Pump testing - Page 3

BillA · 09-26-2003, 05:03 PM

a sorting of multiple voltage pump curves yields this

it does make the pump's performance a bit more explainable
(pHaestus, I've sent you the table of values)

a more accurate of the pump input power is 10.6 to 13W, depending on the flow rate (higher flow, more power)

pHaestus · 09-26-2003, 05:19 PM

Yup got it. Thanks Bill. Eyeballing it, it seems like the MCP600 will fall on Les's fitted equation reasonably well now. I'll put a DMM into the molex when I get home and check the 12V rail. It's been a couple of days so I can check for noise too.

Aardil · 09-26-2003, 05:23 PM

OK you sold me Billa
2 MCP600 pumps have been ordered and hopefully will be here Monday.
When all is over and done with, this project may perform horribly but it wont be for lack of good parts

Les · 09-26-2003, 06:44 PM

Quote:

Originally posted by pHaestus
Yup got it. Thanks Bill. Eyeballing it, it seems like the MCP600 will fall on Les's fitted equation reasonably well now. ...........

Updated(without equation) using revised MCP600 data :-

It does appear that the Eheim1250 may be the only problem child.

KnightElite · 09-27-2003, 08:43 PM

One thing that may or may not be worth noting about my Eheim 1250 (the one pH used for testing) is that it has a 5/8" exit barb, rather than the 3/8" one most commonly sold with the things. I dunno how that would affect results.

BillA · 09-27-2003, 08:56 PM

improve it if anything

Halo_Master · 09-28-2003, 09:43 AM

Quote:

Originally posted by KnightElite
One thing that may or may not be worth noting about my Eheim 1250 (the one pH used for testing) is that it has a 5/8" exit barb, rather than the 3/8" one most commonly sold with the things. I dunno how that would affect results.

No,

The 1250 has a 1/2" exit barb and a 5/8" inlet barb

pHaestus · 09-28-2003, 02:23 PM

KE's Eheim outlet got a crack in it, so I replaced the stock outlet with a 3/8" copper pipe to 1/2" NPT female fitting. That was held in place via JB Weld and it is secure and strong. The resulting setup is no more restrictive than it was before with a 1/2" barb in it (the 3/8" ID of the barb is the limit) and is less restrictive than stock with a 5/8" barb. It is normally used with a 1/2" NPT to 1/2" copper pipe fitting in it (even larger id than the 5/8" barb), but I used a 1/2"NPT to 1/2" barb in the testing since that's what the 1250 comes with stock.

The fact remains that the Eheim has been modified rather substantially even though I don't believe it was something that would affect the flow rates for the worse. Bill sees even worse performance than I do from the 1250, presumably with stock parts. Many people on forums have reported no difference in performance with 1250 vs 1048. I conclude from all of this that the P-Q curve is probably a bit "optimistic" for the 1250. If I ever have a brand new 1250 in my possession again I'll be sure to test it.

Now would you guys care to see this testing repeated with 3/8" tubing and dual rads and a GPU and CPU wb? Cause I think I am doing that next. Basically it would generate another Power line on Les's excellent figure with higher resistance. I'll try to get it started tonight.

Les · 09-29-2003, 04:46 PM

Quote:

Originally posted by pHaestus
............
Now would you guys care to see this testing repeated with 3/8" tubing and dual rads and a GPU and CPU wb? .......

Lacking suggestions, how about an extremely restrictive loop containing MCW-CHILL 452™ with 1/4"ID tubing. Have added some speculative simulations (at 26c and 0c) to graph, and would guess it might goad Bill into producing some data.

BillA · 09-29-2003, 05:57 PM

come on Les, if I give you more data you'll just add more curves
what do you want ?

N.B. its pHaestus' turn to gin out the numbers

Les · 09-30-2003, 10:35 AM

Quote:

Originally posted by unregistered
come on Les, if I give you more data you'll just add more curves
what do you want ?
......

Have not exhausted ColourFul Displays(CFDs?) of existing data.
However am finding it somewhat befuddling :-
Including data from

get:

BillA · 09-30-2003, 11:04 AM

yes, curious
the 1048 flows 'more' than it should
or
the system resistance was lower for its test
(not the case as the pumps were substituted one after another)

did you look at the 1/2" numbers ?
(actual test numbers sent to all)

Les · 09-30-2003, 01:43 PM

Quote:

Originally posted by unregistered
yes, curious
........
did you look at the 1/2" numbers ?
(actual test numbers sent to all)

Some of

which enhances the befuddlement:-

BillA · 09-30-2003, 02:06 PM

curiouser and curiouser . . . .
why the fliers indicating a different resistance ?
and note they do fall correctly on their respective curves
? ? ? ?
I need to review my test setup, but that is probably not going to yield much
what other explainations ?

adaptors on pumps to fitup the tubing sizes seems possible
later

Les · 09-30-2003, 02:19 PM

Quote:

Originally posted by unregistered

why the fliers indicating a different resistance ?

Not intended to imply different resistances. Only to describe the fitted "power".relationships.
At the mpment am also treating as a curiosity which needs some thought.

Ruiner · 11-18-2003, 10:04 AM

One flow comparo I would be interested in would be a pump-res vs. inline res. vs. T-tube setups. Is this easy to do with your setup?

pHaestus · 11-18-2003, 12:10 PM

Not a bad test. I am not sure the pump-res vs. pump and res will give enough difference to pick up with big tubing but perhaps at 3/8". I tell you what; I'll see if I can devise a good test tonight related to your question along with some questions I have been wondering about.

bigben2k · 11-19-2003, 08:43 AM

One test that I ran on all my pumps, is to cap them with a PVC screw cover, into which I drilled a hole, between 3/16 and 3/8".

BillA · 11-19-2003, 08:55 AM

Quote:

Originally posted by bigben2k
One test that I ran on all my pumps, is to cap them with a PVC screw cover, into which I drilled a hole, between 3/16 and 3/8".

demonstrating what ?

bigben2k · 11-19-2003, 09:17 AM

It's a simple way of having the same restriction.

For a cheap test, you can measure the landing point of the jet, sideways, or the "shooting height". Mostly for fun though, not terribly useful. Good for a rough comparison.

For something more accurate (and useful), one can measure the flow rate, at which point a flow meter is required (isn't it always?).

Personally, I'd much rather set the flow rate, and measure the pressure drop: the comparison of the results is easier that way, but it's gotta be done right.

09-27-2003, 08:43 PM	#55
KnightElite Cooling Savant Join Date: Sep 2002 Location: Saskatoon, Saskatchewan Posts: 294	One thing that may or may not be worth noting about my Eheim 1250 (the one pH used for testing) is that it has a 5/8" exit barb, rather than the 3/8" one most commonly sold with the things. I dunno how that would affect results. __________________ Can anyone else here say that they have a watercooled monster that's 45" tall?

09-30-2003, 11:04 AM	#62
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	yes, curious the 1048 flows 'more' than it should or the system resistance was lower for its test (not the case as the pumps were substituted one after another) did you look at the 1/2" numbers ? (actual test numbers sent to all) Last edited by BillA; 09-30-2003 at 11:21 AM.

11-19-2003, 08:43 AM	#68
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	One test that I ran on all my pumps, is to cap them with a PVC screw cover, into which I drilled a hole, between 3/16 and 3/8". __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

11-19-2003, 09:17 AM	#70
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	It's a simple way of having the same restriction. For a cheap test, you can measure the landing point of the jet, sideways, or the "shooting height". Mostly for fun though, not terribly useful. Good for a rough comparison. For something more accurate (and useful), one can measure the flow rate, at which point a flow meter is required (isn't it always?). Personally, I'd much rather set the flow rate, and measure the pressure drop: the comparison of the results is easier that way, but it's gotta be done right. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

09-26-2003, 05:03 PM	#51
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	a sorting of multiple voltage pump curves yields this it does make the pump's performance a bit more explainable (pHaestus, I've sent you the table of values) a more accurate of the pump input power is 10.6 to 13W, depending on the flow rate (higher flow, more power)

09-26-2003, 05:19 PM	#52
pHaestus Big Player Making Big Money Join Date: Aug 2001 Location: irc.lostgeek.com #procooling.com Posts: 4,782	Yup got it. Thanks Bill. Eyeballing it, it seems like the MCP600 will fall on Les's fitted equation reasonably well now. I'll put a DMM into the molex when I get home and check the 12V rail. It's been a couple of days so I can check for noise too.

09-26-2003, 05:23 PM	#53
Aardil Cooling Savant Join Date: Oct 2002 Location: Louisiana Posts: 174	OK you sold me Billa 2 MCP600 pumps have been ordered and hopefully will be here Monday. When all is over and done with, this project may perform horribly but it wont be for lack of good parts

09-27-2003, 08:56 PM	#56
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	improve it if anything

09-28-2003, 02:23 PM	#58
pHaestus Big Player Making Big Money Join Date: Aug 2001 Location: irc.lostgeek.com #procooling.com Posts: 4,782	KE's Eheim outlet got a crack in it, so I replaced the stock outlet with a 3/8" copper pipe to 1/2" NPT female fitting. That was held in place via JB Weld and it is secure and strong. The resulting setup is no more restrictive than it was before with a 1/2" barb in it (the 3/8" ID of the barb is the limit) and is less restrictive than stock with a 5/8" barb. It is normally used with a 1/2" NPT to 1/2" copper pipe fitting in it (even larger id than the 5/8" barb), but I used a 1/2"NPT to 1/2" barb in the testing since that's what the 1250 comes with stock. The fact remains that the Eheim has been modified rather substantially even though I don't believe it was something that would affect the flow rates for the worse. Bill sees even worse performance than I do from the 1250, presumably with stock parts. Many people on forums have reported no difference in performance with 1250 vs 1048. I conclude from all of this that the P-Q curve is probably a bit "optimistic" for the 1250. If I ever have a brand new 1250 in my possession again I'll be sure to test it. Now would you guys care to see this testing repeated with 3/8" tubing and dual rads and a GPU and CPU wb? Cause I think I am doing that next. Basically it would generate another Power line on Les's excellent figure with higher resistance. I'll try to get it started tonight.

09-29-2003, 05:57 PM	#60
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	come on Les, if I give you more data you'll just add more curves what do you want ? N.B. its pHaestus' turn to gin out the numbers

09-30-2003, 02:06 PM	#64
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	curiouser and curiouser . . . . why the fliers indicating a different resistance ? and note they do fall correctly on their respective curves ? ? ? ? I need to review my test setup, but that is probably not going to yield much what other explainations ? adaptors on pumps to fitup the tubing sizes seems possible later

11-18-2003, 10:04 AM	#66
Ruiner Cooling Savant Join Date: May 2003 Location: New Smyrna, FL Posts: 258	One flow comparo I would be interested in would be a pump-res vs. inline res. vs. T-tube setups. Is this easy to do with your setup?

11-18-2003, 12:10 PM	#67
pHaestus Big Player Making Big Money Join Date: Aug 2001 Location: irc.lostgeek.com #procooling.com Posts: 4,782	Not a bad test. I am not sure the pump-res vs. pump and res will give enough difference to pick up with big tubing but perhaps at 3/8". I tell you what; I'll see if I can devise a good test tonight related to your question along with some questions I have been wondering about.

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