Signs of too much flow

mfpmax · 11-12-2002, 05:24 AM

Since i'm cheap...if I make another rig, i'm thinking i'm gonna get a Maxi-Jet 900 instead of another 1200.

Now my question is, what are the common signs of too much flow.

Cathar · 11-12-2002, 06:02 AM

Huh? What's "too much flow" mean?

The higher the flow rate, the better the heat exchange for every component in your system, with some specific abberations.

There's such a thing as too much pump heat that overcomes the benefit of higher flow rates.

Or are we talking about block-splitting, hosing ripping itself of barbs kinda flow/pressure rates?

HMB · 11-12-2002, 06:55 AM

Some radiators work best within a given flow, but waterblocks always work better with greater flow so it might be positive, it might be negative. But it wont be much of a difference i guess.

mfpmax · 11-12-2002, 07:07 AM

too much flow...as in you're not letting your water cool down in the heat transfer medium...as HMB explains.

murray13 · 11-12-2002, 08:02 AM

If you don't think your flow is maximized for your system, and you think it's too high, just insert something to limit flow and test it out.

That is the ONLY way to know for sure.

If you are considering new components 'do your homework'. I know you've seen this many many times around the threads but it does pay off.

mfpmax · 11-12-2002, 08:27 AM

Quote:

Originally posted by murray13
If you don't think your flow is maximized for your system, and you think it's too high, just insert something to limit flow and test it out.

That is the ONLY way to know for sure.

If you are considering new components 'do your homework'. I know you've seen this many many times around the threads but it does pay off.

Can't really insert anything, the system is designed so tightly.

MadDogMe · 11-12-2002, 08:28 AM

I remember BillA saying 2.5GPM was good for a rad, 3 was excellent, 4 was not worth the efoort it would take to achieve it(meaning huge pump?). I'd say from that we've no worries about reaching the speed where coolant is moving to fast to cool (if it even does?, I thought that was theory?, not proven fact?). the graph line gets very flat towards the end...

**Edit**
To test/restrict all you'd have to do is clamp a hose...

mfpmax · 11-12-2002, 09:08 AM

clamping a hose causes it to buldge in which fear of springing a leak occurs

MadDogMe · 11-12-2002, 09:32 AM

Sheeeyiiit!, what kind of 'far eastern manufactured' hose you using?

, or what kind of pressure/head is your pump producing

..., anyways!, should'nt the pressure the pumps producing be the same regardless?...

bigben2k · 11-12-2002, 10:14 AM

I'll refer you all to this webpage.

If you keep the flow speed under 5 fps (feet per second), then everything should be fine.

There really isn't such a thing as too much flow. There is a point where the heat induced by the flow becomes a significant portion of the total heat load on a rig, but we're not there yet.

If your tubing expands because of the pressure caused by the flow, then you might be using tubing that is not meant for this kind of flow rate. If you can't find the specifications for your tubing, then I'd recommend upgrading to vinyl tubing (braided optional).

I'll be testing my pump (Little Giant 2-MDQ-SC) this weekend, with 3/4 inch ID braided tubing.

gmat · 11-12-2002, 11:03 AM

To clarify this, flow doesnt produce heat by itself (yeah, water friction against tube walls exists, but produces nearly zilch compared to CPU / Pump / etc.)
To get a high flow you need a high powered pump. A 80W pump will dump about twice as much heat into the water than a 40W pump, maybe more if it's not rated for high head. So make sure the rad can handle it, and waterblocks profit from the increased flow. Or else you get those dreaded diminishing returns.

(to make a broad comparison, if you put a more powerful engine in a car you must modify the transmission, maybe the gearbox, stiffen the chassis and improve the suspension and drive train.. or else...)

Can O' Beans · 11-13-2002, 12:57 AM

I can't see how a rad/HC can get worse results with higher flow (not including pump heat).

Yes, the water isn't cooled down as much per volumetric unit, BUT the water as a whole is now moving a lot quicker thru the rad - more heat removal overall for the system.

Pump heat output into the water is another part, but that entirely depends on your pump choice and restriction of the system as a whole (affecting pump's efficiency).

I'm actually going to try a few things when my lazy butt gets further along in my current WC system. One of the main things is to try to remove most of the pump heat before it can even get into the water.

gmat · 11-13-2002, 05:31 AM

Quote:

Originally posted by Can O' Beans
I can't see how a rad/HC can get worse results with higher flow (not including pump heat).

Actually it gets better results. But the higher the flow, the lower the increase in efficiency. (am i clear on that one ? it's rather difficult to explain with words)
So if you're already on the 'flat part' of the efficiency curve (ie already decent flow), doubling the flow will only increase rad efficiency by a few %. That means the rad MUST be already able to handle the increased heat produced by the more powerful pump...

Quote:

Yes, the water isn't cooled down as much per volumetric unit, BUT the water as a whole is now moving a lot quicker thru the rad - more heat removal overall for the system.

Exactly, what must be considered is the total volume of water cooled down each second, and not heat per volumetric unit.

Quote:

One of the main things is to try to remove most of the pump heat before it can even get into the water.

Unfortunately that implies active cooling = noise. Or manage to place the pump in the path of the heatercore exhaust...

Can O' Beans · 11-14-2002, 12:39 AM

Yeah, I understood the efficiency part, but I had no idea where his system stood on flow rate & efficiency.

I'm going to be testing pump cooling on my current pump - Danner MagDrive 500gph. Basically fan cooling (not overdoing it to add a lot of fan noise), but pump will need some modification. If we can remove the pump heat, that would be another plus.

Would be nice to see a magdrive pump with at least a SEMI-enclosed coil(or completely non-enclosed) with a radial heatsink on it so you could slap a 7v-ed 120mm on it. Of course, this couldn't serve as a submerged pump.

11-12-2002, 05:24 AM	#1
mfpmax Cooling Savant Join Date: Oct 2000 Location: Rockledge,FL,US Posts: 731	Signs of too much flow Since i'm cheap...if I make another rig, i'm thinking i'm gonna get a Maxi-Jet 900 instead of another 1200. Now my question is, what are the common signs of too much flow. __________________ My old and retired watercooling setup. Watercooled K6-2 450 at 600 Also Retired - Watercooling an XP1800@1782MHz

11-12-2002, 06:55 AM	#3
HMB Cooling Neophyte Join Date: Aug 2002 Location: Sweden Posts: 63	Some radiators work best within a given flow, but waterblocks always work better with greater flow so it might be positive, it might be negative. But it wont be much of a difference i guess. __________________ Asus A7N8X Deluxe, 2500+@2700MHz (13,5x200), 2x256MB TwinMOS pc3200 @ 2-2-2-11, GFFX 5900 Ultra@500/950MHz, 2x120GB Maxtor Diamondmax 9 8MB SATA raid-0, Prometeia, Logitech z-680. 20000+ 3Dmark2001 Duron 600@1521MHz RIP

11-12-2002, 07:07 AM	#4
mfpmax Cooling Savant Join Date: Oct 2000 Location: Rockledge,FL,US Posts: 731	too much flow...as in you're not letting your water cool down in the heat transfer medium...as HMB explains. __________________ My old and retired watercooling setup. Watercooled K6-2 450 at 600 Also Retired - Watercooling an XP1800@1782MHz

11-12-2002, 08:28 AM	#7
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	I remember BillA saying 2.5GPM was good for a rad, 3 was excellent, 4 was not worth the efoort it would take to achieve it(meaning huge pump?). I'd say from that we've no worries about reaching the speed where coolant is moving to fast to cool (if it even does?, I thought that was theory?, not proven fact?). the graph line gets very flat towards the end... Edit To test/restrict all you'd have to do is clamp a hose... Last edited by MadDogMe; 11-12-2002 at 08:33 AM.

11-12-2002, 09:08 AM	#8
mfpmax Cooling Savant Join Date: Oct 2000 Location: Rockledge,FL,US Posts: 731	clamping a hose causes it to buldge in which fear of springing a leak occurs __________________ My old and retired watercooling setup. Watercooled K6-2 450 at 600 Also Retired - Watercooling an XP1800@1782MHz

11-12-2002, 06:02 AM	#2
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Huh? What's "too much flow" mean? The higher the flow rate, the better the heat exchange for every component in your system, with some specific abberations. There's such a thing as too much pump heat that overcomes the benefit of higher flow rates. Or are we talking about block-splitting, hosing ripping itself of barbs kinda flow/pressure rates?

11-12-2002, 08:02 AM	#5
murray13 Cooling Savant Join Date: Sep 2002 Location: Omaha, NE USA Posts: 216	If you don't think your flow is maximized for your system, and you think it's too high, just insert something to limit flow and test it out. That is the ONLY way to know for sure. If you are considering new components 'do your homework'. I know you've seen this many many times around the threads but it does pay off.

11-12-2002, 09:32 AM	#9
MadDogMe Thermophile Join Date: Aug 2002 Location: Just shut up ;) ... Posts: 1,068	Sheeeyiiit!, what kind of 'far eastern manufactured' hose you using? , or what kind of pressure/head is your pump producing ..., anyways!, should'nt the pressure the pumps producing be the same regardless?...

11-12-2002, 10:14 AM	#10
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	I'll refer you all to this webpage. If you keep the flow speed under 5 fps (feet per second), then everything should be fine. There really isn't such a thing as too much flow. There is a point where the heat induced by the flow becomes a significant portion of the total heat load on a rig, but we're not there yet. If your tubing expands because of the pressure caused by the flow, then you might be using tubing that is not meant for this kind of flow rate. If you can't find the specifications for your tubing, then I'd recommend upgrading to vinyl tubing (braided optional). I'll be testing my pump (Little Giant 2-MDQ-SC) this weekend, with 3/4 inch ID braided tubing. __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

11-12-2002, 11:03 AM	#11
gmat Thermophile Join Date: Mar 2001 Location: France Posts: 1,221	To clarify this, flow doesnt produce heat by itself (yeah, water friction against tube walls exists, but produces nearly zilch compared to CPU / Pump / etc.) To get a high flow you need a high powered pump. A 80W pump will dump about twice as much heat into the water than a 40W pump, maybe more if it's not rated for high head. So make sure the rad can handle it, and waterblocks profit from the increased flow. Or else you get those dreaded diminishing returns. (to make a broad comparison, if you put a more powerful engine in a car you must modify the transmission, maybe the gearbox, stiffen the chassis and improve the suspension and drive train.. or else...)

11-13-2002, 12:57 AM	#12
Can O' Beans Cooling Savant Join Date: Aug 2002 Location: Pa - USA Posts: 264	I can't see how a rad/HC can get worse results with higher flow (not including pump heat). Yes, the water isn't cooled down as much per volumetric unit, BUT the water as a whole is now moving a lot quicker thru the rad - more heat removal overall for the system. Pump heat output into the water is another part, but that entirely depends on your pump choice and restriction of the system as a whole (affecting pump's efficiency). I'm actually going to try a few things when my lazy butt gets further along in my current WC system. One of the main things is to try to remove most of the pump heat before it can even get into the water.

11-14-2002, 12:39 AM	#14
Can O' Beans Cooling Savant Join Date: Aug 2002 Location: Pa - USA Posts: 264	Yeah, I understood the efficiency part, but I had no idea where his system stood on flow rate & efficiency. I'm going to be testing pump cooling on my current pump - Danner MagDrive 500gph. Basically fan cooling (not overdoing it to add a lot of fan noise), but pump will need some modification. If we can remove the pump heat, that would be another plus. Would be nice to see a magdrive pump with at least a SEMI-enclosed coil(or completely non-enclosed) with a radial heatsink on it so you could slap a 7v-ed 120mm on it. Of course, this couldn't serve as a submerged pump.

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