Balance between backpressure and WB flow speed?

awesterink · 07-16-2002, 09:21 PM

After digging through tons of threads about backpressure and flow speed, I learned that if the channel in a WB is narrower than the rest of the loop, the flow speed in the block increases, but the backpressure also increases (which I believe reduces flow rate).

I currently use Innovatek blocks, which have fairly narrow 1/4" ID fittings for its crappy 3/8" OD tubing. My Eheim 1048 seems to be fighting backpressure, as it is quite noisy.

I plan to go to 1/2" ID Tygon tubing all the way, using the 1048 and a BIX rad. I can put 1/2" fittings on the blocks (1/4"NPT threads), but the internals will probably still be quite narrow.

So the question is...should I stick with the Innovatek blocks and choose for good WB flow speed with crappy backpressure, or would I be better off with e.g. the DD MAZE3 and DD GF4 blocks?

#Rotor · 07-16-2002, 10:47 PM

go for flow velocity.... it is much more important than the actual volume being pumped.... the higher the velocity, the faster the blocks volume is getting replaced with fresh cold liquid.... also the more turbulent the flow will be, a very good thing.

I always say, massive flow-rate only means a lot of water went through without doing nothing to cool the CPU. flow rate needs to be converted into flow velocity and turbulence. If the pump can not handle the pressure, don't blame the block, get a better pump....

MO. And I live by it.... my blocks are all low volume high velocity designs.

gone_fishin · 07-16-2002, 11:16 PM

Quote:

Originally posted by #Rotor
go for flow velocity.... it is much more important than the actual volume being pumped.... the higher the velocity, the faster the blocks volume is getting replaced with fresh cold liquid.... also the more turbulent the flow will be, a very good thing.

I always say, massive flow-rate only means a lot of water went through without doing nothing to cool the CPU. flow rate needs to be converted into flow velocity and turbulence. If the pump can not handle the pressure, don't blame the block, get a better pump.... MO. And I live by it.... my blocks are all low volume high velocity designs.

If you restrict the flow with narrow channels then there is less fresh cold liquid going through it. Contradicting sentences you have there in the same post:shrug: High flow leads to more turbulence and more velocity. When you use a large pump with a restrictive block then you are going to be at or near it's maximum allowable volumetric flow rate which it will perform better with. A block design that allows even higher flow rates will take advantage of it with an even higher flow rate. Velocity increase goes hand in hand with increased volumetric flow.

#Rotor · 07-17-2002, 06:25 AM

but keep in mind, flow velocity and flow rate is not the same thing. velocity is the actual surface speed at which the molecules are moving, flow rate is the ammount of molecules that is passing a point. channel dimmentions is what creates velocity.

bigben2k · 07-17-2002, 08:29 AM

What it comes down to, is what the effective flow rate is going to be. If it's going to be above 100 gph, I'd recommend 1/2 in. tubing, otherwise, 3/8 would do.

What you need to do, awesterink, is figure out what kind of flow you're getting, with the bucket test. Once you've done this, you can then figure out if you want to:
a) get a better pump, and use 1/2 tubing

or

b) just swap the tubing for 3/8.

That's my $.02 ...

awesterink · 07-17-2002, 08:40 AM

Quote:

Originally posted by gone_fishin

High flow leads to more turbulence and more velocity. When you use a large pump with a restrictive block then you are going to be at or near it's maximum allowable volumetric flow rate which it will perform better with. A block design that allows even higher flow rates will take advantage of it with an even higher flow rate. Velocity increase goes hand in hand with increased volumetric flow.

Yes, a higher flow rate created by the pump with all other factors constant should increase flow speed in the entire loop, but a narrow channel in the block will increase flow speed in just the block while reducing flow rate for the entire loop. This is probably appreciated by the rad, but not by the pump, which has to start fighting backpressure and gets noisy.

On the other hand, a narrow channel in the block also reduces the volume of water that is in the block at any given time, which I would think is a downside.

So..which is the better way to go...1/2" tubing with a fairly narrow block or a wider channel block?

awesterink · 07-17-2002, 08:44 AM

Bigben,

Posted at the same time, sorry :-)

I am actually not trying to decide what tubing I'll go with, but rather what block I should go with given 1/2" tubing and 1/2" fittings on pump and rad?

bigben2k · 07-17-2002, 08:59 AM

Quote:

Originally posted by awesterink
Bigben,

Posted at the same time, sorry :-)

I am actually not trying to decide what tubing I'll go with, but rather what block I should go with given 1/2" tubing and 1/2" fittings on pump and rad?

It's all good...

If you've got 1/2 in. fittings everywhere else, you're very much ahead of the game. If you're considering another block (I think you could stick with what you have, but that's just a money thing, because I don't think that the extra cost to you is worth it), considering that you have a 1048 and that you want to feed two blocks, I'd go for either:
a) DD Maze 3, and their GF block ($42+45=87)
b) Dtek's new spiral, and geminicool's copper GF wb (also at dtek). ($52+36=88)

There's probably a few more choices out there, so it's up to you to shop around.

awesterink · 07-17-2002, 09:28 AM

Bigben,

At this point the money is a lost case already. I went this far, now I want to have it RIGHT!

My Innovatek blocks do not yet have 1/2" fittings, they come with compression fittings for 8mm ID/10mm OD tubing. I can however put 1/4NPT 1/2" fittings on them.

Ah. Decisions decisions. I appreciate all your help!!

Arjen

bigben2k · 07-17-2002, 09:43 AM

Quote:

Originally posted by awesterink
At this point the money is a lost case already. I went this far, now I want to have it RIGHT!

My Innovatek blocks do not yet have 1/2" fittings, they come with compression fittings for 8mm ID/10mm OD tubing. I can however put 1/4NPT 1/2" fittings on them.

Well, it up to you.

What kind of temps are you getting anyways?

awesterink · 07-17-2002, 10:38 AM

I am currently back at aircooling, but MBM reported 44C/45C idle, 49C/46C load. This was with the standard 8mm ID tubing and fittings, and the Maxxxpert 120mm rad using a Panaflow L1A intake at 7V.

The ambient sensor is probably somehere near the NB to report that high (Gigabyte 7VRXP). Plus my case sits in a cabinet, so the hot air exhausted from the back (120mm Panaflow L1A at 7V) gets trapped a little, making the ambient around the case higher than intake ambient.

redleader · 07-17-2002, 10:43 AM

What do the insides of Innovatek blocks look like anyway? Would they be much wider then 1/4ID tube?

awesterink · 07-17-2002, 10:49 AM

Not sure. I can screw in the standard 1/2" barb fitting that comes with the Eheim, so perhaps it is not so bad.....

bigben2k · 07-17-2002, 11:11 AM

Quote:

Originally posted by redleader
What do the insides of Innovatek blocks look like anyway? Would they be much wider then 1/4ID tube?

There really isn't any room to make the opening any bigger.

bigben2k · 07-17-2002, 11:13 AM

Quote:

Originally posted by awesterink
Not sure. I can screw in the standard 1/2" barb fitting that comes with the Eheim, so perhaps it is not so bad.....

That's the way I'd go.

bigben2k · 07-17-2002, 11:15 AM

Quote:

Originally posted by redleader
What do the insides of Innovatek blocks look like anyway? Would they be much wider then 1/4ID tube?

(pic rip-courtesy of OC!)

awesterink · 07-17-2002, 12:08 PM

I'll do that this weekend (using 1/2" fittings, 1/2" ID tubing, BIX rad and 1048) and report on the results compared to the standard kit.

BillA · 07-17-2002, 02:10 PM

read it horizontally,
from the blue curve to the magenta is what you will pickup in flow (from where you are at)
(its actually interactive - but close enough)

with barbed fittings the Innovatek is pretty much the same as the laybrinth types

bigben2k · 07-17-2002, 02:27 PM

See, see, I told you so!!!

Your flow will increase roughly 30%.

1/4 in. ID tubing is bad, very bad.

I believe that in your case 3/8 would have been OK, but what the heck, go for 1/2 and you'll be ready for the next upgrade.

awesterink · 07-17-2002, 02:39 PM

The standard fittings on the Innovatek have a 1/4" ID size, but the standard tubing is 8mm ID (about 5/16"). Still bad, very bad.

Say anonymous...how did you get to this wizardy in Excel?

JimS · 07-17-2002, 06:25 PM

Put a good size reservoir just before the pump and you will eliminate any backpressure problem. For me, the advantages of a reservoir far outweigh any disadvantages.

awesterink · 07-17-2002, 07:03 PM

You got me curious....there are disadvantages to using a reservoir? Like what?

bigben2k · 07-18-2002, 08:15 AM

Quote:

Originally posted by JimS
Put a good size reservoir just before the pump and you will eliminate any backpressure problem. For me, the advantages of a reservoir far outweigh any disadvantages.

You've got me curious as well... What do you mean? What is this thing about backpressure?

As for disadvantage, the only ones I know is that a res can take up a significant amount of space, and since it's best to have it at the top, that can mean getting creative with the case. Advantage: it works better than a T-fitting.

pHaestus · 07-18-2002, 08:47 PM

I wouldn't "upgrade" an Innovatek block to any other block I have used. With 1/2" barbs I find the CPU and GPU blocks to perform extremely well, especially at lower volumetric flow rates (which are easily attainable with smaller pumps). The Rev3 just oozes quality.

Powerhouse · 07-22-2002, 10:51 AM

Quote:

Originally posted by pHaestus
I wouldn't "upgrade" an Innovatek block to any other block I have used. With 1/2" barbs I find the CPU and GPU blocks to perform extremely well, especially at lower volumetric flow rates (which are easily attainable with smaller pumps). The Rev3 just oozes quality.

So pHaestus, by changing the Innovatek block to 1/2' barbs. what kind of performance increase (temp drop) did you get on your system?

07-16-2002, 09:21 PM	#1
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	Balance between backpressure and WB flow speed? After digging through tons of threads about backpressure and flow speed, I learned that if the channel in a WB is narrower than the rest of the loop, the flow speed in the block increases, but the backpressure also increases (which I believe reduces flow rate). I currently use Innovatek blocks, which have fairly narrow 1/4" ID fittings for its crappy 3/8" OD tubing. My Eheim 1048 seems to be fighting backpressure, as it is quite noisy. I plan to go to 1/2" ID Tygon tubing all the way, using the 1048 and a BIX rad. I can put 1/2" fittings on the blocks (1/4"NPT threads), but the internals will probably still be quite narrow. So the question is...should I stick with the Innovatek blocks and choose for good WB flow speed with crappy backpressure, or would I be better off with e.g. the DD MAZE3 and DD GF4 blocks? __________________ Asus A7N8XD - BIOS 1002 T-Bred 2600+ 2x256MB Samsung PC2700 120GB Seagate Barracuda V on SATA1 Pioneer DVR-105 on IDE1 ZIP250 on IDE2 Abit GeForce4 Ti4200 64MB - Det 41.09 Chenbro Junior SR-20503 PC Power & Cooling Silencer 400 ATX Windows XP Home

07-17-2002, 08:29 AM	#5
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	What it comes down to, is what the effective flow rate is going to be. If it's going to be above 100 gph, I'd recommend 1/2 in. tubing, otherwise, 3/8 would do. What you need to do, awesterink, is figure out what kind of flow you're getting, with the bucket test. Once you've done this, you can then figure out if you want to: a) get a better pump, and use 1/2 tubing or b) just swap the tubing for 3/8. That's my $.02 ... __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

07-17-2002, 02:27 PM	#19
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	See, see, I told you so!!! Your flow will increase roughly 30%. 1/4 in. ID tubing is bad, very bad. I believe that in your case 3/8 would have been OK, but what the heck, go for 1/2 and you'll be ready for the next upgrade. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101 Last edited by bigben2k; 07-17-2002 at 02:33 PM.

07-16-2002, 10:47 PM	#2
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	go for flow velocity.... it is much more important than the actual volume being pumped.... the higher the velocity, the faster the blocks volume is getting replaced with fresh cold liquid.... also the more turbulent the flow will be, a very good thing. I always say, massive flow-rate only means a lot of water went through without doing nothing to cool the CPU. flow rate needs to be converted into flow velocity and turbulence. If the pump can not handle the pressure, don't blame the block, get a better pump.... MO. And I live by it.... my blocks are all low volume high velocity designs.

07-17-2002, 06:25 AM	#4
#Rotor Cooling Savant Join Date: Feb 2002 Location: Dione, sector 4s1256 Posts: 852	but keep in mind, flow velocity and flow rate is not the same thing. velocity is the actual surface speed at which the molecules are moving, flow rate is the ammount of molecules that is passing a point. channel dimmentions is what creates velocity.

07-17-2002, 08:44 AM	#7
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	Bigben, Posted at the same time, sorry :-) I am actually not trying to decide what tubing I'll go with, but rather what block I should go with given 1/2" tubing and 1/2" fittings on pump and rad?

07-17-2002, 09:28 AM	#9
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	Bigben, At this point the money is a lost case already. I went this far, now I want to have it RIGHT! My Innovatek blocks do not yet have 1/2" fittings, they come with compression fittings for 8mm ID/10mm OD tubing. I can however put 1/4NPT 1/2" fittings on them. Ah. Decisions decisions. I appreciate all your help!! Arjen

07-17-2002, 10:38 AM	#11
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	I am currently back at aircooling, but MBM reported 44C/45C idle, 49C/46C load. This was with the standard 8mm ID tubing and fittings, and the Maxxxpert 120mm rad using a Panaflow L1A intake at 7V. The ambient sensor is probably somehere near the NB to report that high (Gigabyte 7VRXP). Plus my case sits in a cabinet, so the hot air exhausted from the back (120mm Panaflow L1A at 7V) gets trapped a little, making the ambient around the case higher than intake ambient.

07-17-2002, 10:43 AM	#12
redleader Thermophile Join Date: Jun 2001 Location: The deserts of Tucson, Az Posts: 1,264	What do the insides of Innovatek blocks look like anyway? Would they be much wider then 1/4ID tube?

07-17-2002, 10:49 AM	#13
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	Not sure. I can screw in the standard 1/2" barb fitting that comes with the Eheim, so perhaps it is not so bad.....

07-17-2002, 12:08 PM	#17
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	I'll do that this weekend (using 1/2" fittings, 1/2" ID tubing, BIX rad and 1048) and report on the results compared to the standard kit.

07-17-2002, 02:10 PM	#18
BillA CoolingWorks Tech Guy Formerly "Unregistered" Join Date: Dec 2000 Location: Posts: 2,371.493,106 Posts: 4,440	read it horizontally, from the blue curve to the magenta is what you will pickup in flow (from where you are at) (its actually interactive - but close enough) with barbed fittings the Innovatek is pretty much the same as the laybrinth types

07-17-2002, 02:39 PM	#20
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	The standard fittings on the Innovatek have a 1/4" ID size, but the standard tubing is 8mm ID (about 5/16"). Still bad, very bad. Say anonymous...how did you get to this wizardy in Excel?

07-17-2002, 06:25 PM	#21
JimS Cooling Savant Join Date: Jun 2001 Location: Connecticut Posts: 140	Put a good size reservoir just before the pump and you will eliminate any backpressure problem. For me, the advantages of a reservoir far outweigh any disadvantages.

07-17-2002, 07:03 PM	#22
awesterink Cooling Neophyte Join Date: Jul 2002 Location: Grand Rapids, MI Posts: 53	You got me curious....there are disadvantages to using a reservoir? Like what?

07-18-2002, 08:47 PM	#24
pHaestus Big Player Making Big Money Join Date: Aug 2001 Location: irc.lostgeek.com #procooling.com Posts: 4,782	I wouldn't "upgrade" an Innovatek block to any other block I have used. With 1/2" barbs I find the CPU and GPU blocks to perform extremely well, especially at lower volumetric flow rates (which are easily attainable with smaller pumps). The Rev3 just oozes quality.

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