Question for Cathar

Rayman2k2 · 02-22-2003, 06:47 PM

Would there be any difference in temps if you were to run your block without the middle piece? I am talking about the one that blocks the water and makes it come out faster.

Cathar · 02-22-2003, 08:16 PM

About a 2C difference if run without the nozzle plate.

Rayman2k2 · 02-22-2003, 08:32 PM

hrrrrrmmmmm......would there still be the same amount of flow resistance?

Cathar · 02-22-2003, 09:23 PM

Flow resistance is then very low. From my limited measurements it flows better than a Maze 3 does if the nozzle plate is removed.

Rayman2k2 · 02-22-2003, 10:04 PM

my god cathar, you've invented some sort of SUPER block!!!

Spamz0r · 02-23-2003, 07:21 AM

Quote:

Originally posted by Rayman2k2
my god cathar, you've invented some sort of SUPER block!!!

what the !?

thats one freaky emoticon ! :P

hara · 02-23-2003, 07:42 AM

Is it better than the maze3 with the plate removed?

Rayman2k2 · 02-23-2003, 01:19 PM

from what I hear, Cathar's block's temps are 5c degrees less than Maze3, so with out the middle plate it would probably be 3c below Maze3

theetruscan · 02-23-2003, 06:02 PM

Why would you want to take the restriction out, a moderately high flow pump can be had for around $20US.

I thought the inlet was designed to increase pressure and to spread flow. I wonder if the stable overclocks would drop disproportionately to the 2C increase because of lower flow on the areas over the sides of the die.

Cathar · 02-23-2003, 08:31 PM

Quote:

Originally posted by Rayman2k2
from what I hear, Cathar's block's temps are 5c degrees less than Maze3, so with out the middle plate it would probably be 3c below Maze3

Mind you, that's a very carefully lapped Maze 3. The Maze 3 when I got it performed about 2-3C worse. Once lapped and mounted firmly (throw away the DangerDen stock springs) the Maze 3 is actually a very, very good performer. Far better than the [H]ardOCP review makes out.

So, yes, the answer about "about 3C" better than a properly lapped and mounted Maze 3 when used without the nozzle plate, but keeping the same pump/heater-core.

Rayman2k2 · 02-23-2003, 08:53 PM

not bad....i have another question, when you keep the middle plate in, will the waterflow leaving the block be so minimal that you cant use other blocks? I am planning on using an Ehiem1250.

Cathar · 02-23-2003, 09:18 PM

Quote:

Originally posted by Rayman2k2
not bad....i have another question, when you keep the middle plate in, will the waterflow leaving the block be so minimal that you cant use other blocks? I am planning on using an Ehiem1250.

This seems to be a common misconception after BillA's review, which while I thought detailed the restriction and pressure-drop appropriately via the results, the wording was a little harsh.

Using an Eheim 1250 and a heater-core, I see 6.7lpm flow rates. By the same token, a DTek TC-4 would see about 7.2lpm flow rates, and a Maze 3 around 8.0lpm flow rates

We're talking 5-15% differences here, which if one looks at flow-C/W curves often enough, amounts to some pretty small temperature differences for other blocks in the system.

What does that mean really? It means that if you had a cooling loop with multiple blocks and a Maze 3, and then pulled out the Maze 3 and stuck a White Water in its place, you'd be unlucky if you noticed that the performance of the other blocks changed so much that what they are cooling is as much as 0.5C warmer than it was before.

My educated guess, remembering that GPU's and NB's are much lower heat emitting devices than CPU's, is that you'd see less than a 0.2C difference on a GPU block, and less than a 0.1C difference on a NB block as a result of the slightly lower flow rates.

pokpok · 02-23-2003, 11:16 PM

hi cathar!

quick question for you, do you have any US resellers?

Blackeagle · 02-24-2003, 01:03 AM

From posts I've read from those who have your block and are happy with it they are reporting up to 1C higher water temps at idle and 1-1.5C higher water temps at load in O/Ced systems. This makes sense as the extra heat removed from the chip has to go somewhere, and with a lower flow helps it show up more.

So would not the temps on other blocks have to rise closer to 1.5C due to the warmer water combined with the slightly lower flows?

This shouldn't be taken as a slam on the WW as the CPU is the most important point for cooling. And a 1-1.5C change in temp of the NB & GPU will make far less differance in a system's total O/Cing effort than a 5C differance will at the CPU!

And could you not also increase the width of the impingment slit slightly combined with a higher grade pump than the Ehiem like a Iwaki MD-15 or MD-20Z? This should allow the stronger pump to maintain the lower temps reported in testing at lower psi while also keeping flow high for best performance of other blocks? This would seem to be the solution for those concerned with the lower flow of the WW compared to other blocks when used in a 3 block system.

Cathar, just how much would you estimate the slit should be opened up to gain this? .5-1mm??

Although as noted above the CPU is the more important place for cooling. So the MD-20Z could also increase the velocity through the WW with a smaller increas in total flow but gaining perhaps a 6C + advantage at the CPU.

Some interesting possible options here, depending on where you value the cooling most.

Cathar · 02-24-2003, 04:08 AM

Quote:

Originally posted by Blackeagle
From posts I've read from those who have your block and are happy with it they are reporting up to 1C higher water temps at idle and 1-1.5C higher water temps at load in O/Ced systems. This makes sense as the extra heat removed from the chip has to go somewhere, and with a lower flow helps it show up more.

Under what controlled conditions were such tests taken? This is total trash.

I'm sorry to be so blunt, but such statements really are total crap.

The water won't be warmer. Why? Because the CPU is pumping the same amount of heat into the water. There is no "extra heat". If the CPU has clocked at a certain speed, and runs at a certain voltage, running a certain program, it emits the same amount of heat energy.

The technical reason why the White Water cools so well is because it lower the thermal gradients from the CPU surface to the point that the heat is absorbed by the water.

It's easiest to think of it like electrical resistance. The White Water offer less thermal resistance for the heat to dissipate, therefore the "voltage drop/potential difference" (temperature difference) from input to output is less.

theetruscan · 02-24-2003, 04:24 AM

Under what controlled conditions were such tests taken? This is total trash.

I'm sorry to be so blunt, but such statements really are total crap.

The water won't be warmer. Why? Because the CPU is pumping the same amount of heat into the water. There is no "extra heat". If the CPU has clocked at a certain speed, and runs at a certain voltage, running a certain program, it emits the same amount of heat energy.
__________________________________________

Are you sure? I mean obviously the CPU releases the same amount of heat, so some of the reasoning may be faulty, but it seems reasonable that the reduction in flow rate from the additional pressure drop could result in something on the order of 1C temperature increase. I could be completely off base here, but it was my understang that lower flow rates result in higher water temps.

Cathar · 02-24-2003, 04:30 AM

Quote:

Originally posted by theetruscan
I could be completely off base here, but it was my understang that lower flow rates result in higher water temps.

Rather than me explaining why the water temps won't be higher (as I already just have), perhaps you could take the time to explain just why you think the water temps could possibly be higher by as much as 1C (or 1.5C as stated above).

Cathar · 02-24-2003, 04:38 AM

Quote:

Originally posted by Blackeagle
And could you not also increase the width of the impingment slit slightly combined with a higher grade pump than the Ehiem like a Iwaki MD-15 or MD-20Z? This should allow the stronger pump to maintain the lower temps reported in testing at lower psi while also keeping flow high for best performance of other blocks? This would seem to be the solution for those concerned with the lower flow of the WW compared to other blocks when used in a 3 block system.

Cathar, just how much would you estimate the slit should be opened up to gain this? .5-1mm??

If anything, for higher pressure pumps, it would be beneficial to reduce the width of the nozzle slot, not widen it.

Widening the nozzle slot further reduces the impingement velocity at a faster rate than the resultant increase in flow rates would account for. The block's performance starts to drop off.

Narrowing the nozzle slot with a higher pressure pump allows that higher pressure pump to sustain the same (or even higher) volumetric flow rates as is seen with something like an Eheim 1250, but the corresponding impingement nozzle velocity will be much higher, possibly dropping up to 1C off the CPU's temperature while not really affecting the performance of other blocks in the system much at all.

theetruscan · 02-24-2003, 04:39 AM

could possibly be higher.

1. Radiator performance is dependent on flow rate. A dumbed down equation for all of us is provided by Bill Adams over on www.overclockers.com, it is.

"Heat Extracted = Btu/hr = 499 x flow rate in gpm (T - Ti)."

2. most hobby/aquarium pumps have very low pressure output (may be wrong word, sorry), and will lose a not insignificant amount of flow when you switch to the WWR block.

Combining these two, I believed that the water would stabilize at a slightly higher temp due to the lower flow rate. How am I wrong?

Cathar · 02-24-2003, 04:49 AM

Quote:

Originally posted by theetruscan
Combining these two, I believed that the water would stabilize at a slightly higher temp due to the lower flow rate. How am I wrong?

Sorry, you're going to have to do better than this. You're on the right path though. You're quoting BillA's results, but now please apply them to demonstrate how the water temps could possibly rise by 1C or more...

Hint: Yes - they will be ever so slightly warmer. By 1C? Not even close.

Hint #2: Compare radiator dissipation performance for 8lpm and 6.7lpm for something like the Big Momma, which is what the DTek Pro cores are (but with 1/2" fittings instead of the restrictive 3/8" fittings that BillA tested with).

Hint #3: Take a look at this thread and understand it:
http://forum.oc-forums.com/vb/showth...hreadid=148310

Join #2 and #3 together, citing performance data as given by BillA, to arrive at a total water temperature gain, and please report what you find.

Sorry for being a little harsh, but wouldn't you rather understand the full impact of what's going on here by working it out for yourself, rather than me just giving the answer and then not knowing how I derived it?

theetruscan · 02-24-2003, 04:53 AM

Thank you for the further information. I will look at it sometime, unfortunately probably not soon enough. Sorry for posting a poorly constructed thought. I'm less than two months from finishing my thesis, and all other coherent thoughts are a thing of vague memory.

Cathar · 02-24-2003, 04:57 AM

[Edit: Just saw your above post - the following is still relevant though if you wish to investigate the data properly and arrive at a real result]

Hint #4. BillA's corrected radiator heat dissipation test data can be found here:

http://www.thermal-management-testin...issipation.htm

6.7lpm = 1.7gpm
8.0lpm = 2.0gpm

1000BTU = 293W

A extremely hot CPU typically radiates about 70-90W of heat under full load, not the 100-150W that some thermal calculators like to pretend that is emitted.

Fan's that most people use will typically follow the 0.05inH2O air-pressure line (the yellow line on the graphs)

Blackeagle · 02-24-2003, 08:00 AM

Well I did screw up and started useing someone's reported water temp changes for the changes in block temps while adding the smaller real GPU & NB block changes to those temps.

But I made no claim those numbers were from any proper testing, in fact I said clearly they were not, only numbers from those who have been reporting the improvements they've seen, including changes in water temps. In this case from a thread over at Extreme systems I read. He was comparing your block to a Spir@l instead of a Maze. And you took no exception to his reporting a increase in water temps along with lower CPU temps of 4C at idle and of 6C at load. And it was in a thread you started and posted to a number of times both before and after he posted those numbers.

A simple quote of the part of my post that was in error with a polite question as to what could I have meant by that would have been plenty. No need for a major rant.

Cathar · 02-24-2003, 08:34 AM

That would be this thread here, with the post in question by SuperFreak.

At no time did the post mention that the ambient temperatures were constant, and this directly affects the resultant water temperatures.

Sorry for the rant, but I really wanted to squash the myth that more efficient water-blocks lead to (noticably) higher water temperatures. This is a statement that Steve Lynch made at [H]ardOCP when explaining/defending his waterblock roundup, and it's false and needs to be addressed before it becomes one of those "pseudo-facts".

That, and I must admit that I'm a little sensitive about it. Part of the problem with designing a block like the White Water is that it does a lot of things different to what people commonly accepted as "good", so it's always been a bit of an uphill battle to educate just how and why it works, and the side-effects of using it, and cutting through a lot of (sometimes quite strong) negativity along the way.

Maybe I'm just tired and grumpy. Time to go to bed.

Blackeagle · 02-24-2003, 09:45 AM

It's cool man, everybody's entitled to a bad day sometime. And as many times as I'm sure you've needed to answer the same questions and address the wrong impressions I can't blame you for being a bit tired of it.

When it's gett'n to ya just remember, the only stat that shines above all others is that it cools a CPU better than any other block now on the market.

02-22-2003, 06:47 PM	#1
Rayman2k2 Cooling Savant Join Date: Feb 2003 Location: Annapolis, MD Posts: 190	Question for Cathar Would there be any difference in temps if you were to run your block without the middle piece? I am talking about the one that blocks the water and makes it come out faster. __________________ -Insert Signature Here-

02-22-2003, 10:04 PM	#5
Rayman2k2 Cooling Savant Join Date: Feb 2003 Location: Annapolis, MD Posts: 190	my god cathar, you've invented some sort of SUPER block!!! __________________ -Insert Signature Here-

02-23-2003, 07:42 AM	#7
hara Cooling Savant Join Date: Oct 2002 Location: Malta, Mediterranean Posts: 662	Is it better than the maze3 with the plate removed? __________________ - Every great HD crash day is the day before back-up day. - My Past System - "Better to reign in Hell than serve in Heaven." - Milton, Paradise Lost. - FMZ

02-23-2003, 01:19 PM	#8
Rayman2k2 Cooling Savant Join Date: Feb 2003 Location: Annapolis, MD Posts: 190	from what I hear, Cathar's block's temps are 5c degrees less than Maze3, so with out the middle plate it would probably be 3c below Maze3 __________________ -Insert Signature Here-

02-23-2003, 08:53 PM	#11
Rayman2k2 Cooling Savant Join Date: Feb 2003 Location: Annapolis, MD Posts: 190	not bad....i have another question, when you keep the middle plate in, will the waterflow leaving the block be so minimal that you cant use other blocks? I am planning on using an Ehiem1250. __________________ -Insert Signature Here-

02-22-2003, 08:16 PM	#2
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	About a 2C difference if run without the nozzle plate.

02-22-2003, 08:32 PM	#3
Rayman2k2 Cooling Savant Join Date: Feb 2003 Location: Annapolis, MD Posts: 190	hrrrrrmmmmm......would there still be the same amount of flow resistance? __________________ -Insert Signature Here-

02-22-2003, 09:23 PM	#4
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Flow resistance is then very low. From my limited measurements it flows better than a Maze 3 does if the nozzle plate is removed.

02-23-2003, 06:02 PM	#9
theetruscan Cooling Neophyte Join Date: Jan 2003 Location: OR/CA/NY Posts: 81	Why would you want to take the restriction out, a moderately high flow pump can be had for around $20US. I thought the inlet was designed to increase pressure and to spread flow. I wonder if the stable overclocks would drop disproportionately to the 2C increase because of lower flow on the areas over the sides of the die.

02-23-2003, 11:16 PM	#13
pokpok Cooling Neophyte Join Date: Feb 2003 Location: rsm,ca Posts: 87	hi cathar! quick question for you, do you have any US resellers?

02-24-2003, 01:03 AM	#14
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	From posts I've read from those who have your block and are happy with it they are reporting up to 1C higher water temps at idle and 1-1.5C higher water temps at load in O/Ced systems. This makes sense as the extra heat removed from the chip has to go somewhere, and with a lower flow helps it show up more. So would not the temps on other blocks have to rise closer to 1.5C due to the warmer water combined with the slightly lower flows? This shouldn't be taken as a slam on the WW as the CPU is the most important point for cooling. And a 1-1.5C change in temp of the NB & GPU will make far less differance in a system's total O/Cing effort than a 5C differance will at the CPU! And could you not also increase the width of the impingment slit slightly combined with a higher grade pump than the Ehiem like a Iwaki MD-15 or MD-20Z? This should allow the stronger pump to maintain the lower temps reported in testing at lower psi while also keeping flow high for best performance of other blocks? This would seem to be the solution for those concerned with the lower flow of the WW compared to other blocks when used in a 3 block system. Cathar, just how much would you estimate the slit should be opened up to gain this? .5-1mm?? Although as noted above the CPU is the more important place for cooling. So the MD-20Z could also increase the velocity through the WW with a smaller increas in total flow but gaining perhaps a 6C + advantage at the CPU. Some interesting possible options here, depending on where you value the cooling most.

02-24-2003, 04:24 AM	#16
theetruscan Cooling Neophyte Join Date: Jan 2003 Location: OR/CA/NY Posts: 81	Under what controlled conditions were such tests taken? This is total trash. I'm sorry to be so blunt, but such statements really are total crap. The water won't be warmer. Why? Because the CPU is pumping the same amount of heat into the water. There is no "extra heat". If the CPU has clocked at a certain speed, and runs at a certain voltage, running a certain program, it emits the same amount of heat energy. __________________________________________ Are you sure? I mean obviously the CPU releases the same amount of heat, so some of the reasoning may be faulty, but it seems reasonable that the reduction in flow rate from the additional pressure drop could result in something on the order of 1C temperature increase. I could be completely off base here, but it was my understang that lower flow rates result in higher water temps.

02-24-2003, 04:39 AM	#19
theetruscan Cooling Neophyte Join Date: Jan 2003 Location: OR/CA/NY Posts: 81	Why I think the water temps . . . could possibly be higher. 1. Radiator performance is dependent on flow rate. A dumbed down equation for all of us is provided by Bill Adams over on www.overclockers.com, it is. "Heat Extracted = Btu/hr = 499 x flow rate in gpm (T - Ti)." 2. most hobby/aquarium pumps have very low pressure output (may be wrong word, sorry), and will lose a not insignificant amount of flow when you switch to the WWR block. Combining these two, I believed that the water would stabilize at a slightly higher temp due to the lower flow rate. How am I wrong?

02-24-2003, 04:53 AM	#21
theetruscan Cooling Neophyte Join Date: Jan 2003 Location: OR/CA/NY Posts: 81	Thank you for the further information. I will look at it sometime, unfortunately probably not soon enough. Sorry for posting a poorly constructed thought. I'm less than two months from finishing my thesis, and all other coherent thoughts are a thing of vague memory.

02-24-2003, 04:57 AM	#22
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	[Edit: Just saw your above post - the following is still relevant though if you wish to investigate the data properly and arrive at a real result] Hint #4. BillA's corrected radiator heat dissipation test data can be found here: http://www.thermal-management-testin...issipation.htm 6.7lpm = 1.7gpm 8.0lpm = 2.0gpm 1000BTU = 293W A extremely hot CPU typically radiates about 70-90W of heat under full load, not the 100-150W that some thermal calculators like to pretend that is emitted. Fan's that most people use will typically follow the 0.05inH2O air-pressure line (the yellow line on the graphs)

02-24-2003, 08:00 AM	#23
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	Ranting Well I did screw up and started useing someone's reported water temp changes for the changes in block temps while adding the smaller real GPU & NB block changes to those temps. But I made no claim those numbers were from any proper testing, in fact I said clearly they were not, only numbers from those who have been reporting the improvements they've seen, including changes in water temps. In this case from a thread over at Extreme systems I read. He was comparing your block to a Spir@l instead of a Maze. And you took no exception to his reporting a increase in water temps along with lower CPU temps of 4C at idle and of 6C at load. And it was in a thread you started and posted to a number of times both before and after he posted those numbers. A simple quote of the part of my post that was in error with a polite question as to what could I have meant by that would have been plenty. No need for a major rant.

02-24-2003, 08:34 AM	#24
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	That would be this thread here, with the post in question by SuperFreak. At no time did the post mention that the ambient temperatures were constant, and this directly affects the resultant water temperatures. Sorry for the rant, but I really wanted to squash the myth that more efficient water-blocks lead to (noticably) higher water temperatures. This is a statement that Steve Lynch made at [H]ardOCP when explaining/defending his waterblock roundup, and it's false and needs to be addressed before it becomes one of those "pseudo-facts". That, and I must admit that I'm a little sensitive about it. Part of the problem with designing a block like the White Water is that it does a lot of things different to what people commonly accepted as "good", so it's always been a bit of an uphill battle to educate just how and why it works, and the side-effects of using it, and cutting through a lot of (sometimes quite strong) negativity along the way. Maybe I'm just tired and grumpy. Time to go to bed.

02-24-2003, 09:45 AM	#25
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	It's cool man, everybody's entitled to a bad day sometime. And as many times as I'm sure you've needed to answer the same questions and address the wrong impressions I can't blame you for being a bit tired of it. When it's gett'n to ya just remember, the only stat that shines above all others is that it cools a CPU better than any other block now on the market.

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