Maximum flow rates possible??

Blackeagle · 11-04-2004, 02:41 PM

In a system with a low resistence core like the 2-342 & a block with restrictive interior structures such as a TDX, RBX, Cascade or Storm what would you expect the max flow rate to be when making use of a pump such as a Iwaki MD20RLT??

What if a MD20RLZT or MD30RLZT were used in place of the MD20RLT?

I'm asking this due to a referance I've seen at another site which gave 2.5gpm as the AVERAGE flow rate for most users systems. I disagreed with this, saying a Iwaki MD20 would be hard up to provide that in many loops, and would not be able to provide it if a block for the GPU like the Fusion were to be added.

Guy claimed he'd seen posts HERE that supported a 2.5 gpm average for most systems flow rates.

I've not been following on the newest developments real close of late (except the Storm blocks), so thought I'd better check to see if the fool is I. I may be wrong but I thought the MD20Z would be very hard put to make 2.5gpm even with a loop designed to be very low resistence except in the block. I thought 2.25gpm would be about it for the MD20Z with a high resistence CPU block.

How's about it guys, are the MD30 and other more powerfull pumps much like the MD30 in performance becoming much more common in use? ? Or can the MD20Z or MD20RLT perform at hgiher levels than I"m thinking?

Cathar · 11-04-2004, 03:11 PM

I am presently part way through a larger article that covers the same issues, but in the time-being you may be interested in this post I made at OCAU:

How much pump is enough? How much pump is too much?

The MD30 is almost always a substantial step backwards away from using the MD20 in almost any realistics radiator based scenario. The MD30's simply put WAY too much heat into the loop, and the radiator allows the water to warm up by more than the waterblock improves CPU temps by.

Check it out. There are some surprises there. Some of the differences being touted by people when changing pumps simply are not realistic, with barely 1.5C differences between an Eheim 1048 and a "best possible case" pump for hot systems.

As stated, I am fleshing out a more detailed article, and I'll cover issues like varying CPU heat load, and GPU blocks with added restriction, and include a few different CPU blocks with different shaped flow vs C/W curves and pressure drop characteristics. From what I'm seeing here, and I have stated this in the past, at no time is a pump that draws more than a rated 50W of power a step forwards for better performance.

mastermind2004 · 11-04-2004, 04:04 PM

Could you perhaps copy/paste the post here, as OCAU won't let me register with gmail, and I don't feel like giving out my other e-mail address?
:shrug:

Blackeagle · 11-04-2004, 04:22 PM

Quote:

Originally Posted by mastermind2004

Could you perhaps copy/paste the post here, as OCAU won't let me register with gmail, and I don't feel like giving out my other e-mail address?
:shrug:

Same here ! I'm useing Yahoo.

Cathar · 11-04-2004, 04:34 PM

Quote:

Originally Posted by mastermind2004

Could you perhaps copy/paste the post here, as OCAU won't let me register with gmail, and I don't feel like giving out my other e-mail address?
:shrug:

OCAU NEVER Spams. Just register.

However, for the overly paranoid I have reposted the pertinent posts here:

http://forums.procooling.com/vbb/showthread.php?t=10825

mastermind2004 · 11-04-2004, 04:37 PM

Thanks. I just don't like giving out my address, I'm not so paranoid about spam and such.

BillA · 11-04-2004, 04:42 PM

Iwaki pumps are not representative of "average systems", period
"average systems" with better pumps will see 1.1 to 1.5 gpm depending on the pump and specific configuration
this is based on data, and Swiftech should be qualified to characterize "average systems"

Blackeagle · 11-04-2004, 05:02 PM

Quote:

Originally Posted by unregistered

Iwaki pumps are not representative of "average systems", period
"average systems" with better pumps will see 1.1 to 1.5 gpm depending on the pump and specific configuration
this is based on data, and Swiftech should be qualified to characterize "average systems"

I agree with you Iwaki, Little Giant, Panworld ect are pumps of the really extreme users, not the average.

I pointed out that a Iwaki MD20RLT (his pump) isn't average IMO. I also suggested that a average system would be more around half his estimate (1.25gpm). Then when he responded that he'd gotten that figure from data posted at PC, I decided I should check to see if I were wrong prior to posting anything more regarding such. I have been seeing more guys making use of Panworlds, Iwaki ect than a year of so ago, but still thought they were beyond the average, which the responses here seem to support.

Blackeagle · 11-04-2004, 05:31 PM

BTW,

I'm going to order this heater core to go with one of Cathars Storm blocks. GDI # 399030 @ PartsAmerica.com this is the same as the 2-192 core from Ford trucks. Has a finned core area that is 2" thick and is 10.75 X 10.25".

Based on Cathar's specs of watts vs rad area, this thing should be able to handle a MD30 type pump and match or beat a MD20. So should do a great job of taking care of my smaller MD20's heat. Which is good as in my case I'm also going to have my Fusion block on the GPU to disapate heat from as well.

Even with great care in routing of lines, I know I won't make 2gpm, let alone 2.5gpm.

11-04-2004, 02:41 PM	#1
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	Maximum flow rates possible?? In a system with a low resistence core like the 2-342 & a block with restrictive interior structures such as a TDX, RBX, Cascade or Storm what would you expect the max flow rate to be when making use of a pump such as a Iwaki MD20RLT?? What if a MD20RLZT or MD30RLZT were used in place of the MD20RLT? I'm asking this due to a referance I've seen at another site which gave 2.5gpm as the AVERAGE flow rate for most users systems. I disagreed with this, saying a Iwaki MD20 would be hard up to provide that in many loops, and would not be able to provide it if a block for the GPU like the Fusion were to be added. Guy claimed he'd seen posts HERE that supported a 2.5 gpm average for most systems flow rates. I've not been following on the newest developments real close of late (except the Storm blocks), so thought I'd better check to see if the fool is I. I may be wrong but I thought the MD20Z would be very hard put to make 2.5gpm even with a loop designed to be very low resistence except in the block. I thought 2.25gpm would be about it for the MD20Z with a high resistence CPU block. How's about it guys, are the MD30 and other more powerfull pumps much like the MD30 in performance becoming much more common in use? ? Or can the MD20Z or MD20RLT perform at hgiher levels than I"m thinking?

11-04-2004, 03:11 PM	#2
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	I am presently part way through a larger article that covers the same issues, but in the time-being you may be interested in this post I made at OCAU: How much pump is enough? How much pump is too much? The MD30 is almost always a substantial step backwards away from using the MD20 in almost any realistics radiator based scenario. The MD30's simply put WAY too much heat into the loop, and the radiator allows the water to warm up by more than the waterblock improves CPU temps by. Check it out. There are some surprises there. Some of the differences being touted by people when changing pumps simply are not realistic, with barely 1.5C differences between an Eheim 1048 and a "best possible case" pump for hot systems. As stated, I am fleshing out a more detailed article, and I'll cover issues like varying CPU heat load, and GPU blocks with added restriction, and include a few different CPU blocks with different shaped flow vs C/W curves and pressure drop characteristics. From what I'm seeing here, and I have stated this in the past, at no time is a pump that draws more than a rated 50W of power a step forwards for better performance. Last edited by Cathar; 11-04-2004 at 03:17 PM.

11-04-2004, 04:04 PM	#3
mastermind2004 Cooling Neophyte Join Date: Aug 2003 Location: MIT Posts: 78	Could you perhaps copy/paste the post here, as OCAU won't let me register with gmail, and I don't feel like giving out my other e-mail address? :shrug:

11-04-2004, 04:37 PM	#6
mastermind2004 Cooling Neophyte Join Date: Aug 2003 Location: MIT Posts: 78	Thanks. I just don't like giving out my address, I'm not so paranoid about spam and such.

11-04-2004, 04:42 PM	#7
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	Iwaki pumps are not representative of "average systems", period "average systems" with better pumps will see 1.1 to 1.5 gpm depending on the pump and specific configuration this is based on data, and Swiftech should be qualified to characterize "average systems"

11-04-2004, 05:31 PM	#9
Blackeagle Thermophile Join Date: Oct 2002 Location: U.S.A = Michigan Posts: 1,243	BTW, I'm going to order this heater core to go with one of Cathars Storm blocks. GDI # 399030 @ PartsAmerica.com this is the same as the 2-192 core from Ford trucks. Has a finned core area that is 2" thick and is 10.75 X 10.25". Based on Cathar's specs of watts vs rad area, this thing should be able to handle a MD30 type pump and match or beat a MD20. So should do a great job of taking care of my smaller MD20's heat. Which is good as in my case I'm also going to have my Fusion block on the GPU to disapate heat from as well. Even with great care in routing of lines, I know I won't make 2gpm, let alone 2.5gpm.

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