Go Back   Pro/Forums > Site News and Blogs > ProCooling workBlogs > pHaestus's WorkLog
Password
Register FAQ Members List Calendar JavaChat Mark Forums Read

Reply
Thread Tools
Unread 12-09-2003, 09:46 AM   #1
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default Let's talk about waterblock reviews...

Ok guys I should be starting testing off this week or early next (whenever the chiller gets here) and I believe that I have testing methods in place that are acceptable:

Days 1 and 2: 10-15 mounts of the waterblock with flow rate held constant at 1.00GPM and one hour equilibration time to get a good feel for reproducibility of results, quality of mounting hardware, etc. These numbers will give me an average to normalize the below tests to as well. These tests will get me acquainted with the wb mostly but should be useful to you in that you can see the variance in wb mounting.

Days 3-5 would be the more interesting tests for you all. Performance vs. Flow rate measured at 0.5, 0.75, 1.0, 1.25, 1.5, and 2.0 GPM flow rates. Three replicates run and the results scaled so that they are equal to the average from the 1.00GPM tests above at 1.00 GPM. Tests are run from 6pm-11pm with one hour equilibration time between every point. At 11pm the waterblock is removed, thermal paste is reapplied, and wb is remounted. Then at 6am the next day the system is turned on at 0.5GPM and left to run CPUBurn all day until that night's 6pm testing begins.

I KNOW that everyone wants to see pressure drop numbers (myself included) but that will have to wait until the new year when I can win an auction on ebay for the needed equipment.

But now the real question: What of review format? I enjoyed the way Bill asked manufacturers to contribute to the beginning of reviews by explaining their design goals and methods in his reviews; would you guys want to see that format discussed as well? I am planning to cover the following in review: Finish and machining quality, mounting hardware and installation, performance, and overall opinion on the block.

Am I missing anything? Any comments on format or testing, or anything else? When I get everything assembled then I will put up a short article on the test setup itself and then just refer to it in the individual reviews.
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 10:16 AM   #2
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

how do you intend to 'control' the clamping force ?
dare I say 'quantify' ?

measure the curvature of the mobo ?
jk jk
BillA is offline   Reply With Quote
Unread 12-09-2003, 10:19 AM   #3
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

I haven't come up with a better plan than "5 complete turns of a wingnut from when the wingnut touches the springs" that Cathar uses to achieve proper mounting pressure. Other bright ideas?
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 10:31 AM   #4
nikhsub1
c00ling p00n
 
nikhsub1's Avatar
 
Join Date: Jun 2002
Location: L.A.
Posts: 758
Default

Sounds good to me! I really miss Bill's style (and accuracy) of testing. You do realize the can of worms you are opening right? But someone had to do it, I'm just glad it's you! Anyway, if there is anything I can do, give me a shout.

PS- The 2 WW's should be there tomorrow.
__________________

*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*
E6700 @ 3.65Ghz / P5W DH Deluxe / 2GB 667 TeamGroup / 1900XTX
PC Power & Cooling Turbo 510 Deluxe
Mountain Mods U2-UFO Cube
Storm G5 --> MP-01 --> PA 120.3 --> 2x DDC Ultras in Series --> Custom Clear Res
"Artificial intelligence is no match for natural stupidity."
1,223,460+ Ghz Folding@Home
aNonForums
*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*'``'*:-.,_,.-:*
nikhsub1 is offline   Reply With Quote
Unread 12-09-2003, 11:06 AM   #5
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

I am counting on the fact that JoeC will review blocks his way MUCH faster than I do. So the bulk of mfgrs will prefer to send things his way. That leaves me to test at a slower speed in more detail and post one review a week or so. Hell I am currently booked up til February just with the blocks I have or that are en route.

It's only fair that people come criticize my methods now; I've been doing so with others for a long time without posting test results.

I MAY also run what I'd call "typical system" test: Chevette core, Panaflo M1A, 1/2" tubing, Eheim 1250. This wouldn't strictly speaking be needed if I had pressure drop data to correlate C/W and head loss, but in the meantime it should help stop super restrictive wbs from looking a lot better on my testbed than they actually would be in end user rigs...
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 11:12 AM   #6
jaydee
Put up or Shut Up
 
jaydee's Avatar
 
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
Default Re: Let's talk about waterblock reviews...

Little confused here. If days 1 and 2 are going to have all the remounts then how are we to know the mount for days 3-5 are any good? Hummm, I guess by mounting it 10-15 times you will probably have a good idea if it is mounted well or not.

Last edited by jaydee116; 12-09-2003 at 11:53 AM.
jaydee is offline   Reply With Quote
Unread 12-09-2003, 11:24 AM   #7
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

Well jaydee currently I am mounting 5 times and running flow tests days 1-5. The problem with this is what if I get an especially good mount on day 5 (it's happened). Say temps drop 1-2C? Then do I try to figure out what happened and test for ANOTHER week?

My thought was that if you have a lot of data points and take an average, that you can then scale your later tests to that average number from the big set of remounts.

Maybe that's not such a great idea though? To get input is why I posted this thread.
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 11:24 AM   #8
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

Quote:
Originally posted by pHaestus
I haven't come up with a better plan than "5 complete turns of a wingnut from when the wingnut touches the springs" that Cathar uses to achieve proper mounting pressure. Other bright ideas?
I have not found this to work well at all
the difficulty is establishing a similar 'contact' at all 4 points
-> would strongly suggest using a 4" dial caliper to actually measure the springs' compression
(use flat washers top and bottom)

jd
once one gets the hang of it, and some wbs can be a challenge initially, 2 mountings say it all - the rest is just validation
its all about technique, and the goop
since using "Artic BillA Goop" I rarely make multiple mounts any more
-> inspect the grease pattern, tells all (including lack of flatness)
BillA is offline   Reply With Quote
Unread 12-09-2003, 11:32 AM   #9
Since87
Pro/Guru - Uber Mod
 
Join Date: Sep 2002
Location: Indiana
Posts: 834
Default Re: Let's talk about waterblock reviews...

Quote:
Originally posted by pHaestus
Flow rate measured at 0.5, 0.75, 1.0, 1.25, 1.5, and 2.0 GPM flow rates.
Any chance your pump could do 3 gpm with 'reasonably' restrictive blocks? (White Water, Cascade)

If so, I'd prefer to see the 1.25 gpm point exchanged for a 3 gpm point.

Actually, to get really geeky, I'd like to see the following points:

0.5
0.707
1
1.414
2
2.828

(The log10 of each of these numbers is evenly spaced. Aside from myself and possibly Les, I doubt anyone would care about this though.)

Edit: Photographers with really fast lenses may notice the similarity to F-stop values.
Since87 is offline   Reply With Quote
Unread 12-09-2003, 11:41 AM   #10
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

At the moment I am using a big heatercore (1 pass) and the pumps I have can't do more than 2.3-2.5 GPM through the test loop. On Friday (fingers crossed) the lab chiller will be here and I can replace the res and heatercore with a lab chiller. That MAY be able to get me closer to 3GPM but I suspect that the flowmeter is making that difficult. I had left points above 2GPM out because my thought was that the vast majority of users would never make it out in that region and that posting performance out there might be a bit misleading (you KNOW those are the numbers people will look at).
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 11:55 AM   #11
jaydee
Put up or Shut Up
 
jaydee's Avatar
 
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
Default

Quote:
Flow rate measured at 0.5, 0.75, 1.0, 1.25, 1.5, and 2.0 GPM flow rates.
If you have the above points taken then can't you just make a graph with the curve and have a pretty good estimation of what it would do at 3GPM or any GPM for that matter?
jaydee is offline   Reply With Quote
Unread 12-09-2003, 12:33 PM   #12
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

a post on OCers of relevance to this topic

I ran to 3.0gpm because of the way Excel plots curves, the end portions are not so good
- in fact virtually all wbs show only slight improvement between 2 and 3gpm
BillA is offline   Reply With Quote
Unread 12-09-2003, 01:19 PM   #13
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

Sorta a slippery slope tho Bill. If you are going to test for more than one point then you might as well generate a complete curve (much easier way to present data). That's exactly what JoeC is trying to avoid though (time constraints I assume).

I still remain confused over whether he is actually testing performance at 1.00GPM or if he is testing at 1.00GPM freeflow (no wb) flow.
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 01:34 PM   #14
Since87
Pro/Guru - Uber Mod
 
Join Date: Sep 2002
Location: Indiana
Posts: 834
Default

Quote:
Originally posted by jaydee116
If you have the above points taken then can't you just make a graph with the curve and have a pretty good estimation of what it would do at 3GPM or any GPM for that matter?
Yes, but I'm lazy. Having data for those points specifically let's me see some things at a glance. (Like how 'normal' the PQ curve for the block is. With a 'normal' PQ curve, the pressure drop at each of those flowrates would be twice the pressure drop at the next lower flowrate.)

Like I said, not something that most people would care about.
Since87 is offline   Reply With Quote
Unread 12-09-2003, 01:37 PM   #15
RoboTech
Cooling Savant
 
RoboTech's Avatar
 
Join Date: Sep 2002
Location: Cincinnati, OH
Posts: 229
Default

Sounds like a solid starting point to me...

1) Flow - I too would like to see flow go up to 3 GPM on the high end (but understand the limitation$)
2) Delta P across block - very useful data. How about a simple water manometer (instead of digital press gauges or digital Delat P meter)? Easy to make, inexpensive, with good repeatability and accuracy.

What does your flow loop look like pH? (order of things - particularly how are you planning to integrate the chiller?)
RoboTech is offline   Reply With Quote
Unread 12-09-2003, 01:38 PM   #16
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

Only 0.01GPM resolution also Since87 so you'd be off by a bit.
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 01:40 PM   #17
Since87
Pro/Guru - Uber Mod
 
Join Date: Sep 2002
Location: Indiana
Posts: 834
Default

Quote:
Originally posted by unregistered
a post on OCers of relevance to this topic

I ran to 3.0gpm because of the way Excel plots curves, the end portions are not so good
Good point. I hadn't thought about how the interpolated curve would be distorted.
Since87 is offline   Reply With Quote
Unread 12-09-2003, 01:44 PM   #18
BillA
CoolingWorks Tech Guy
Formerly "Unregistered"
 
Join Date: Dec 2000
Location: Posts: 2,371.493,106
Posts: 4,440
Default

what I meant to say on OCers re JoeC's testing is that the single datum should be based on the wb's connection size
- the assumption being that an Innovatek XX using 1/4" ID tubing would not be used in a 1/2" ID (low flow restriction) high flow system

JoeC is going to test it at 1gpm which makes no sense at all considering the dp @ 1gpm is 1.417psi, 1.0mH2O, a rather high number

I am under the impression that all JoeC's data is generated at 1gpm,
then he sets the system flow at 1gpm w/o a wb, and then inserts the wb and records the change in flow
- obviously the 'flow reduction data' is not comparable between different pumps

it is a mess the way it is poorly described and presented
BillA is offline   Reply With Quote
Unread 12-09-2003, 02:30 PM   #19
Cathar
Thermophile
 
Cathar's Avatar
 
Join Date: Sep 2002
Location: Melbourne, Australia
Posts: 2,538
Default

Quote:
Originally posted by unregistered
I am under the impression that all JoeC's data is generated at 1gpm,
then he sets the system flow at 1gpm w/o a wb, and then inserts the wb and records the change in flow
- obviously the 'flow reduction data' is not comparable between different pumps

it is a mess the way it is poorly described and presented
Yes, that's my understanding of the test procedure as well.

What I find awry is the conclusions, especially for the RBX. JoeC tested the RBX with a higher-pressure pump, but then goes and artificially restricts the flow rate anyway to 1GPM.

To then make the conclusion that the RBX (and presumably this would apply to all waterblocks of similar pressure drop) then requires a high pressure pump to work well, strikes me as odd.

Had JoeC stuck with using the 1048, and measured the performance and the RBX was still shown to be in front, then that totally disproves his statement. Since he did not test with the 1048, DangerDen are now left with the scenario that one of the testers that the masses look to for guidance has just laid out the statement that large pumps are required for higher pressure drop blocks to work well, all on the basis of a single data point using a pump that is not the same as the other tests.

I know what score I'd get on my physics practical class if I ran my tests in the same manner and attempted to draw similar conclusions...
Cathar is offline   Reply With Quote
Unread 12-09-2003, 11:12 PM   #20
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

Ok so two WWs should be here tomorrow. Dangerden RBX will be here Dec 11th. Chiller is SUPPOSED to be here the 13th. I already have a Cascade and a MCW5000-A. That's a lot of testing to do

Any preference as to first wb to test? I was thinking Cascade, but that's just because I want to see how it really does myself...
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 11:21 PM   #21
jaydee
Put up or Shut Up
 
jaydee's Avatar
 
Join Date: Dec 2001
Location: Spokane WA
Posts: 6,506
Default

Quote:
Originally posted by pHaestus
I was thinking Cascade, but that's just because I want to see how it really does myself...
Yes.
jaydee is offline   Reply With Quote
Unread 12-09-2003, 11:42 PM   #22
satanicoo
Cooling Savant
 
satanicoo's Avatar
 
Join Date: Nov 2002
Location: portugal
Posts: 635
Default

Please. And RBX after .
satanicoo is offline   Reply With Quote
Unread 12-09-2003, 11:44 PM   #23
Cathar
Thermophile
 
Cathar's Avatar
 
Join Date: Sep 2002
Location: Melbourne, Australia
Posts: 2,538
Default

Quote:
Originally posted by pHaestus
Any preference as to first wb to test? I was thinking Cascade, but that's just because I want to see how it really does myself...
How would testing it first allow you to know how well it does (comparitively) if there's nothing else that has yet been tested on the test bed to compare it to?
Cathar is offline   Reply With Quote
Unread 12-09-2003, 11:48 PM   #24
pHaestus
Big Player
Making Big Money
 
pHaestus's Avatar
 
Join Date: Aug 2001
Location: irc.lostgeek.com #procooling.com
Posts: 4,782
Default

Cathar: Just because it is a first review with chiller and current setup doesn't mean that I have never tested any other blocks for my own information on previous iterations. I know roughly CPU wattage so I can take a fair stab at C/W. And always of interest are C/W vs. flow rate curves.
__________________
Getting paid like a biker with the best crank...
-MF DOOM
pHaestus is offline   Reply With Quote
Unread 12-09-2003, 11:50 PM   #25
Cathar
Thermophile
 
Cathar's Avatar
 
Join Date: Sep 2002
Location: Melbourne, Australia
Posts: 2,538
Default

Quote:
Originally posted by pHaestus
Cathar: Just because it is a first review with chiller and current setup doesn't mean that I have never tested any other blocks for my own information on previous iterations. I know roughly CPU wattage so I can take a fair stab at C/W. And always of interest are C/W vs. flow rate curves.
Sorry, I really should've put a smiley there. The statement wasn't meant to be taken too seriously.
Cathar is offline   Reply With Quote
Reply


Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 
Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


All times are GMT -5. The time now is 03:59 AM.


Powered by vBulletin® Version 3.7.4
Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
(C) 2005 ProCooling.com
If we in some way offend you, insult you or your people, screw your mom, beat up your dad, or poop on your porch... we're sorry... we were probably really drunk...
Oh and dont steal our content bitches! Don't give us a reason to pee in your open car window this summer...