Work in progress - Extreme Mod - Page 2

airspirit · 11-18-2002, 10:47 PM

Woohoo! The new case is installed and working wonderfully, with my laser printer sitting on top of the stack. Now all I need to do is install my Caravels and I'll be all set. This case is frickin' quiet, too. I'll have pictures up on Wednesday for you all (maybe tomorrow if I can sneak home with the digicam at lunchtime). I strung the hoses in through some empty PCI slots, so there are no holes on the sides of the case. The only problem with that was that I used more Clearflex 60 than I wanted to ... I'm going to have to buy more to put the third case on, but that's a while down the road. *sigh*

I do have to say that this case doesn't breathe very well. I'm running 37C on my CPU, and 39 in the case. Very odd.

airspirit · 11-19-2002, 11:41 AM

Pics will be up in a couple of hours. I'm going to have to run home, and I'll be snappin' pix then.

airspirit · 11-19-2002, 01:14 PM

Here are the pics with the case style I'll be using. To envision the end of the project, try to picture three of those identical desktops stacked on each other with the printer up top. Here's the pics:

airspirit · 11-19-2002, 01:17 PM

I'm still using the desk fan until the Caravels come in. It isn't really a good way of doing the cooling since the cooling chest is a closed box. I blow air in one side, and only a tiny flow comes out. I estimate I'm getting less than 20 CFM net, and this is a radiator than needs 100s of CFM to function correctly (I'll be at a peak of 1100 CFM with the two fans I'll be installing). If I'm getting 38C hard idle now, I wonder how low it'll hover when allowed to breathe.

And striking a sexy pose here:

airspirit · 11-19-2002, 01:20 PM

Story about the router:

It feeds to a seperate 4pt WAP router, a 4/5pt hub, and a second dedicated WAP. There is too much networking going on in my borgpartments. That is going to be the greatest benefit is that it'll clean up my cable mess significantly.

bigben2k · 11-19-2002, 01:52 PM

Nice work!

So, is it quiet?

mo · 11-19-2002, 02:11 PM

Excellent. Very neat... I really like the way it turned out. And with the wooden container its probably very quiet

airspirit · 11-19-2002, 03:08 PM

If I turn off the desk fan, it is nearly silent. There is slight vibration from the pump, but your ear must be within 1ft of the box to barely hear it. The front case fan is 5V and the rear fans are temp controlled by the PSU (Antec Truepower). I'm putting the Caravels on a ceiling fan controller, so overall I suspect that even when I have all three PCs on the loop, it will be less noisy than one typical OEM PC.

airspirit · 11-20-2002, 08:52 PM

It's funny, but the loudest thing in the case is the hard drive. I didn't realize how noisy that bastard was until it got to sing solo, so to speak. I spent a couple of hours wracking my brain trying to figure out why the bugger was making any noise at all until it finally hit me. Go figure.

Can O' Beans · 11-20-2002, 11:30 PM

Nice!

Looks good, AND is functional (table, etc..)

airspirit · 11-24-2002, 01:59 PM

Just got the ball rolling on a new pump ($150 Gen-X Mak 4 for only $65 shipped, hehe), I have the Caravels on the way, and in a week I'm ordering the block/clamp/case for machine number 2. By December 15th, I should be done with the second phase of my Evil Project.

airspirit · 12-02-2002, 12:41 PM

The Caravels and pump are here. The fans pump air like crazy (do you know how fast a desk can be cleared of paperwork? I do, now ... ), and they are not quiet by no stretch of the imagination. They are like a middle/loud HSF, or a vacuum at about 100 ft. I'm going to set up a fan controller tonight in a non-regulated power strip that should allow me to quiet these mofos down. With how much air they pump, I should be able to slow them significantly (making them virtually silent in the confines of the cooling box) while retaining excellent air flow between the two of them.

The pump might take a little longer before I have real statistics on it since I won't be ordering the second case/block/clamp until tomorrow (it'll be complete by the middle of next week, or faster if Becooling is on the ball).

I may set up a dummy second loop with my BIX in place of the second water block tonight if I'm really bored, and string the new pump into it. It'll allow me to see what kind of flow to expect out of the new pump, as well as sound check it. The only thing I dread about that is the risk of miscutting the tubing (I don't want to order any more!!!), and having to drain and purge the system, however easy that may be. Plus, it'll throw off my temp measurements (they'll be completely invalid for this test) since instead of a second block adding about 60W of heat into the system, I'll have a radiator pulling some out. This will give me a chance to check the validity of the much maligned bypass theory, though. By design, the entire loop should now be at about 1" ID except for the individual parallel loops. This pump better not suck. No, wait, I mean, aw, f*ck it.

airspirit · 12-02-2002, 11:28 PM

I have the new pump and fans on it. The fans are tuned to 50% speed, and the second loop (also new) has a BIX to simulate the resistance of a second block. Since there is no air going through it and it is wrapped in four layers of towel, it won't be doing much cooling. I opened the bypass to 30%. Time to do some prime95 and see where the temps take me.

airspirit · 12-03-2002, 12:13 AM

Idle: 34C
Peak: 42C

I'm going to try this again later with the water flow optimized further and the fans all the way up. Mwahaha!

airspirit · 12-03-2002, 01:40 PM

Idle: 34C
Peak: 41C

That's all, folks. That seems to be my best with everything tuned perfectly. The pump is in horrible condition (EBay unit) so I am returning it to the vendor for a refund and I have a nice shiny new one coming in a week with the other parts. The fans buzz something fierce from the controller, so I'm probably going to pick up a Variac to do the job right. In the meantime I'm going to go back to my old pump and plug my fans directly into the wall. The noise from the fans and pump combined sounds like a circular saw at about 20 ft ... very f*cking loud, and I can't hear myself think over it. I would estimate the sound levels to somewhere in the 70 dB range or higher (I've been in busy woodshops that were quieter).

All the parts are now on the way, though, so I'll have much to share next week.

airspirit · 12-07-2002, 01:10 PM

I've finished phase 2. I'll be posting some pictures soon.

The hardest part of that phase was dremeling the damn cartridge off of an old Slot A Athlon 700 ... the bastard wouldn't come loose!

So far, there is little to no temperature change, but I'm still normalizing the reservoir temperature for testing purposes and trying to tune the bypass volume.

I'll have pictures and data early next week.

iroc409 · 12-09-2002, 02:56 AM

man, definitely props on the setup, it looks nice.

one piece of advice tho, with such a nice box, get rid of the duct tape

duct tape rules, and is the fixer of all, but get some nice polished hose clamps or soemthing

airspirit · 12-09-2002, 08:44 AM

The duct tape on the hoses protects the hoses from damage from the hose clamps (it is soft silicone) and damage from the entry/exit points from the case (sharp metal edges). The only other duct tape is holding my res manifold down onto my res, and I don't think they make hose clamps that big.

Arcturius · 12-09-2002, 10:01 AM

Airspirit:
While reading your article, and your comments about the restrictiveness of the radiator, I noticed that it looks the radiator is gravity fed. Am I correct in this, or do I have inlet/outlet confused in my understanding of Danner pumps?

Case "Gravity fed radiator"
Have you considered simply adding another inline pump to push the water through the radiator, as opposed to a bigger pump pushing the water up through your PCs, to the cooling tower?

Case "me confused"
I'm doubting this, since this case would mean that the pump is trying to push water into the res, and then the PCs are gravity fed...and the warm water rises to the top to be fed to the PCs...this must be wrong

Case "me really confused"
In this case, I'm simply having difficulty following hoses inside your cooling box, and I'm all messed up.

airspirit · 12-09-2002, 02:39 PM

This is how it works. The reservoir serves as a gravity trap. What I mean by that is that the weight of the water in the res forces down through the tubing and piping in the cooling chest and primes the pump. Once the pump is turned on, it just needs to pump the slug of air out of the block loops and any excess air out of the radiator, and it is bled. The pump is fed a guaranteed supply from the res, and it pumps directly up to the blocks which drain into the res. The only effort that pump has to make is to pump upwards through the blocks, and suck water through the radiator and the restricted bypass. If the bypass is opened fully, nearly no water is pulled through the radiator. If the bypass is only partially open causing enough restriction to the pumps suction, it will also pull water through the radiator to make up for the slack. The one thing that the pump needs never worry about, though, is its water supply, because even if the bypass is completely closed, water will be forced by gravity through the radiator providing an endless supply for it to pump.

Does that make sense?

iroc409 · 12-09-2002, 07:18 PM

actually, they make hose clamps to clamp guages (like a tach) onto steering columns and the like, that might just work. then maybe you can put some padding over the hose so it doesn't get damaged, like some foam.

but still, i like the setup

airspirit · 12-10-2002, 05:14 PM

Duct tape is doing good for me, and I don't mind the look. The benefit is that since it is so thin I can gauge the compression of the hose by the clamp to confirm adequate pressure on the hose.

I may change it in the future, and probably will need to when I start using the bottom port on the res. manifold. Until then, I don't see much of a point. Besides, with the positioning of my desks, it is hard to notice anyway.

airspirit · 12-12-2002, 04:31 PM

Here are some goodies for mah bruthaz:

airspirit · 12-12-2002, 04:31 PM

And again:

airspirit · 12-12-2002, 04:32 PM

And inside the box:

11-18-2002, 10:47 PM	#26
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Woohoo! The new case is installed and working wonderfully, with my laser printer sitting on top of the stack. Now all I need to do is install my Caravels and I'll be all set. This case is frickin' quiet, too. I'll have pictures up on Wednesday for you all (maybe tomorrow if I can sneak home with the digicam at lunchtime). I strung the hoses in through some empty PCI slots, so there are no holes on the sides of the case. The only problem with that was that I used more Clearflex 60 than I wanted to ... I'm going to have to buy more to put the third case on, but that's a while down the road. sigh I do have to say that this case doesn't breathe very well. I'm running 37C on my CPU, and 39 in the case. Very odd. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

11-19-2002, 11:41 AM	#27
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Pics will be up in a couple of hours. I'm going to have to run home, and I'll be snappin' pix then. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

11-19-2002, 01:20 PM	#30
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Story about the router: It feeds to a seperate 4pt WAP router, a 4/5pt hub, and a second dedicated WAP. There is too much networking going on in my borgpartments. That is going to be the greatest benefit is that it'll clean up my cable mess significantly. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

11-19-2002, 01:52 PM	#31
bigben2k Responsible for 2% of all the posts here. Join Date: May 2002 Location: Texas, U.S.A. Posts: 8,302	Nice work! So, is it quiet? __________________ WBTA (Water Block Testing Alliance) Nordic Hardware WC 101

11-19-2002, 03:08 PM	#33
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	If I turn off the desk fan, it is nearly silent. There is slight vibration from the pump, but your ear must be within 1ft of the box to barely hear it. The front case fan is 5V and the rear fans are temp controlled by the PSU (Antec Truepower). I'm putting the Caravels on a ceiling fan controller, so overall I suspect that even when I have all three PCs on the loop, it will be less noisy than one typical OEM PC. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

11-19-2002, 02:11 PM	#32
mo Cooling Savant Join Date: Aug 2002 Location: Canada, Montreal Posts: 136	Excellent. Very neat... I really like the way it turned out. And with the wooden container its probably very quiet

11-20-2002, 08:52 PM	#34
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	It's funny, but the loudest thing in the case is the hard drive. I didn't realize how noisy that bastard was until it got to sing solo, so to speak. I spent a couple of hours wracking my brain trying to figure out why the bugger was making any noise at all until it finally hit me. Go figure. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

11-20-2002, 11:30 PM	#35
Can O' Beans Cooling Savant Join Date: Aug 2002 Location: Pa - USA Posts: 264	Nice! Looks good, AND is functional (table, etc..)

11-24-2002, 01:59 PM	#36
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Just got the ball rolling on a new pump ($150 Gen-X Mak 4 for only $65 shipped, hehe), I have the Caravels on the way, and in a week I'm ordering the block/clamp/case for machine number 2. By December 15th, I should be done with the second phase of my Evil Project. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-02-2002, 12:41 PM	#37
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	The Caravels and pump are here. The fans pump air like crazy (do you know how fast a desk can be cleared of paperwork? I do, now ... ), and they are not quiet by no stretch of the imagination. They are like a middle/loud HSF, or a vacuum at about 100 ft. I'm going to set up a fan controller tonight in a non-regulated power strip that should allow me to quiet these mofos down. With how much air they pump, I should be able to slow them significantly (making them virtually silent in the confines of the cooling box) while retaining excellent air flow between the two of them. The pump might take a little longer before I have real statistics on it since I won't be ordering the second case/block/clamp until tomorrow (it'll be complete by the middle of next week, or faster if Becooling is on the ball). I may set up a dummy second loop with my BIX in place of the second water block tonight if I'm really bored, and string the new pump into it. It'll allow me to see what kind of flow to expect out of the new pump, as well as sound check it. The only thing I dread about that is the risk of miscutting the tubing (I don't want to order any more!!!), and having to drain and purge the system, however easy that may be. Plus, it'll throw off my temp measurements (they'll be completely invalid for this test) since instead of a second block adding about 60W of heat into the system, I'll have a radiator pulling some out. This will give me a chance to check the validity of the much maligned bypass theory, though. By design, the entire loop should now be at about 1" ID except for the individual parallel loops. This pump better not suck. No, wait, I mean, aw, f*ck it. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-02-2002, 11:28 PM	#38
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	I have the new pump and fans on it. The fans are tuned to 50% speed, and the second loop (also new) has a BIX to simulate the resistance of a second block. Since there is no air going through it and it is wrapped in four layers of towel, it won't be doing much cooling. I opened the bypass to 30%. Time to do some prime95 and see where the temps take me. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-03-2002, 12:13 AM	#39
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Idle: 34C Peak: 42C I'm going to try this again later with the water flow optimized further and the fans all the way up. Mwahaha! __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-03-2002, 01:40 PM	#40
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Idle: 34C Peak: 41C That's all, folks. That seems to be my best with everything tuned perfectly. The pump is in horrible condition (EBay unit) so I am returning it to the vendor for a refund and I have a nice shiny new one coming in a week with the other parts. The fans buzz something fierce from the controller, so I'm probably going to pick up a Variac to do the job right. In the meantime I'm going to go back to my old pump and plug my fans directly into the wall. The noise from the fans and pump combined sounds like a circular saw at about 20 ft ... very f*cking loud, and I can't hear myself think over it. I would estimate the sound levels to somewhere in the 70 dB range or higher (I've been in busy woodshops that were quieter). All the parts are now on the way, though, so I'll have much to share next week. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-07-2002, 01:10 PM	#41
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	I've finished phase 2. I'll be posting some pictures soon. The hardest part of that phase was dremeling the damn cartridge off of an old Slot A Athlon 700 ... the bastard wouldn't come loose! So far, there is little to no temperature change, but I'm still normalizing the reservoir temperature for testing purposes and trying to tune the bypass volume. I'll have pictures and data early next week. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-09-2002, 02:56 AM	#42
iroc409 Cooling Savant Join Date: Oct 2002 Location: midwest side, yo Posts: 596	man, definitely props on the setup, it looks nice. one piece of advice tho, with such a nice box, get rid of the duct tape duct tape rules, and is the fixer of all, but get some nice polished hose clamps or soemthing

12-09-2002, 08:44 AM	#43
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	The duct tape on the hoses protects the hoses from damage from the hose clamps (it is soft silicone) and damage from the entry/exit points from the case (sharp metal edges). The only other duct tape is holding my res manifold down onto my res, and I don't think they make hose clamps that big. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

12-09-2002, 10:01 AM	#44
Arcturius Cooling Savant Join Date: Aug 2001 Location: Chippewa Falls, Wisconsin Posts: 294	Airspirit: While reading your article, and your comments about the restrictiveness of the radiator, I noticed that it looks the radiator is gravity fed. Am I correct in this, or do I have inlet/outlet confused in my understanding of Danner pumps? Case "Gravity fed radiator" Have you considered simply adding another inline pump to push the water through the radiator, as opposed to a bigger pump pushing the water up through your PCs, to the cooling tower? Case "me confused" I'm doubting this, since this case would mean that the pump is trying to push water into the res, and then the PCs are gravity fed...and the warm water rises to the top to be fed to the PCs...this must be wrong Case "me really confused" In this case, I'm simply having difficulty following hoses inside your cooling box, and I'm all messed up.

12-09-2002, 07:18 PM	#46
iroc409 Cooling Savant Join Date: Oct 2002 Location: midwest side, yo Posts: 596	actually, they make hose clamps to clamp guages (like a tach) onto steering columns and the like, that might just work. then maybe you can put some padding over the hose so it doesn't get damaged, like some foam. but still, i like the setup

12-10-2002, 05:14 PM	#47
airspirit Been /.'d... have you? Join Date: Jul 2002 Location: Moscow, ID Posts: 1,986	Duct tape is doing good for me, and I don't mind the look. The benefit is that since it is so thin I can gauge the compression of the hose by the clamp to confirm adequate pressure on the hose. I may change it in the future, and probably will need to when I start using the bottom port on the res. manifold. Until then, I don't see much of a point. Besides, with the positioning of my desks, it is hard to notice anyway. __________________ #!/bin/sh {who;} {last;} {pause;} {grep;} {touch;} {unzip;} mount /dev/girl -t {wet;} {fsck;} {fsck;} {fsck;} {fsck;} echo yes yes yes {yes;} umount {/dev/girl;zip;} rm -rf {wet.spot;} {sleep;} finger: permission denied

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