How to do disaster management?

[Solarian League]

Originally posted by Pyrotechnic
If you still want watercooling, consider using a separate case, like a external SCSI drive enclosure to house the radiator and pump, and run to lines to the computer, this minimizes the amount of places leaks could occur within the case, you only have the waterblock, and a properly sealed block should be no problem. Of course, setup the temperature and fan shutdowns, and that should be pretty safe. You could even put a few leak dectors on the block, which would be semi effective since this is the only place for leaks within the computer.

i would love to have an exterial watercooling system atleast it would make it easier for me to bleed off crap and refill etc... and probably safer but problem is dorm room is very cramped so i'm going to have to pack it into my computer but anyway i'll probably never do any of this stuff heh just throwing the idea in

anyway i think the best way to prevant failure is to just do a regular mantaiance check say onice a week, you just pop open the case check the horse, check the comps to make sure they look like theyre functing correctly etc. or maybe extend that peroid to onice a month do a check up but then again if you bother the system more chance of failure so.... but if you do regularly check upon the system u could possiable avert an severe system failure

[Gooserider]

Many folks may view me as excessivly paranoid for even starting this thread, but I do see a need for balance in design. Just where the balance is drawn can be debated, but it is easy to stake out the extremes.

On the one side, not taking ANY precautions makes the system an expensive failure waiting to happen. The other extreme is to add so many safety functions that the system becomes overwhelmed and non-functional. It is not unlike the debate in the computer security business between ease of use and security, it is difficult to do both.

My opinion is that it is important to have failure detection, and to design so as to reduce the odds of failure. However good design is more critical than failure detection (it is better NOT to have the failure than it is to detect it!)

Failure detection should be limited to failures that are fairly likely. Shouldn't be expensive, and must not be so complex as to become a possible failure source of their own. Ideally it should provide useful information even when NOT detecting a failure. A thermometer is a good example - it tells how hot the system is normally, but can trigger a shutdown if it goes over range. A leak detector isn't good this way, but it provides a valuable and low cost safety function.

Solarian League is IMNSHO going way into the overkill range, and possibly making some bad design decisions as well.

Solarian League: i'm going to get an water cooling system that probably will have pelts on it

Pelt runners might argue with me, but my reading has led me to the conclusion that UNLESS one is doing really radical OC stuff, they are a BAD idea. They seem to add a multitude of additional failure points, consume great amounts of extra power, etc. Looks to me like avoiding them will in and of itself produce a more reliable system.

for the leaking detector,...place one below each barb joint....now you could place a little funnel below both barb then have the leak sensor in the neck of the funnel.

This would get real complex real fast, especially since alot of small leaks start as 'sprayers' that might not hit your funnel. If I *WAS* to do something like that; I would be more inclined to wrap the plumbing in absorbent cloth of some sort (maybe some of that 'tube bandage' stuff?) that would absorb water and wick it to the low point on each tube. The sensor could then be a couple of bare wires on the side of the tube, which is cleaner.

However this seems overly complex, the water will reach the bottom of the case soon enough, so put a sensor or two there and be done with it. (My server case has two chambers, so I'd need a sensor for each side) If one wanted to get a bit fancier, the pump and the rad seem to be the biggest potential leakers, so put each in a tray and add a sensor to the trays.

...you could optionaly have an additional setup. ... put in a fewlights, ... small say 9 volt battery ... hook the lights up into ur auto kill system ... and one of the light on the little pannel comes on and it will stay on ... so the user will know why the system triped and where the leak was.

It's called a 'First out' system, and I first encountered them back in the late 70's, when they were done with relays that made a very complex and expensive setup. With modern circuitry it would probably be a lot easier / cheaper, but still adds another layer of complexity which isn't really needed - you can SEE a leak, etc. Most of the failure modes we are talking about are pretty obvious with minimal testing.

the total cost should be i hope maybe $50 or less, depending on what you want to do with it i guess. its just an idea, and admitly i like the idea myself, i'm newbie to watercooling and this idea might help me push my watercooling system to.

I think you are WAY underestimating the cost of what you are describing, It looks to me like over $100 just for sensors and indicators...

I'm thinking more in terms of ONE relay, and maybe a 'wall wart' power supply. A $10 flood alarm from Home Depot. The rest of the sensors would be in the system already for other monitoring, and just feed the 'kill relay' as an extra step.
------------------

BladeRunner: I think that maybe you are all going overboard a little.

I don't think I am, but agree that SL is... However it's debateable.

I'll have to do an air pressure test with my clipless pipes but I can assure you they will not pop off, slip off, or leak, Just think about it with the barb in a vice using all my strength and body weight all I can do is stretch the tube a bit.

Ever hear of a 'Chinese Finger Trap'? Pulling on the hose is like trying to get out of one with a straight pull. Water pressure is going to work more in a direction to make the hose slip.

I used to have the whole system with barbs done in a similar fashion, and none ever leaked or came off. I much prefer the festo push fittings now because I can make smaller neater blocks and connect and disconnect them quickly & easily without worry of leaks. The tube isn't an issue once you have some of the correct stuff, I think it was about £15 for 20 meters of the 8mm

As I said, I'm not arguing (much ;-) ) with your methods, just commenting that they aren't to my preference. I'll admit that what you are doing has worked well so far, which is a strong point in it's favor.

The liquid level sensor as said I haven't done yet but I'll be using a sealed optical type liquid sensor I have here atm.

Do you have a reccomended source / brand, or is it just something you happened to find? I'm in the process (in a different window) of getting a nice flow sensor for $10 here http://sales.goldmine-elec.com/Index.asp as I send this message. It's a rotor type, and I'm not sure how easily it can do flow measurement, but it has an adjustable flow failure detection relay and appears to be a nice unit otherwise.

Part of the reason for soldering my blocks is with the more complex shapes of the VGA ram blocks and Splitter-X in my system, it would be very difficult to make a seal grove without access to CNC, and like you say the blocks can be thinner walled and smaller overall if soldered.

Your blocks certainly are beautiful in the pictures, your finishing work is incredible. Aside from looks, is there a reason for such a high polish on the non-contact surfaces? (I agree with lapping the contact surface, but was thinking more in terms of wire wheeling the rest...) I also noticed that you say you pressure test before finishing, but not after. Seems to me like there is a slight but present risk that a leak could be created by disturbing/removing the solder that was covering it before finishing. Is there reasoning for doing it this way, or did you leave something out of your process descriptions?
As to seal grooves, I wasn't planning on them. I was planning to use flat surfaces with a liberal layer of RTV or aviation Form-a-gasket between them, with 6/32 bolts holding things together. I might also use a thin gasket, but didn't think it was really needed.

It doesn't really matter what target you set yourself but I think you have to be realistic when using water cooling and that's why I say "major disaster every two years". I would probably double that to every four years if I just built the system and then didn't dick around with it at all, but as I'm constantly modding it and upgrading parts of it adding new blocks etc, the possibility of a screw up on my part is much higher. It wont change the world but if you beat your target you'll get a nice warm feeling about it too.

Seems reasonable to me. I'd like to get at least 4 years between major failures, but then I'm not planning to mess with the system much once it's built.
-------------------

Solarian League you might think it is overkill but heh if its what $50 to say $100 to do a little bit of disaster prevation

The problem is that what you describe is NOT 'failure prevention' but 'failure detection'. It appears to me that you are working to hard on the former, and not enough on the latter.

its well worth it like that funnel and the water sensor with little lightboard on front of computer heh atleast if something goes to hell you will know what it probably was before opening the case

If the 'kill relay' tripped, odds are I'm going to need to open the case anyways. I can wait to find out the bad news until then :shrug: Your proposed 'first out' system appears to me like it would add little value compared to it's cost.
-----------------

bigben2k Here's the latest flow switch

Thanks, I'm getting one from the last source mentioned there.
[quote]bigben2k: No, a hose split is not reasonable, neither is a blow off. To make sure it doesn't happen you use hose clamps and forget about it. That's what I meant by "good design". [quote]
Not saying it's LIKELY, but I am saying it's possible. It isn't WC'ing a PC, but I've seen each of those failures in automotive fluid systems. The stresses are lower, but I don't see anything magic in a WC system that makes them impossible. (But I'm planning design to make them less likely!)

What has happened historically, is that a brass barb in an Aluminium top would corrode, and break a seal: just don't do it. Hose split has happened once, when the water was allowed to get so hot, that it melted the tubing, so don't allow that to happen either, that's all. The same heat can melt a plastic top, and cause an o-ring failure: I've seen that twice. They were the result of either a lack of OCP, or OCP failure.

Thanks for the historical record, it is useful. Is this just your experience, or has there been some effort to collect this data from multiple sources? (IOW, how big is the sample?)
I am planning to do all copper and brass, so corrosion shouldn't be an issue, I'm also not planning on O-rings or plastic parts other than PVC pipe and the braided poly reinforced hose. I'm guesstimating that I would set my 'panic shutdown' water temp around 50-75*C, which should be well within safe limits for the plumbing.
I assume that OCP is 'Over Current Protection', correct? Is there a low cost way of ensuring that it works?

Thanks,

Gooserider

[Solarian League]

haha i aknowledge that i'm going overboard, but i tend to do that

i probably as in 99.9% won't even do any of the ideas that i have presented in this thread, its just some idea that i throwed into the mix

anyway what probably will happen is i will setup the watercooling system and make sure its very secure and double or triple check everything etc... then turn it on and keep a eye on it for the first few days onice it seems to be working good close up then do a mantaiance check say onice a week to onice a month or so.

i think that is an fairly reasonable setup, along with the always important temperature probe on the coldblock and the waterblock.

probably will buy digidoc5+ and stick those 8 probe on the system, probably 2 on the cpu, flat one on the coldplate, then another one on the waterblock, another one on video card, then a couple lying around to check ambident temperature, maybe one on northbridge chip etc... then if any of the temperature is abnormably high shut down the system and i'll open it up and check out why.

i think that is an pretty reasonable safty check heh, that along with including those good ol worm clamp on the hoses.


*shungs* i guess my first reaction is to setup an increditable complex and uber system that will detect repair every single failure possiable, heh its just a little fault of mine oh well that's life.

anyway do u guys think my procedudle above is more reasonable


now to address your comment about the pelt idea: well yes they might add some complexity to the system but imho the benefit is well worth it, some users are able to get sub zero temperature with a pelt system. and i'm planning on majorly oc' my cpu. and consdering the temperature outhere is hotter than hell, that subzero temp that some people get, i probably won't but atleast it will be cooler than just watercooling so.... worth it in my option.

[bigben2k]

For an OCP (CPU Overheat Protection) alternative, someone suggested bolting a thermal switch to the CPU block: too hot and it shuts down the power supply. The switch is designed specifically for a set temperature, and I think it's available from www.digikey.com .

[Gooserider]

Pyrotechnic: I'm not sure if a leak detector would work 100%

You make a good point, and are correct about the problem of detecting a small leak. Presumably protecting against that is where the conformal coating discussion comes in. The other option which might be possible is to put some sort of plastic 'umbrella' under the bulk of the plumbing designed to deflect most leaks away from the board so the water would go to the case bottom.

Your idea of a seperate enclosure for the pump and rad is also good, although probably to awkward for most people (including me!) However I may approach that method, as the case I'll be using is a dual chamber 'server cube'. I'm already planning to put the pump on the hard drive side of the case, and am debating about where to put the rad - I can either hang it off the back outside of the case on the drive side with the system exhaust fans blowing through it, or stick it on the inside on the mobo side with so the intake fans suck (or blow) through it. Either way, If the rad leaked it wouldn't drip on the mobo, unless the fans blow the water there.

I'm not going to even consider air cooling though it's safer in some respects. I don't like loud computers, and even on my ancient boxes with minimal fans, AC = NOISE... (I've also had multiple fan failures, each box is on at least it's second CPU fan)

Solarian League: ... anyway i think the best way to prevant failure is to just do a regular mantaiance check say onice a week, you just pop open the case check the horse, check the comps to make sure they look like theyre functing correctly etc. or maybe extend that peroid to onice a month do a check up but then again if you bother the system more chance of failure so.... but if you do regularly check upon the system u could possiable avert an severe system failure

A bit more reasonable, (especially since you don't want the horse to escape ) but you do point out the dilema of excessive checking, not sure what the solution is. Of course with a Linux box you don't want to shut down so often so I need to design so as not to need frequent checks.

bigben2k For an OCP (CPU Overheat Protection) alternative, someone suggested bolting a thermal switch to the CPU block: too hot and it shuts down the power supply. The switch is designed specifically for a set temperature, and I think it's available from www.digikey.com .

Simple and neat, but I'm mildly inclined against it. I don't know if those switches would be able to handle the power without getting into relays, etc. (Not bad but then you haven't gained much...) I'd also be concerned about what the trigger temp was, and how that would match what your system needs (dependent on the block properties and location of switch among other things)

If one is already using a digidoc, or other on board monitoring, and has a spare fan header then just hook a kill relay to it and tell the monitor system to turn on that 'fan' if temps get to high... Why add another temp switch?

[Solarian League]

Originally posted by Gooserider
A bit more reasonable, (especially since you don't want the horse to escape ) but you do point out the dilema of excessive checking, not sure what the solution is. Of course with a Linux box you don't want to shut down so often so I need to design so as not to need frequent checks.

hah true i'm going to be running a dual OS box, on raptor 1 i will have window 2k pro, then on raptor 2 i will probably have some sort of linux running not sure what version but obiviously something that can handle my hardware

another benefit of my system setup is the lian-li 75 case i'm buying comes with a preinstalled window so i can just peek into the window say everytime i'm around just peek into the window to see if anything looks bad, if so then open it up other wise stay the **** out.

then say maybe onice a month or so, open that window and closely inspect all the critical spot for rust, corrosion, etc... all of that fun stuff

[BladeRunner]

Gooserider

"Ever hear of a 'Chinese Finger Trap'? Pulling on the hose is like trying to get out of one with a straight pull. Water pressure is going to work more in a direction to make the hose slip".

I get what you are saying but we are not using high pressure supply pumps, I would doubt most systems would achieved 1 bar and the higher rated pumps maybe 2 or 3 max. If you knew just how solid the tubes are on the barbs you'd see why I'm confident they wont slip or blow off, I'm not suggesting you should follow but my set-up is 50% about image as well, to me hose clamps look ugly and are not required in my setup. I can get a max of 150psi (10.5 bar) with my compressor so I will try it sometime. my guess is the tube will not blow off even though it will expand somewhat with 10.5bar.

Do you have a recommended source / brand, or is it just something you happened to find? I'm in the process (in a different window) of getting a nice flow sensor for $10 here http://sales.goldmine-elec.com/Index.asp as I send this message. It's a rotor type, and I'm not sure how easily it can do flow measurement, but it has an adjustable flow failure detection relay and appears to be a nice unit otherwise.

The liquid level sensor I have is a soild state. go here www.rswww.com and type 317-803 it their search engine. Not used it yet so can't comment on it further. I bought two flow switches from the same site, (one was a flow sensor), and the sensor was to restrictive and had an impeller which = noise so no good. The simple float switch I'm using has been 100% reliable so far & I test it twice a week. It is very low restriction and silent after the initial flow created click.

Your blocks certainly are beautiful in the pictures, your finishing work is incredible. Aside from looks, is there a reason for such a high polish on the non-contact surfaces? (I agree with lapping the contact surface, but was thinking more in terms of wire wheeling the rest...) I also noticed that you say you pressure test before finishing, but not after. Seems to me like there is a slight but present risk that a leak could be created by disturbing/removing the solder that was covering it before finishing. Is there reasoning for doing it this way, or did you leave something out of your process descriptions?
As to seal grooves, I wasn't planning on them. I was planning to use flat surfaces with a liberal layer of RTV or aviation Form-a-gasket between them, with 6/32 bolts holding things together. I might also use a thin gasket, but didn't think it was really needed.

This is where I'm coming from, you are both getting carried away with overly complex protection systems and your block sealing idea is in my opinion very flawed. We use RTV and other sealants at work and while they are good to stop most leaking they are bad when it comes to small leaks weeps. I would NEVER seal any block that I was going to use in my system with silicon sealant. In a water-cooling system you want zero leaks IMO and that is what BB2K and myself are trying to get over... prevention is much better than any cure...

I use soldering for the reasons mention before and if the soldering is done correctly is 99.99999999999% never going to leak. I do leak test the block at all stages but if I put every intricate detail in my build articles it would be a tedious read for the less technically minded folks, so I have to strike a balance. I do test them less now than I did, as I'm more confident of getting a 100% soldered seal everytime now than perhaps I was to start with, experience counts I guess

I never originally intended getting mad with the finish of the blocks when I first started but I like them looking nice. If cooper is left it will tarnish and look horrid, some people may not care about this but I do.

[Gooserider]

Solarian League: hah true i'm going to be running a dual OS box, on raptor 1 i will have window 2k pro, then on raptor 2 i will probably have some sort of linux running not sure what version but obiviously something that can handle my hardware

A couple of notes from my Linux experience - Unless you are planning to do separate drives that you change to change O/S's (not reccomended, as it's a PITA, and you always find you need data on the drive you aren't running) be careful doing the installs - MS products don't play nicely with other O/S's (Just like MS itself...) it is generally best to install Windows 1st, then put Linux on second as Linux is more willing and able to re-arrange things so that both O/S's can co-exist smoothly.
As to hardware support, most REAL hardware devices are pretty well supported, the problem is with 'Win-modems' and other such O/S dependent hardware, which you don't want anyways for the way it kills your system performance under any O/S. The other problem is with a few manufacturers that won't release enough information on their interfaces to allow the Open Source folks to do good drivers for their products. However for at least the major distributions there isn't a lot of difference in the hardware supported.
I currently use Red Hat and find it's OK as long as you don't try to do much upgrading. I am inclined on the next box to try Debian, or possibly one of the new 'Source based' distributions (Which are only reccomended if you have broadband speed connectivity)

another benefit of my system setup is the lian-li 75 case i'm buying comes with a preinstalled window so i can just peek into the window say everytime i'm around just peek into the window to see if anything looks bad, if so then open it up other wise stay the **** out. then say maybe onice a month or so, open that window and closely inspect all the critical spot for rust, corrosion, etc... all of that fun stuff

Reasonable, but note that unless you have a leak, you will NOT see any rust or corrosion. To get either requires exposure to water or high humidity. (BTW, it is only steel that 'rusts' non-ferrous materials like copper either oxidize or corrode...)

BladeRunner: Gooserider quote: "Ever hear of a 'Chinese Finger Trap'? ..." I get what you are saying but we are not using high pressure supply pumps, I would doubt most systems would achieved 1 bar and the higher rated pumps maybe 2 or 3 max. ... I'm confident they wont slip or blow off, I'm not suggesting you should follow but my set-up is 50% about image as well, to me hose clamps look ugly and are not required in my setup. I can get a max of 150psi (10.5 bar) with my compressor so I will try it sometime.

You might be surprised - I've mentioned being a motorcyclist, and one of the things I've found is that a properly fitted hand grip just about can't be moved by twisting or pulling on it. But stick an airhose between the grip and the bar, and it will come off with no trouble at all (volume counts more than pressure BTW).
I agree that hose clamps are ugly, which is why I was suggesting use of zip-ties. Pulled tight and cut, a good zip-tie is almost as solid as a hose clamp, and close to invisible. I use them on our swimming pool robot's hoses, the stock fittings kept blowing apart, and you don't want to use a standard hose clamp because of problems with the pool chemistry and the sharp edges. But I haven't had a failure since I put zip ties on the problem fittings.

The liquid level sensor I have is a soild state. go here ...

Thanks for the link, I just checked it out though, and I don't think it does what I want. I am thinking in terms of a 'gas guage' equivalent, this switch looks like it just turns on and off when it reaches a certain depth. I could be wrong though, the site didn't want to give me a data sheet w/o registering, and the US site they sent me to didn't seem to have the part, so all I had to go on was the description on the page.

This is where I'm coming from, you are both getting carried away with overly complex protection systems and your block sealing idea is in my opinion very flawed. We use RTV and other sealants at work and while they are good to stop most leaking they are bad when it comes to small leaks weeps. I would NEVER seal any block that I was going to use in my system with silicon sealant. In a water- cooling system you want zero leaks IMO and that is what BB2K and myself are trying to get over... prevention is much better than any cure... I use soldering for the reasons mention before and if the soldering is done correctly is 99.99999999999% never going to leak.

I agree that ANY leakage is unacceptable, so I'm glad I picked up some plumbing solder and flux the other day (I noticed you use electronics solder, but that seems problematic to me because of the lead.) I know I don't have trouble on my bike most of the time, but I'm a bit less fussy on it as well, and am working with larger parts that HAVE to be dis-assembleable.

BTW, do you solder in your non-Festo barbs (or reccomend it) or do you seal them some other way? (Teflon tape and plumbers dope is what some seem to use)

Yes, I am perhaps over concerned about protection systems, but I *HOPE* they would never be needed. If they are, it means I made a mistake in my design. I'm just trying to allow for being human and also fend of that B*****D lawyer Murphy...

I do leak test the block at all stages but if I put every intricate detail in my build articles it would be a tedious read for the less technically minded folks, so I have to strike a balance. I do test them less now than I did, as I'm more confident of getting a 100% soldered seal everytime now than perhaps I was to start with, experience counts I guess

Thanks for the clarification, it was just something that bothered me given the care you took everywhere else. I understand about wanting to keep things breif, although at the same time you seem to put in the same process in each article - perhaps a more detailed single article 'How I finish my blocks' with a pointer to it in the others?

I never originally intended getting mad with the finish of the blocks when I first started but I like them looking nice. If cooper is left it will tarnish and look horrid, some people may not care about this but I do.

I agree that it is good to make them look nice, I just didn't know if there was some other reason as well. I'm a bit less fussy than you are about appearances, so I think I'll stick with my original notion of wire-wheeling the non contact surfaces and then clear coating with a spray bomb. Sort of an 'anodised' look instead of the polished look.

Gooserider

[Solarian League]

gooserider: ive worked with linux a bit but nothing execessive, but anyway. i know how much mircosoft loves other foreign OS heh

anyway what i was going to do was just format all harddrive, install window 2k pro on raptor 1. then, for raptor 2 i would just leave it as fat32 or something like that for linux because so far from what ive heard is that linux can't write to NTFS yet or something, ive seen few linux that can read from it but they all says its not recommened to write to it heh.

anyway the raptor is only 36 gig so shouldn't have too much trouble with fat32, i can't rember the largest harddrive you can have with fat32, i used to have an old 20 gig with fat32 and it worked fine i think.


i ditto what version or type of linux i'm going to get, probably slackware or deban i'm guessing.

and yeah i sort of got broadband, its not terific fast but it can download a whole cdrom in a day, hellva lots better than old 56k modem u know

i could probably optionaly hold off installing linux untill i hit college in about 2 month and college probably will have a damn good internet acess so i could then there download my linux goodies

[bigben2k]

FWIW...


In the initial setup, many people report doing a test run of the cooling components, which are put aside for the purpose.

I suggested pressure testing, which really ought to accelerate the process: just make yourself one of these (see pic) and dunk your cooling components in a container of water: any leak should be immediately observable.

[MMZ_TimeLord]

Keep in mind if you use hose clamps on barbs like this...

|\|\|\|\
        |
|/|/|/|/

You are looking for problems down the road. As hose clamps force the tubing to cut into the barb's sharp spots and can harm the tubing.

Use barbs like this...

  _ 
 ( )
 | |
 | |
|___|

if you plan on using clamps.

As to prevention... in the three years my system has been running, I have had ONE true failure. This was due to some really cheap (free) tubing that I used when I first built the system. This tubing is the type that is used for turnaques in doctors offices. Kind of a tan color, really rubbery, and VERY flexible (kinks easy).

It took over a year for this tubing to fail. It was a short piece I had in the water reservior that connects my pump (submerged type) to the barb in the wall of the reservior. It finally degraded and blew out. the pump was still running, and the temps began to rise... I was using MBM5, as I do now and have alarms set on it. The system went into auto shutdown shortly after the temps reached the alarm level.

After correcting the problem, the system was fine. No damage.

IMHO the worst problems you have to fear are micro leaks in your block (hence pressure testing BEFORE the final build) and hose leaks at the barbs/connectors.

Pump and fan failures and hose blockages begin to raise temps pretty fast. And if the box is in a controlled environment (air conditioned home), you can set your alarm level 5° or 10° C above your hotest running temp (using CPUBurn to max the processor) and have the system auto shutdown if it exceeds that level.

Other than that... monthly inspections of the system to verify no small leaks have started. The UV dye is especially good for this as mentioned earlier.

[MMZ_TimeLord]

Okay... my ASCII art sux... here is another look at the barbs and what I think is bad and good with hose clamps...

[bigben2k]

I stuck to the polypropylene barbs. I had to shave some of the molding leftovers, but I don't foresee any problems, even if you call them "bad"

[MMZ_TimeLord]

Okay... I guess I should be more specific... BRASS barbs of that type. The plasic ones should be okay as they are a softer material and will compress some as opposed to cutting into your tubing.

Mine are all brass of the "bad" type with NO clamps on the blocks and simple Acrylic straight tubing on the reservior and the elbows I made. These are larger than the inner diameter of the tubing and so the tubing has to stretch over it and is VERY tight, also with no clamps.

[Gooserider]

Solarian League: ive worked with linux a bit but nothing execessive, but anyway. i know how much mircosoft loves other foreign OS heh anyway what i was going to do was just format all harddrive, install window 2k pro on raptor 1. then, for raptor 2 i would just leave it as fat32 or something like that for linux because so far from what ive heard is that linux can't write to NTFS yet or something, ive seen few linux that can read from it but they all says its not recommened to write to it heh.

Linux should go on it's own partitions. There are a few distributions that will run off MS type partitions but they are very limited, mostly intended to let people experiment a bit w/o needing to mess with the disk partitions. As I recollect Linux can read and write all types of MS partitions, but I could be wrong on that. What I would reccomend is to look online for one of the appropriate 'HowTos' on how best to set up the multi-boot combo you want, and follow it. Most seem to involve installing MS on one or two small partitions, then creating Linux partitions on the rest of the drive and putting Linux on those.
(BTW, don't feel bad about only having 36Gb, I just doubled my drive size and am now up to EIGHT Gb!)

bigben2k FWIW... I suggested pressure testing, which really ought to accelerate the process: just make yourself one of these (see pic) and dunk your cooling components in a container of water: any leak should be immediately observable.

Good idea, I was planning to do something like that. Not sure what pressure to use though... I'm figuring 50-75 psi for blocks, etc. but I'm not sure what the rad should be able to handle. (I'll be using an auto heater core)

MMZ_TimeLord Keep in mind if you use hose clamps on barbs like this... You are looking for problems down the road. As hose clamps force the tubing to cut into the barb's sharp spots and can harm the tubing. Use barbs like this... if you plan on using clamps.

Hmmm... That's a new one on me, I've never heard of that type of problem unless the clamp was WAY overtightened, or the hose was taken on and off alot. I will keep it im mind however. Unfortuneately all my local hardware stores seem to carry are the 'bad' kind, however I have a lathe, so I can always reshape them a bit, possibly just making a flat spot for the clamp to sit on. (I may end up using zip-ties anyways in order to save space) I'm also going to be using good hose, the braid reinforced clear poly stuff, among other things it has very thick walls.

As to prevention... IMHO the worst problems you have to fear are micro leaks in your block (hence pressure testing BEFORE the final build) and hose leaks at the barbs/connectors. Pump and fan failures and hose blockages begin to raise temps pretty fast. And if the box is in a controlled environment (air conditioned home), you can set your alarm level 5° or 10° C above your hotest running temp (using CPUBurn to max the processor) and have the system auto shutdown if it exceeds that level. Other than that... monthly inspections of the system to verify no small leaks have started. The UV dye is especially good for this as mentioned earlier.

Thanks for the experience input, it helps.

MMZ_TimeLord: Okay... my ASCII art sux...

It's not that bad, better than mine in some cases. I thought it was rather, ummm, phallic...

So far, most of the barbs I've gotten have been brass of the 'bad' variety. (but I can change that, lathes are wonderful) I also got a couple of grey plastic barbs with the 2nd hand Iwaki pump I just picked up. I don't know why, but I'm more inclined to trust the brass items than the plastic ones. The plastic just feels flimsy to me. I'm planning to use hose that matches the barbs, but even then it's a tight fit and I feel sure that if I put a clamp or possibly a tie wrap on it, it will be OK.

Gooserider

[bigben2k]

New heatercores are tested to 25 psi, and I wouldn't exceed that.

[Solarian League]

Originally posted by Gooserider
Linux should go on it's own partitions. There are a few distributions that will run off MS type partitions but they are very limited, mostly intended to let people experiment a bit w/o needing to mess with the disk partitions. As I recollect Linux can read and write all types of MS partitions, but I could be wrong on that. What I would reccomend is to look online for one of the appropriate 'HowTos' on how best to set up the multi-boot combo you want, and follow it. Most seem to involve installing MS on one or two small partitions, then creating Linux partitions on the rest of the drive and putting Linux on those.
(BTW, don't feel bad about only having 36Gb, I just doubled my drive size and am now up to EIGHT Gb!)

gotcha, and btw i have more than 36 gig, i have two raptor which both each are 36 gig, and two storage IDE drive which both are each at 200 gig.

anyway thanks for ur help on the linux thing, i'll take a look at some of the to do guide, anyway i want to setup so linux has one whole harddrive to itself, aka the second raptor, hopfully it should be able to access the IDE storage drive with no problems.

[BladeRunner]

Gooserider

You might be surprised - I've mentioned being a motorcyclist, and one of the things I've found is that a properly fitted hand grip just about can't be moved by twisting or pulling on it. But stick an airhose between the grip and the bar, and it will come off with no trouble at all (volume counts more than pressure BTW).
I agree that hose clamps are ugly, which is why I was suggesting use of zip-ties. Pulled tight and cut, a good zip-tie is almost as solid as a hose clamp, and close to invisible. I use them on our swimming pool robot's hoses, the stock fittings kept blowing apart, and you don't want to use a standard hose clamp because of problems with the pool chemistry and the sharp edges. But I haven't had a failure since I put zip ties on the problem fittings.

As said I get your point, I work in motorsport and we have oil pipes that are pushed over barbs with no clamps fitted and they don't blow off, as said It is fine and hard to get over in text without you being here and me showing you... trust me they wont come off with the amount of flow pressure an aquarium pump can muster, but do use clamps if it makes you happier

agree that ANY leakage is unacceptable, so I'm glad I picked up some plumbing solder and flux the other day (I noticed you use electronics solder, but that seems problematic to me because of the lead.) I know I don't have trouble on my bike most of the time, but I'm a bit less fussy on it as well, and am working with larger parts that HAVE to be dis-assembleable.

BTW, do you solder in your non-Festo barbs (or reccomend it) or do you seal them some other way? (Teflon tape and plumbers dope is what some seem to use)

Yes, I am perhaps over concerned about protection systems, but I *HOPE* they would never be needed. If they are, it means I made a mistake in my design. I'm just trying to allow for being human and also fend of that B*****D lawyer Murphy...

I use electrical solder because it is much easier to work with, flows better at low temps, and contains a flux core, and I doubt the tiny amount of lead in the solder in the water-cooling will cause any problems. I've soldered in fittings, but mostly I tap a hole in the block and seal the barb in with either araldite, (epoxy resin), or an expensive motorsport thread locking agent I "acquired" from work

Again this is partly for the image factor as visible white tape looks ugly, but also the sealing agents are much less likely to leak or weep than PTFE tape. I think it is good to be concerned about protection systems, just don't loose sight of the real importance....... designing and implementing the system so that they are not actually required or used, even though they are there. For instance I made that proximity switch system to ensure the PC cannot be started without the pump plugged into the PC psu, even if another cable is plugged in it wont start up. Overkill? maybe but it really is so easy to forget something this basic when you have been doing things to the system and it could turn into a disaster, simply because the pump wasn't plugged in. If the real truth were known there is probably more water cooling mishaps related to this that actual real pump death.

Thanks for the clarification, it was just something that bothered me given the care you took everywhere else. I understand about wanting to keep things brief, although at the same time you seem to put in the same process in each article - perhaps a more detailed single article 'How I finish my blocks' with a pointer to it in the others?

Yeah I'm going over my site updating stuff, but time is always an issue and to be honest I hate webby stuff, I find it tedious and would rather be making blocks. Probably be a good idea to have a link to a detailed block soldering testing page, (and maybe another link on that page to an abridged version?)

Not sure braided hose is an ideal as it tends to be stiff so may stress block fittings or their mountings. Can't really beat Tygon for standard barbs, it is thick walled but flexible and kink resistant, all in all good stuff if not cheap.

[Gooserider]

bigben2k: New heatercores are tested to 25 psi, and I wouldn't exceed that.

Thanks for the info, it's about what I was guessing. After all the rads are probably the flimsiest items in a typical WC system. This isn't a fault, just an observation noting that a heavily built rad would probably also really stink as a heat exchanger.

Solarian League: ... anyway thanks for ur help on the linux thing, i'll take a look at some of the to do guide, anyway i want to setup so linux has one whole harddrive to itself, aka the second raptor, hopfully it should be able to access the IDE storage drive with no problems.

Double check the HowTo. I believe that if both drives are present in the system at the same time, there is a boot loader that Linux needs to put on the primary boot drive if you want to avoid using a boot floppy. This is just a small boot screen that pops up and asks which OS to load (you can have several) and then redirects the boot process to the appropriate boot blocks for the desired OS. This is the way my system works as I keep a small DOS 6.x partition with some of my legacy SW on it. Both OS's should be able to work with the IDE storage drive as long as you format it in something both can read and write to. I know FAT isn't a problem, I think NTFS is OK, but I'm not certain offhand.

BladeRunner: ... but do use clamps if it makes you happier

I may go with zip ties, but yes, some kind of clamps will make me feel better. Like I said, I'm paranoid...

I use electrical solder because it is much easier to work with, flows better at low temps, and contains a flux core, and I doubt the tiny amount of lead in the solder in the water-cooling will cause any problems.

I guess I'll experiment a bit... I'll be using either a propane torch or an oxy-acetelyne rig, so I may be able to do more heat than you get with your torch. (I've even wondered about brass brazing, but am inclined against it overall.)

I've soldered in fittings, but mostly I tap a hole in the block and seal the barb in with either araldite, (epoxy resin), or an expensive motorsport thread locking agent I "acquired" from work Again this is partly for the image factor as visible white tape looks ugly, but also the sealing agents are much less likely to leak or weep than PTFE tape.

I was thinking of soldering in my barbs because I'm not sure about how well the threads would hold in my top (only 1/4" thick). Additionally, I plan to trim some or all of the thread that sticks through the inside of the top so it doesn't obstruct the coolant flow. I don't want to take the barb out after I do that because of possible problems getting it back in, and if I cut it on an angle, aligning it.
I noticed your mentions that you had to deal with the top drifting out of position when soldering the block up, I was thinking that if I put the barbs in place and trimmed them ahead of time, then I could use them as locating pins when doing the solder up, and just flow some solder into the threads while things were still hot.
I am surprised that you consider teflon tape leak prone however - We have teflon taped joints in the plumbing all over the house, and no leaks despite the 110psi mains pressure we were getting. (I agree it can be ugly though)

I think it is good to be concerned about protection systems, just don't loose sight of the real importance....... designing and implementing the system so that they are not actually required or used, even though they are there.

Agreed completely, I don't want to ever need to have installed them, I'd just feel foolish if I hadn't and then had a disaster they could have saved me from.

For instance I made that proximity switch system to ensure the PC cannot be started without the pump plugged into the PC psu, even if another cable is plugged in it wont start up. Overkill? maybe but it really is so easy to forget something this basic when you have been doing things to the system and it could turn into a disaster, simply because the pump wasn't plugged in. If the real truth were known there is probably more water cooling mishaps related to this that actual real pump death.

It is an elegant solution, though I'm planning something a bit different. I plan to make a single power cord going to my 'kill relay' with the pump hard wired to the output of the relay. Unless I have a reason that I must, I don't want to open or mod the PSU if I don't have to, as I prefer to keep the warranty intact. I'm using a PCP&C unit that's expensive enough I want to keep it under warranty.

Yeah I'm going over my site updating stuff, but time is always an issue and to be honest I hate webby stuff, I find it tedious and would rather be making blocks. Probably be a good idea to have a link to a detailed block soldering testing page, (and maybe another link on that page to an abridged version?)

Understandable, and what you have works, I just thought it might save repetition in any future articles you do.

Not sure braided hose is an ideal as it tends to be stiff so may stress block fittings or their mountings. Can't really beat Tygon for standard barbs, it is thick walled but flexible and kink resistant, all in all good stuff if not cheap

The stuff at the local store seems pretty flexible, though I haven't compared it to Tygon. I may see if I can find some Tygon locally to see if I think it's enough better to justify switching to it. I might also do a mixed setup with Tygon for the block plumbing and braided elsewhere - I know I'll be using at least 3-4 different sizes of tubing in my setup so it wouldn't be difficult. I do need to worry some about my block mountings since I have to do a clip-on mount (my mobo won't have the holes for a bolt-on mount, one of it's few flaws IMHO) I'm going to make a 3 lug clip out of 304 stainless, but still...

Gooserider

[BladeRunner]

It is an elegant solution, though I'm planning something a bit different. I plan to make a single power cord going to my 'kill relay' with the pump hard wired to the output of the relay. Unless I have a reason that I must, I don't want to open or mod the PSU if I don't have to, as I prefer to keep the warranty intact. I'm using a PCP&C unit that's expensive enough I want to keep it under warranty.

It's hard sometimes to get over a point in text sometimes, I wasn't suggesting you copy my system of preventing the pump not being plugged in, more just highlighting the types of things that are more likely to cause disasters, (assuming the design and application of the system is solid), before one gets involved in many overly complex protection systems. I believe user error causes most of these things. I once refilled my system and stopped when I heard trickling water. I'd forgotten to plug in one of the PSU fetsos at some point when I'd had the set-up apart. It was in the fitting but not tight. The result was water all over the mobo, vga card & sound card. As I had unplugged the PSU and de-powered the system before working on it, this was only a minor annoyance that ment I had to spend half an hour with a hairdryer and an blow gun before I could get back to where I was previously.

I guess I'll experiment a bit... I'll be using either a propane torch or an oxy-acetylene rig, so I may be able to do more heat than you get with your torch. (I've even wondered about brass brazing, but am inclined against it overall.)

Using an oxy-acetylene rig can prove hard as the copper will tarnish with the heat quicker making it more difficult to get the solder to flow and take. Also heat warpage of the copper can be an issue if you get it too hot.

I was thinking of soldering in my barbs because I'm not sure about how well the threads would hold in my top (only 1/4" thick). Additionally, I plan to trim some or all of the thread that sticks through the inside of the top so it doesn't obstruct the coolant flow. I don't want to take the barb out after I do that because of possible problems getting it back in, and if I cut it on an angle, aligning it.
I noticed your mentions that you had to deal with the top drifting out of position when soldering the block up, I was thinking that if I put the barbs in place and trimmed them ahead of time, then I could use them as locating pins when doing the solder up, and just flow some solder into the threads while things were still hot.
I am surprised that you consider teflon tape leak prone however - We have teflon taped joints in the plumbing all over the house, and no leaks despite the 110psi mains pressure we were getting. (I agree it can be ugly though)

It's unlikely if done well but I've had weeps and minor leaks from household plumbing fittings using PTFE tape in a no pressurised UK style system. As said in PC water-cooling zero leaks/weeps are the order of the day.

The stuff at the local store seems pretty flexible, though I haven't compared it to Tygon. I may see if I can find some Tygon locally to see if I think it's enough better to justify switching to it. I might also do a mixed set-up with Tygon for the block plumbing and braided elsewhere - I know I'll be using at least 3-4 different sizes of tubing in my set-up so it wouldn't be difficult. I do need to worry some about my block mountings since I have to do a clip-on mount (my mobo won't have the holes for a bolt-on mount, one of it's few flaws IMHO) I'm going to make a 3 lug clip out of 304 stainless, but still...

I'd send you a small piece of tygon to try if you were not on the other side of the pond, however a foot of it wont cost much if you can find a local supplier. I used a socket lug waterblock when I first started over two years ago, and it was a bit of a concern but the choice of blocks was very limited back then. Now I would never put any mobo on my shopping list if it didn't come with socket holes, for waterblock retention.


Just for interest I did this basic schematic in ms paint the other day that clearly shows how my system flow is configured.

[Boli]

Back to the barbs... I have never had any trouble with the brass barbs AT ALL.

I use no clamps and even this way I find that a good twisting is required to level the hose of the barb. No leaks of anykind at the connections.

The only problem I have with leaks is the pump where I tightened the barb too much a cracked the housing slightly ... Do'h! but the barb itself was not at fault

I use the 1/2" brass barbs with generic dagerden 1/2" tubing and the fix is as tight as you need it. (I am not running at extreme presures).

~ Boli

[Gooserider]

BladeRunner: It's hard sometimes to get over a point in text sometimes, ... more just highlighting the types of things that are more likely to cause disasters, (assuming the design and application of the system is solid), before one gets involved in many overly complex protection systems. I believe user error causes most of these things. I once ... heard trickling water. I'd forgotten to plug in one of the PSU fetsos ... half an hour with a hairdryer and a blow gun before I could get back to where I was previously.

I got the point, I was just commenting that it was a very elegant solution... I'm well aware of the problem of user error, I've done more than a few of them myself. One of my favorites is to close the lid on a tool box, but not latch it. Later I go to move it, and dump the box all over the floor. I've now tried to make a rule that I don't close the lid UNLESS I latch it. I try to make analagous policy about assembling things, and avoid putting parts together unless I lock them down as well.

Using an oxy-acetylene rig can prove hard as the copper will tarnish with the heat quicker making it more difficult to get the solder to flow and take. Also heat warpage of the copper can be an issue if you get it too hot.

I mention it as a possibility, since I have both. If I can do it with the propane I will, if only because it's less expensive and less hassle to get the tanks refilled... What I'm thinking of doing is what I see plumbers do. They clean the surfaces with emery cloth and/or a wire brush, then apply a coat of plumbing flux and fit everything together. They then heat the joint with a torch, and the solder wicks into the joint and seals everything up. (Note that in the US it is a serious legal violation to use lead bearing solder in a potable water system, I'm not planning to drink my coolant, but the claim is that the lead free solder they use now is actually better than the lead stuff.

It's unlikely if done well but I've had weeps and minor leaks from household plumbing fittings using PTFE tape in a no pressurised UK style system. As said in PC water-cooling zero leaks/weeps are the order of the day.

Not arguing, just suprised is all, I just had PTFE on my list of more reliable things. (Though I note that our plumber uses PTFE with plumbers dope on top of it)
Of course the threads in a water system will likely be thicker / longer than what I will have in most of my system, making sealing a bit more critical.

I'd send you a small piece of tygon to try if you were not on the other side of the pond, however a foot of it wont cost much if you can find a local supplier.

I appreciate the 'non-offer', thanks. I'm sure I can find someone around here that carries it, the challenge will be to find someone that is willing to sell it in 'retail' quantities. BTW, it's just 'TYGON' correct, or is there some additional descriptive like a formula number or some such?
[

I used a socket lug waterblock when I first started over two years ago, and it was a bit of a concern but the choice of blocks was very limited back then. Now I would never put any mobo on my shopping list if it didn't come with socket holes, for waterblock retention.

I sort of agree, but not to the same degree - I wouldn't put not having the holes in the 'no-sale' category, but more in the 'choosing among equals' class. However, I want rock solid, reliable performance, and the Tyan mobo I'm using was consistently named as tops in its class (dual CPU, workstation/server class) for both performance and reliability by sources I trust (not to mention AMD themselves) There really wasn't another board in the same class as this one IMHO, so it really isn't possible to pick a hole-equipped alternative.

Just for interest I did this basic schematic in ms paint the other day that clearly shows how my system flow is configured.

Nice... What was the reasoning behind the routing you chose, was it trying to balance flow restrictions, temperature, or just ease of plumbing?

Boli: Back to the barbs... I have never had any trouble with the brass barbs AT ALL. I use no clamps and even this way I find that a good twisting is required to level the hose of the barb. No leaks of anykind at the connections. The only problem I have with leaks is the pump where I tightened the barb too much a cracked the housing slightly ... Do'h! but the barb itself was not at fault I use the 1/2" brass barbs with generic dagerden 1/2" tubing and the fix is as tight as you need it. (I am not running at extreme presures). ~ Boli

My pressure won't be extreme, but it will be a bit higher than average perhaps - I have an Iwaki MD20RT pump, which has a potential head of 14+ feet, or about 7psi. I believe the pumps most used for WCing have head numbers closer to 8-10 feet, or 4-5 psi. I dont think this will be a problem, just something I need to be aware of. My current plan (Unless someone talks me into something better ) is to use mostly brass barbs which I will stick in a lathe and flatten one or two rings on. (so one edge will look like this: |\|\---|\|\ ) I will then put the hose on, and apply a tight zip tie around the flat area to prevent any possibility of blow offs. On my drive blocks, which will have terminations consisting of the copper tubes they are made from, I will form a 'good' type barb end, and put a zip tie behind, so the hose can't come off accidentally.

Gooserider

[BladeRunner]

Gooserider

I see what you mean about the mobo choice as it's more limited with dual CPU systems, It would still annoy me however, and it's beyond comprehension that some makers see fit to make newer single CPU boards without the holes, (unless of course they are indicating that they don't want the modders to buy their products)

Not sure what you are asking about the flow configuration, but I split the flow in distrubition as it's the most effective way to avoid overall system flow restrictions with that many blocks. The CPU only adds at most 0.5C to the coolant temp in one pass, and all the other system blocks combined about the same added together. If they were run in series it would be highly restrictive to flow for the most important block, (CPU). Running it in series the blocks would have to be much larger and non restrictive.

If you mean the layout then to be 100% honest it's not configured like that yet, as the HDD block is still 6mm festo, which requires an extra splitter, but It's how I intend to run it when the HDD block is replaced. The reasoning is that some or the blocks will need doubling up as there are 6 extra blocks after the CPU Splitter-X and only four outputs. I consider the VGA to be the next most important, so each block gets it's own outlet from the Splitter-X where as the less important HDD/PSU & NB/Mosfet will be fine sharing an output respectively. With good flow and cool coolant I doubt you could even measured any real world difference even if it actually mattered, which I'm pretty sure it doesn't, as long as all ouputs have reasonable flow through them.

The inflow/ouflow tube I used is Tygon R-3603 which is 1/2" ID, that as I said is heat stretched, (boiling water immersion), over the 5/8" OD (1/2" ID barb), keeping the ID "pure" throughout. It really is best as 1/2" Tygon on a 1/2" barb is too loose IMO, and could present a leak / weep issue even with a clip. If I were using 1/2" OD barbs I would get 3/8" Tygon and do the same I guess. Tygon is thick walled and a brass barb wont cut through it even if the edge is sharp, this edge is what gives it good grip. If you intend to use socket clip retention I highly recommend Tygon tube, (or silicon, but you'd definately need good clips with silicon)

[Gooserider]

BladeRunner: I see what you mean about the mobo choice as it's more limited with dual CPU systems, It would still annoy me however, and it's beyond comprehension that some makers see fit to make newer single CPU boards without the holes, (unless of course they are indicating that they don't want the modders to buy their products)

I agree it's annoying, but at least according to Tyan tech support, it appears that either AMD has removed the holes from the current design spec, or hadn't put them in as of when the board was first built. (And bear in mind that Tyan is the board co. selected to do the REFERENCE design for dual Athlon MP mobos, so they should know...)

Below is a partial quote from the response I got from Tyan when I asked about the holes...

Tyan Tech Support: ... Only the S2462 has these holes designed around the sockets. All the other AMD products do not have these type of mounting holes available. At the point that these boards were designed AMD had no official design that incorporated mounting through the board itself. There are plenty of sources, including AMD's own, that will mount and work just fine without being mounted through the board itself.

If this represents a new change in AMD design spec policy, it may bode ill for the modding community. OTOH, I'm not sure how we could fight it, somehow I have trouble visualizing the effect of a 'Save the Holes' campaign...

Not sure what you are asking about the flow configuration, ...

I was just wondering why you chose the distribution pattern you did, and gave some of the reasons I could think of. Your explanation was exactly what I was asking, and makes good sense.

Since I'll be doing at least some flow splitting as well, I was interested in your analysis. I'm going to have manifolds with loops running between them, as my CPU blocks will only have single in/out barbs.
Elsewhere I've been discussing flow patterns, and it appears to me that the best pattern will be to have two loops, one with the two CPU blocks in series using 1/2" barbs/tubing, and the second connecting all my small blocks (hard drives, possibly northbridge, etc.) with 1/4" or 3/8" tubing. That should keep the bulk of the water flow going through the CPU blocks, without excessive restriction.

The only real question was whether to have the CPU's in series or parrallel, but barring testing, I expect the series setup will work better from what the other thread was saying.

... The inflow/ouflow tube I used is Tygon R-3603 which is 1/2" ID, that as I said is heat stretched, (boiling water immersion), over the 5/8" OD (1/2" ID barb), keeping the ID "pure" throughout. It really is best as 1/2" Tygon on a 1/2" barb is too loose IMO, and could present a leak / weep issue even with a clip. If I were using 1/2" OD barbs I would get 3/8" Tygon and do the same I guess. Tygon is thick walled and a brass barb wont cut through it even if the edge is sharp, this edge is what gives it good grip. If you intend to use socket clip retention I highly recommend Tygon tube, (or silicon, but you'd definately need good clips with silicon)

Well, I'll check around to see if I can find a local Tygon source. Thanks for the spec. on the tube number, I'll have to mike out my barbs to see what their actual dimensions are, I'm pretty sure they're 1/2" OD, 3/8" ID though.

Gooserider

[BladeRunner]

Gooserider

Good luck with your system, you should be well armed now to make it very good and potentially disaster free for a long time.

Originally posted by Gooserider
Below is a partial quote from the response I got from Tyan when I asked about the holes...

If this represents a new change in AMD design spec policy, it may bode ill for the modding community. OTOH, I'm not sure how we could fight it, somehow I have trouble visualizing the effect of a 'Save the Holes' campaign...
Gooserider

I was more talking of single CPU mobos in my last reply, as some makes like Abit and Asus (that often target the overclocking market with new mobos), have curiously released new "Overclocking" mobos without the holes in the past, (at least at first), I think the NF7 came without holes to begin with.

I'd fight it simply by putting my money elsewhere if AMD were daft enough to remove the holes from the spec, and force board makers to do the same. The current Retail Intel P4 sink uses the P4 mobo holes for retention, so I doubt Intel will be removing their holes for a while.......... I also deliberately made the Splitter-X with a base area just large enough to fit a P4 CPU just in case I Change CPU type