to play with. Right now im eyeing a Cuplex DI.
personally this block is looking like a good performer but i think the thick acryllic is ruining the looks :S
What's wrong with acrylic? I think it looks awesome :p:Quote:
Originally Posted by Giannis86
I can help ya out NaeKuh:Quote:
Originally Posted by NaeKuh
http://i82.photobucket.com/albums/j2...g/IMG_1223.jpg
@Broken, WHOA! :shocked::shocked: That looks sweet! :up:
Worklog somewhere?? :D
Not really, this was my 680i build and I now have a 780i so my Twinplex's don't work on the 780i so... I got some EK stuff and decided to get a Supreme to match, I know mental that I like stuff to match. Same thing with the build before the AC stuff. I had Alphacool XP Bold and two Alphacool chipset blocks that matched.
I am working on a G70 and will post some pics and a small work log in a little bit.
See, save some money and buy mine. That way you can feel better, and I will feel better...Quote:
Originally Posted by mcoffey
:p:
$75 shipped. Can't go wrong.
i have first calls on it. :D
Yeah i know, freya isnt going to be a high end performer like persephone is. I wanted to go on bling.
The chipset blocks, are those nickle platted copper as well? I might take all 3 from you. :]
LMP.
WOAH!Quote:
Originally Posted by Broken
Now that's a mans chair!
:rofl:
I want one, Anyone see them selling in the US yet?
That ought to work nicely with the RD. ;)Quote:
Originally Posted by NaeKuh
I think it looks awesome too, but I think a glass top would look even more awesome.Quote:
Originally Posted by [XC] NetburstXE
How were you going to bow this?
Plastic/rubber/neoprene or some sort of flat sheet to lay on top of the channels between the channels and the top.Quote:
Originally Posted by DavidNJ
Either that or put it on the mill and cut a step in the base. Leaning towards the first though.
Bowing with this TIM consultants grease and my lapped IHS doesn't seem to do very much though. The fuzion is already slightly bowed just by tightening down the bolt well. My V2 also has a slight bow.