Need Water Block Design for 60L/M

[Snowman] · 05-22-2005, 03:25 PM

yep 60L/M the fitting is a 2 inch nippel so i have to make it like BladeRunner 1 2inch in and 4 1/2inch out. but is a spiral, cone and pins to rekomend?

jaydee · 05-22-2005, 04:09 PM

2in. x 2in. is larger than the socket for the CPU?

[Snowman] · 05-22-2005, 04:54 PM

Quote:

Originally Posted by jaydee116

2in. x 2in. is larger than the socket for the CPU?

yes the socket is 37,5mm and the nippel is 51mm

Cathar · 05-22-2005, 05:33 PM

Tail wagging the dog.

Brians256 · 05-23-2005, 12:09 AM

That'd be a decent pump to move water from a buried tank to cool multiple computers. For one computer...

Free advice: use really really good hose clamps.

Edit: Also, please take pics.

JSimmons · 05-23-2005, 08:55 AM

I knew a girl once in the Phillipines with a 2-inch nipple...

She had a good hose clamp, too, come to think of it...

maxSaleen · 05-23-2005, 07:32 PM

LOL! Nice one J.

Moto7451 · 05-24-2005, 08:21 PM

You'll need to adapt that to 1/2" or at least 3/4" or your tubing will be... absurd? I think thats a good word for it. BTW is that an electric pump or will it be run off of a gas engine? (the latter wouldn't be advisable seeing as you'll need to probably sell any children you have to pay for the fuel needed for a year of use)

Long Haired Git · 05-24-2005, 10:20 PM

Are you serious?

[Snowman] · 05-25-2005, 08:56 AM

Quote:

Originally Posted by Moto7451

BTW is that an electric pump or will it be run off of a gas engine?

its a electric 230V pump

Quote:

Originally Posted by Long Haired Git

Are you serious?

YES

The fitting is 35,5 mm outside and 20,5mm inside and the CPU is 37,5mm

Belenar · 05-25-2005, 09:07 AM

Hmmm ... an article comes in mind here ...
It's Cathar's 'Hom much pump power is enough' article ...
I don't even have to do the math on this one ... I'm sure it's too much.
With all the pressure you are also prone to tubing jumping off, ruptures of any kind, and so on ...
And the boost in performance (if any) over a normal pump won't really measure up to the boost on your electricity bill...
But then again ... I'd love to see you solve this problem.

[Snowman] · 05-25-2005, 12:43 PM

Quote:

Originally Posted by Belenar

With all the pressure you are also prone to tubing jumping off, ruptures of any kind, and so on ...

thats wy i want help to do a water block besign with high flow capability

jman1310 · 05-25-2005, 01:27 PM

be sure to use double layer nylon reinforced tubing and stainless worm drive hose clamps

Long Haired Git · 05-25-2005, 07:36 PM

The problem you'll have is getting the coolant close to the die.
See, 60LPM is no good if its no where near the heat.

Designs like the LRR, Cascade and **next big thing that I won't reveal*** all sacrifice some flow to accelerate the water, so that 5 LPM can come within 1mm of the copper of the block by it moving at 15 m/s. This causes a fair bit of backpressure to the pump, which causes the drop in flow rate.

If you take 60 LPM, and you try to do the same, you're going to drop down a fair bit of flow to get the velocity you require to get all the water close.
That pressure is then going to try to break your hoses, radiators and particularly the clamp for the water block to the motherboard.

Its like the footage of a single bloke trying to hold onto a fire hose as it whirls him around.

I've got no idea of the backpressure in this case, but given my garden hose on full-whelly struggles to fill my 40L storage container in two minutes, I'd suggest back pressure will be "significant".

Do you have a PQ graph for the pump? At what head height does it stop flowing?

This design will be totally impractical. You'll need to bolt that pump down, and you'll need industrial fasteners to bolt the motherboard to something strong and the block to both.

Lets just say the three-lugs on the old Socket-A are not going to suffice!

Other compromises are the radiators. I doubt they will hold the pressure you will be generating across them. I'd suggest a full car radiator - chances are the pipe sizes will roughly correlate too.

Which brings me to my last point. Remember the "pump selection criteria"?

How do you go:

1) Material of the pump won't cause galvanic corrossion
2) Silent
3) Reliable

etc etc....

[Snowman] · 05-26-2005, 04:11 PM

Quote:

Originally Posted by Long Haired Git

Do you have a PQ graph for the pump? At what head height does it stop flowing?
at 7 meters

How do you go:

1) Material of the pump won't cause galvanic corrossion: its a house water pump so ther no problem with galvanic corrossion
2) Silent: dont know but i will isolat it in a special box
3) Reliable: yes

i im thinking of having this desing on the CPU block

and the system is going to like this (sorry have no time to make it "nice" so paint dos the job)

MadHacker · 05-26-2005, 11:47 PM

Quote:

Originally Posted by Long Haired Git

...Its like the footage of a single bloke trying to hold onto a fire hose as it whirls him around....
... You'll need to bolt that pump down, and you'll need industrial fasteners to bolt the motherboard to something strong and the block to both.

Lets just say the three-lugs on the old Socket-A are not going to suffice!...

I don't think that will happen.
as you state "fire hose as it whirls him around" but that is due to the water leaving a closed system.
the rest of the long hose doesn't realy move much... don't ask me about the physics behind that... I don't know...
granted there will be a lot of back preasure.. but i don't think it will want to move once it is running.

Quote:

Originally Posted by Long Haired Git

...This design will be totally impractical....

This i do have to agree...

Quote:

Originally Posted by [Snowman]

i im thinking of having this desing on the CPU block and the system is going to like this (sorry have no time to make it "nice" so paint dos the job)

perhaps consider cooling your hardrives as well...
you do have one connection going directly from the CPU block to a radiator...
it isn't like you won't have enough flow

[Snowman] · 05-27-2005, 03:45 AM

The flow on the 1/2 is 15LPM and thats no problem.
but i dont realy get that with back preasure. (im not so brigt)
But i have a big problem the school wer i am going to make the block ar going to be clost to late agust

so i dont have any CNC to work with (i have A+ (MVG in 5 CNC curses

) yust have to brag) . so i have to make a chep test block with a rigd drill at my dads jobb :/
i gess i will have to weld the Hose Barbs to the block.
and i im going to macke a relay thas make the pump to do a slow start so it dosent pump 60L att the start, like after 30sec it have maximum power

FL3JM · 05-30-2005, 02:19 AM

If you try to run that pump at max power, 60 LPM, things are bound to break. Though if you manage to make it work without breaking, then hats off to you!

Please consider recording the first power up on video!

I can imagine people willing to chop your balls off if you manage to come with that rig to a LAN, the noise must be anything but silent. ^^

HammerSandwich · 05-30-2005, 10:05 AM

Quote:

Originally Posted by Long Haired Git

That pressure is then going to try to break your hoses, radiators and particularly the clamp for the water block to the motherboard.

Keeping everything sealed will be a problem, but the mobo clamp doesn't see this force with a fully contained WB. This would be a problem with direct-die cooling, and this pump would be a poor choice for that approach.

AFA successful sealing, I'd think about using threaded fittings in the components with AN adapters and hose. Twisting the WBs would be a concern.

Long Haired Git · 05-30-2005, 10:27 PM

For the WB design, I'd think about a ridge-less LRR design.
That is basically the waterblock you should above with four outlets.
However, you have a plate which forms a nozzle to accelerate the water and hence get more of it interacting with the base of the block.
I'd go for a slot rather than a hold, and aim to have the slot the width of the block, and then run it at full width initially, and then squeeze it in until the flow dropped off too much, and then bring it back a bit.

FL3JM · 06-02-2005, 05:29 AM

Have you thought about the pressure caused by the hoses when they try to straighten themselves out when you start pumping around that massive amount of water?
The hoses will get really really hard due to the pressure.

I thought of this as i saw a rolled up garden hose getting really stiff when waterpressure was applied. I tried to bend the hose before and after pressure was applied, and it was alot more difficult to bend it when the water was flowing through.

Not sure if this applies to a water cooling loop tho, food for thought anyways.

Correct me if im wrong!

mad mikee · 06-09-2005, 01:19 PM

(And I call myself one at least once a day

)

Easy way to show yourself using a high and low power fan.

Mount it about 1/2 - 1/4" away from a hard surface (maybe less, not sure) w/ airflow toward surface. Turn it on and place hand 1" away behind it.

1. Lower power fan will blow most of the air down and you won't feel much blowing back on your hand.

2. Higher power will feel more like the fan is blowing toward your hand instead of away. This is because the air and other resistence makes an obstruction to the flow, the harder you press on it, the harder it presses back.

I actually ran into this w/ a new TT Big Typhoon HSF when trying out different fans (Fins are close together (~ 1mm)). My Papst 100 CFM fan felt like it was blowing onto my hand instead of though the fins. Also very little flow was felt going through the fins.

My 120mm Panaflo M was the best balance, with most of the air going THROUGH the heatsink instead of backward out the top of the heatsink.

I think that this pump is extreme overkill, and you are tempting fate, but if you are determined some extra suggestions:
1. Go for the big (55+Gals), buried tank (extra structural support and cooling) and / or 1-2 car radiators for cooling.

1a. Where safety glasses and other equip when testing - this is a POWER tool

2. Leak / pressure test for a week (since if this does fail, it WILL be catastrophically) NO computer nearby !

3. Use a cheap system first to see how it works, then try it on something worthwhile

4. Since any failure would be major, YOU must inspect it directly @ least once / week.

5. WE WANT PICS !!!!!!!!!!!!!!

6. Post this @ xtremesystems.org in the liquid cooling section since it certainly qualifies as XTREME

Good luck

|kbn| · 06-09-2005, 07:58 PM

When the pump is turned on, if its not on a soft-start, the instant pressure will probably push your hoses off.

You will want hoseclips - maybe even two or three for each hose connection.

Look on the pump to see if it says how much current it takes?

I think the extra heat it will add will make this perform worse than a smaller pump.

FL3JM · 06-14-2005, 05:36 AM

Heat shmeat, who cares about performance, this is about who has got the biggest cohones!

Pull this one off and you will surely become famous.

Brians256 · 06-17-2005, 11:05 PM

FL3JM, you are indeed correct. Someone doing this will have cojones large enough to prevent normal walking!

05-22-2005, 03:25 PM	#1
[Snowman] Cooling Neophyte Join Date: Sep 2004 Location: Sweden Posts: 41	Need Water Block Design for 60L/M yep 60L/M the fitting is a 2 inch nippel so i have to make it like BladeRunner 1 2inch in and 4 1/2inch out. but is a spiral, cone and pins to rekomend?

05-23-2005, 08:55 AM	#6
JSimmons Cooling Savant Join Date: Apr 2003 Location: 4-sided room with an exit going east, and an exit going south Posts: 392	I knew a girl once in the Phillipines with a 2-inch nipple... She had a good hose clamp, too, come to think of it... __________________ My Watercooling Stuff

05-24-2005, 08:21 PM	#8
Moto7451 Cooling Neophyte Join Date: Jan 2005 Location: LA, CA Posts: 53	You'll need to adapt that to 1/2" or at least 3/4" or your tubing will be... absurd? I think thats a good word for it. BTW is that an electric pump or will it be run off of a gas engine? (the latter wouldn't be advisable seeing as you'll need to probably sell any children you have to pay for the fuel needed for a year of use) __________________ Athlon 64 3200+ Rev F @ 2.4Ghz (and climbing) @ 1.53v, DFI Infinity UltraII-M2, 512MB Apple Micron D9 DDR2 @ 400MHz (DDR2 800) 5-5-5-12 @ 2v, 128MB Radeon X800 @ 460/900, Dual 120GB HDs in RAID 0 Loop 1: D5 Pump • Bonneville w/ AC Heater Core • Swiftech MCW6002 • Gemini Tsunami VGA block • Gemini Tsunami S478 block (cold side chiller block) 110W 16v TEC @ 60W 12v Loop 2: PC Watercooling Shop 12v Water Pump • PCWCS 80.2 Rad • Gemini Tsunami (hot side chiller block)

05-24-2005, 10:20 PM	#9
Long Haired Git Cooling Savant Join Date: Jan 2003 Location: Sydney, Oz Posts: 336	Are you serious? __________________ Long Haired Git "Securing an environment of Windows platforms from abuse - external or internal - is akin to trying to install sprinklers in a fireworks factory where smoking on the job is permitted." (Prof. Gene Spafford) My Rig, in all its glory, can be seen best here AMD XP1600 @ 1530 Mhz \| Soyo Dragon + \| 256 Mb PC2700 DDRAM \| 2 x 40 Gb 7200rpm in Raid-0 \| Maze 2, eheim 1250, dual heater cores! \| Full specifications (PCDB)

05-25-2005, 09:07 AM	#11
Belenar Cooling Neophyte Join Date: Oct 2004 Location: Belgium Posts: 51	Hmmm ... an article comes in mind here ... It's Cathar's 'Hom much pump power is enough' article ... I don't even have to do the math on this one ... I'm sure it's too much. With all the pressure you are also prone to tubing jumping off, ruptures of any kind, and so on ... And the boost in performance (if any) over a normal pump won't really measure up to the boost on your electricity bill... But then again ... I'd love to see you solve this problem. __________________ The fastest cross team around: Team Supreme

05-22-2005, 04:09 PM	#2
jaydee Put up or Shut Up Join Date: Dec 2001 Location: Spokane WA Posts: 6,506	2in. x 2in. is larger than the socket for the CPU?

05-22-2005, 05:33 PM	#4
Cathar Thermophile Join Date: Sep 2002 Location: Melbourne, Australia Posts: 2,538	Tail wagging the dog.

05-23-2005, 12:09 AM	#5
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	That'd be a decent pump to move water from a buried tank to cool multiple computers. For one computer... Free advice: use really really good hose clamps. Edit: Also, please take pics.

05-23-2005, 07:32 PM	#7
maxSaleen Cooling Savant Join Date: Dec 2004 Location: Orlando, Florida Posts: 383	LOL! Nice one J.

05-25-2005, 01:27 PM	#13
jman1310 Cooling Savant Join Date: Oct 2004 Location: Sunny Florida Posts: 246	be sure to use double layer nylon reinforced tubing and stainless worm drive hose clamps

05-25-2005, 07:36 PM	#14
Long Haired Git Cooling Savant Join Date: Jan 2003 Location: Sydney, Oz Posts: 336	The problem you'll have is getting the coolant close to the die. See, 60LPM is no good if its no where near the heat. Designs like the LRR, Cascade and next big thing that I won't reveal* all sacrifice some flow to accelerate the water, so that 5 LPM can come within 1mm of the copper of the block by it moving at 15 m/s. This causes a fair bit of backpressure to the pump, which causes the drop in flow rate. If you take 60 LPM, and you try to do the same, you're going to drop down a fair bit of flow to get the velocity you require to get all the water close. That pressure is then going to try to break your hoses, radiators and particularly the clamp for the water block to the motherboard. Its like the footage of a single bloke trying to hold onto a fire hose as it whirls him around. I've got no idea of the backpressure in this case, but given my garden hose on full-whelly struggles to fill my 40L storage container in two minutes, I'd suggest back pressure will be "significant". Do you have a PQ graph for the pump? At what head height does it stop flowing? This design will be totally impractical. You'll need to bolt that pump down, and you'll need industrial fasteners to bolt the motherboard to something strong and the block to both. Lets just say the three-lugs on the old Socket-A are not going to suffice! Other compromises are the radiators. I doubt they will hold the pressure you will be generating across them. I'd suggest a full car radiator - chances are the pipe sizes will roughly correlate too. Which brings me to my last point. Remember the "pump selection criteria"? How do you go: 1) Material of the pump won't cause galvanic corrossion 2) Silent 3) Reliable etc etc.... __________________ Long Haired Git "Securing an environment of Windows platforms from abuse - external or internal - is akin to trying to install sprinklers in a fireworks factory where smoking on the job is permitted." (Prof. Gene Spafford) My Rig, in all its glory, can be seen best here AMD XP1600 @ 1530 Mhz \| Soyo Dragon + \| 256 Mb PC2700 DDRAM \| 2 x 40 Gb 7200rpm in Raid-0 \| Maze 2, eheim 1250, dual heater cores! \| Full specifications (PCDB)

05-27-2005, 03:45 AM	#17
[Snowman] Cooling Neophyte Join Date: Sep 2004 Location: Sweden Posts: 41	The flow on the 1/2 is 15LPM and thats no problem. but i dont realy get that with back preasure. (im not so brigt) But i have a big problem the school wer i am going to make the block ar going to be clost to late agust so i dont have any CNC to work with (i have A+ (MVG in 5 CNC curses ) yust have to brag) . so i have to make a chep test block with a rigd drill at my dads jobb :/ i gess i will have to weld the Hose Barbs to the block. and i im going to macke a relay thas make the pump to do a slow start so it dosent pump 60L att the start, like after 30sec it have maximum power

05-30-2005, 02:19 AM	#18
FL3JM Cooling Neophyte Join Date: Jul 2003 Location: Sweden Posts: 66	If you try to run that pump at max power, 60 LPM, things are bound to break. Though if you manage to make it work without breaking, then hats off to you! Please consider recording the first power up on video! I can imagine people willing to chop your balls off if you manage to come with that rig to a LAN, the noise must be anything but silent. ^^

05-30-2005, 10:27 PM	#20
Long Haired Git Cooling Savant Join Date: Jan 2003 Location: Sydney, Oz Posts: 336	For the WB design, I'd think about a ridge-less LRR design. That is basically the waterblock you should above with four outlets. However, you have a plate which forms a nozzle to accelerate the water and hence get more of it interacting with the base of the block. I'd go for a slot rather than a hold, and aim to have the slot the width of the block, and then run it at full width initially, and then squeeze it in until the flow dropped off too much, and then bring it back a bit. __________________ Long Haired Git "Securing an environment of Windows platforms from abuse - external or internal - is akin to trying to install sprinklers in a fireworks factory where smoking on the job is permitted." (Prof. Gene Spafford) My Rig, in all its glory, can be seen best here AMD XP1600 @ 1530 Mhz \| Soyo Dragon + \| 256 Mb PC2700 DDRAM \| 2 x 40 Gb 7200rpm in Raid-0 \| Maze 2, eheim 1250, dual heater cores! \| Full specifications (PCDB)

06-02-2005, 05:29 AM	#21
FL3JM Cooling Neophyte Join Date: Jul 2003 Location: Sweden Posts: 66	Have you thought about the pressure caused by the hoses when they try to straighten themselves out when you start pumping around that massive amount of water? The hoses will get really really hard due to the pressure. I thought of this as i saw a rolled up garden hose getting really stiff when waterpressure was applied. I tried to bend the hose before and after pressure was applied, and it was alot more difficult to bend it when the water was flowing through. Not sure if this applies to a water cooling loop tho, food for thought anyways. Correct me if im wrong!

06-09-2005, 01:19 PM	#22
mad mikee Cooling Savant Join Date: Oct 2001 Location: US of A!!!! Posts: 146	Intuitive Backpressure description for idiots (And I call myself one at least once a day ) Easy way to show yourself using a high and low power fan. Mount it about 1/2 - 1/4" away from a hard surface (maybe less, not sure) w/ airflow toward surface. Turn it on and place hand 1" away behind it. 1. Lower power fan will blow most of the air down and you won't feel much blowing back on your hand. 2. Higher power will feel more like the fan is blowing toward your hand instead of away. This is because the air and other resistence makes an obstruction to the flow, the harder you press on it, the harder it presses back. I actually ran into this w/ a new TT Big Typhoon HSF when trying out different fans (Fins are close together (~ 1mm)). My Papst 100 CFM fan felt like it was blowing onto my hand instead of though the fins. Also very little flow was felt going through the fins. My 120mm Panaflo M was the best balance, with most of the air going THROUGH the heatsink instead of backward out the top of the heatsink. I think that this pump is extreme overkill, and you are tempting fate, but if you are determined some extra suggestions: 1. Go for the big (55+Gals), buried tank (extra structural support and cooling) and / or 1-2 car radiators for cooling. 1a. Where safety glasses and other equip when testing - this is a POWER tool 2. Leak / pressure test for a week (since if this does fail, it WILL be catastrophically) NO computer nearby ! 3. Use a cheap system first to see how it works, then try it on something worthwhile 4. Since any failure would be major, YOU must inspect it directly @ least once / week. 5. WE WANT PICS !!!!!!!!!!!!!! 6. Post this @ xtremesystems.org in the liquid cooling section since it certainly qualifies as XTREME Good luck __________________ His PCV2000 DFI ULTRA D (SLI Modded just because) 4200 x2 @ ~ 10 x 260 /1.49 Vcore , G5 + PA160 + 150mm Papst@5V + Maze 4 GPU + mcp350. Temp 36-37(load-2 x D20L) 2x1GB Ballistix PC4000, x800xl 256m, 74G rap + 160G Hitachi SATA , 21" mit/nec 1920x1440 Crt + 19 Neovo F-419 1280x1024 LCD, OCZ Powerstream 600 Hers: (my beloved who lets me do all this) LianLi PCV1000, SD3700 @ 9x300, 1GB GBLA, 250+ 120 SATA, X700pro, 19" Neovo F-419 LCD, another OCZ 600

06-09-2005, 07:58 PM	#23
\|kbn\| Cooling Savant Join Date: Feb 2004 Location: W. Sussex, UK Posts: 329	When the pump is turned on, if its not on a soft-start, the instant pressure will probably push your hoses off. You will want hoseclips - maybe even two or three for each hose connection. Look on the pump to see if it says how much current it takes? I think the extra heat it will add will make this perform worse than a smaller pump. __________________ Waterblocks

06-14-2005, 05:36 AM	#24
FL3JM Cooling Neophyte Join Date: Jul 2003 Location: Sweden Posts: 66	Heat shmeat, who cares about performance, this is about who has got the biggest cohones! Pull this one off and you will surely become famous.

06-17-2005, 11:05 PM	#25
Brians256 Pro/Staff Join Date: Oct 2001 Location: Klamath Falls, OR Posts: 1,439	FL3JM, you are indeed correct. Someone doing this will have cojones large enough to prevent normal walking!

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