Actually, spring clamps work better than gear-driven.chylld wrote:my current plan is to use worm-gear clamps with regular hose barbs as that is the most secure method.
Introducing my new watercooling design concept!
Moderators: NeilBlanchard, Ralf Hutter, sthayashi, Lawrence Lee
-
HammerSandwich
- *Lifetime Patron*
- Posts: 1288
- Joined: Sat Oct 25, 2003 3:21 pm
- Location: 15143, USA
- Contact:
You may have already considered this, but you could equip the unit with the usual "kettle lead" socket for taking mains from the wall, and a UPS style male "kettle lead" socket to pass through power to the PC, along with a female to male "kettle lead" cable to take power to the PC. The user can then use whatever power lead their PC/PSU came with to plug the watercooling unit into whatever wall socket their country uses, saving you some localisation issues.
I'm guessing that there will be a switch at the front of the system to supply power to the pump/fans and to the PC at the same time.
Will it cope with the full range of voltages (110 - 250V)?
Will you ship the GPU - ready version with RAMsinks?
Wouldn't wood be too resonant? Or have you decided not to use wood after all?
Consider these questions rhetorical if you like, I realise you haven't finalised the design yet. Good luck with it, I hope to see it available in the UK soon!
Spod
I'm guessing that there will be a switch at the front of the system to supply power to the pump/fans and to the PC at the same time.
Will it cope with the full range of voltages (110 - 250V)?
Will you ship the GPU - ready version with RAMsinks?
Wouldn't wood be too resonant? Or have you decided not to use wood after all?
Consider these questions rhetorical if you like, I realise you haven't finalised the design yet. Good luck with it, I hope to see it available in the UK soon!
Spod
chylld wrote:ahh
what are spring clamps?
An example:
TDX w/ socket hold down.
http://www.dangerden.com/mall2/more_blo ... fmmore=209
DrCR
______________
Spod - that's exactly what i plan on doing with the power connectors/cables.
the gpu block may come with heatsinks yes but as with every other added cost i will have to look into it thoroughly.
wood is easier to work with and looks much better than plastic or steel imo. vibration from the fans will not be an issue but the pumps might be a different story.
DrCR: ahh, those. yeh, we'll see
the gpu block may come with heatsinks yes but as with every other added cost i will have to look into it thoroughly.
wood is easier to work with and looks much better than plastic or steel imo. vibration from the fans will not be an issue but the pumps might be a different story.
DrCR: ahh, those. yeh, we'll see
-
HammerSandwich
- *Lifetime Patron*
- Posts: 1288
- Joined: Sat Oct 25, 2003 3:21 pm
- Location: 15143, USA
- Contact:
No, we're talking about hose clamps. Go to McMaster, search on "spring clamp", then click the "spring hose clamps" link.DrCR wrote:An example:chylld wrote: what are spring clamps?
TDX w/ socket hold down.
http://www.dangerden.com/mall2/more_blo ... fmmore=209
HammerSandwich wrote: No, we're talking about hose clamps. Go to McMaster, search on "spring clamp", then click the "spring hose clamps" link.
Ah, OK. My bad.
In reference to this subject though, I prefer a bolt or worm design like those on page 210 or 213 so I can determine my own tension level. Your wise in choosing a constant tension type though as it is more newbie friendly--no second guessing as to how tight to make the clamp.
And yet you say spring are better than worm/bolt driven. Why? Because the pressure is better equalized? Perhaps I should consider using the "Wide-Band Self-Tightening Spring Hose and Tube Clamps" on page 215 in the future (pic). I use clamps similar to these for the fuel lines in my RC aircraft (pic).
DrCR
_____________
They are. But its PITA to have flat head screw driver handy just to rotate it endlessely (it seems), in very awkard angle. Stay away from it for user friendliness is my opinion.chylld wrote:HammerSandwich: ahh, those thingies. yeh they're pretty economical but i'm not sure if they're tight enough. has anyone tried these in a watercooling rig?
BTW this is what I was talking about passing through PCI for 3/8" ID system.
Basically splitting the 3/8ID 1/2OD to 2 1/4ID 3/8OD tubing somehow with out adding much resistance.
Or using something thats elleptical for connection of size of maybe 1/4"x1/2" ish thing.
What about something like these? They should be really easy to work with.
Quick-Release Worm-Drive Hose and Tube Clamps
http://www.mcmaster.com , page 211
DrCR
____________
Quick-Release Worm-Drive Hose and Tube Clamps
http://www.mcmaster.com , page 211
DrCR
____________
-
HammerSandwich
- *Lifetime Patron*
- Posts: 1288
- Joined: Sat Oct 25, 2003 3:21 pm
- Location: 15143, USA
- Contact:
I doubt chylld would fit the 2-3/4" OD tubing through a PCI slot!DrCR wrote:Quick-Release Worm-Drive Hose and Tube Clamps
http://www.mcmaster.com , page 211
The advantages of spring clamps are that they offer constant tension, cannot be over tightened, and cannot loosen (from vibration, heat cycles, whatever). For chylld's intended customer (i.e. less DIY savvy), eliminating the possibility of clamp failure is appropriate.
The clamps at the top of page 211 look like a decent alternative, but they start at 9/16".
-
HammerSandwich
- *Lifetime Patron*
- Posts: 1288
- Joined: Sat Oct 25, 2003 3:21 pm
- Location: 15143, USA
- Contact:
I love this idea chyld. As a noob to W/C i would like the sytem to have a conversation starter look, kind of like the reserator. How would a clear case looK? Maybe an option you could charge a little more for?
EDIT: maybe even add a guage, thermometer place where you could see the water flow. Somethign to make it visually inteesting yet somewhat relevent. No neor, uv lights or cold cathodes. I love the metalic structual look of the reserator. It would be a shame if this thing looked like a a simple utilitarian rectangle w/2 fans. Radiators can look cool sometime
EDIT: maybe even add a guage, thermometer place where you could see the water flow. Somethign to make it visually inteesting yet somewhat relevent. No neor, uv lights or cold cathodes. I love the metalic structual look of the reserator. It would be a shame if this thing looked like a a simple utilitarian rectangle w/2 fans. Radiators can look cool sometime
DrCR wrote:What about something like these? They should be really easy to work with.
Quick-Release Worm-Drive Hose and Tube Clamps
http://www.mcmaster.com , page 211
HammerSandwich wrote:I doubt chylld would fit the 2-3/4" OD tubing through a PCI slot!
chylld wrote:2-3/4" OD tubing is well beyond the max useful size for watercooling
Perhaps I should have placed a little more emphasis on the word "like."
HammerSandwich wrote: The advantages of spring clamps are that they...cannot loosen (from vibration, heat cycles, whatever).
Interesting. I'll have to make sure I use some thread-locker if I ever go for the bolt type. Thanks for the info!
DrCR
____________
I understand that this watercooling box will be twice as expensive as the Reserator. Theoretically will this box offer more silent cooling power than dual reserators???
On looks, I know that a black and chome finish of the polar water block is prohibitiively expensive. How would seasonic-blue on the sides and bottom with a clear acryic top look?
On looks, I know that a black and chome finish of the polar water block is prohibitiively expensive. How would seasonic-blue on the sides and bottom with a clear acryic top look?
Those csp750's are powerfull for thier size. Wont they cause galvanic corrosion being made completly out of aluminum. I thought that mixing copper and anodized aluminum would corode everything.chylld wrote:the pump was an eheim 1048 and i don't think i'll be using that pump now because dual csp750's seem to be the way to go now especially since i want to be running an inline filter. (don't want crap coming out of my box
Using this watercooling box would create vastly superior cooling abilities than dual reserators with about the equivalent noise of a single reserator.
And those up and coming CSP-750's look very promising, thought of using those chylld? 2 of those are smaller than a single eheim 1048 and have the pressure+head higher than a 1250.
Im looking into buying one of those very soon. They look great, very compact and quiet looking too.
Anodized aluminium, i heared prevents corrosion. Anyway if your using water in any watercooling setup you need water additive (radiator additive - which is an anti corrosive), although i heared that the likes of zalman im shocked to hear, reccomend tap water. In a setup like zalmans i doubt it matters too much with non transparent tubes to let sunlight in which would trigger photosynthesis and therefore algae growth.
And those up and coming CSP-750's look very promising, thought of using those chylld? 2 of those are smaller than a single eheim 1048 and have the pressure+head higher than a 1250.
Im looking into buying one of those very soon. They look great, very compact and quiet looking too.
Anodized aluminium, i heared prevents corrosion. Anyway if your using water in any watercooling setup you need water additive (radiator additive - which is an anti corrosive), although i heared that the likes of zalman im shocked to hear, reccomend tap water. In a setup like zalmans i doubt it matters too much with non transparent tubes to let sunlight in which would trigger photosynthesis and therefore algae growth.
seal, i read the instructiona of the zalman website and they state that, distilled water is recomended,but that you can use tap water instead if you have to. I think saying tap water is okay is a big mistake, but they want to sell a product. Unfortunatly they dont even reomend adding a biocide.
Anodized aluminum only slows down galvanic corrosion. It doesnt stop it.Also you have to be carefull which additive you put in your water. Most super coolants and water weters increase the PH of your water to make the soution more basic(alkaline). That is beacause acidic water corrodes metal and promotes bacteria. Over time your water tends to become more acidic and you have to replace your coolant. However, unlike copper, anodized aluminum starts breaking donwn in solutiions with high PH levels also. This means that you have to make sure that your coolant is mildly basic (PH below 8.5), and thus your going to have to change it more often than copper and add more biocide.
I wouldnt mind using ananodized aluminum sytem, or a copper sytem. Mixing them togtehr is a bad idea. Note how the reserator uses anodized aluminum and gold platings. None of the copper comes in contact with the water. (its all gold plated to prevent galvanic corrosion).
Opaque tubes dont prevent aneroabic bacteria from groing in warm moist places. They just preven you from seeing them.
How do you know that this thing will have vaslty superrior cooling to dual resorators? A single reserator have been rated at 0.26 C/W and has a surface area of 1.3meters. Does chylds box have twice that cooling power at a quite level? Im not saying your wrong, id just interested in seeing the numbers.
Anodized aluminum only slows down galvanic corrosion. It doesnt stop it.Also you have to be carefull which additive you put in your water. Most super coolants and water weters increase the PH of your water to make the soution more basic(alkaline). That is beacause acidic water corrodes metal and promotes bacteria. Over time your water tends to become more acidic and you have to replace your coolant. However, unlike copper, anodized aluminum starts breaking donwn in solutiions with high PH levels also. This means that you have to make sure that your coolant is mildly basic (PH below 8.5), and thus your going to have to change it more often than copper and add more biocide.
I wouldnt mind using ananodized aluminum sytem, or a copper sytem. Mixing them togtehr is a bad idea. Note how the reserator uses anodized aluminum and gold platings. None of the copper comes in contact with the water. (its all gold plated to prevent galvanic corrosion).
Opaque tubes dont prevent aneroabic bacteria from groing in warm moist places. They just preven you from seeing them.
How do you know that this thing will have vaslty superrior cooling to dual resorators? A single reserator have been rated at 0.26 C/W and has a surface area of 1.3meters. Does chylds box have twice that cooling power at a quite level? Im not saying your wrong, id just interested in seeing the numbers.
-
HammerSandwich
- *Lifetime Patron*
- Posts: 1288
- Joined: Sat Oct 25, 2003 3:21 pm
- Location: 15143, USA
- Contact:
The 50k-hour rating is simply the MTBF of the motor and does not reflect on the pump in toto. Having read the thread at Procooling, I'm pretty confident that C-Systems knows what they're doing, but that has yet to be proven adequately IMO.mpteach wrote:those dual csp's do have a typeIII anodized coating and are rated to run 50,000 hours continuously(5 years).
thanks for all the replies guys, i've been busy these past few days sorting out stuff but can pretty much get back on it soon.
mpteach: a "conversation starter" look - possibly. i think anything external to the computer that has tubing running out of it is a conversation starter i don't like using a clear casing because it makes dust all that more visible. my favourite option atm is to use lacquered pine because it's cheap but also looks very very cool and it's something i can machine myself in my own workshop, can't do that for plastics/metal.
the reservoir will be visible from the front/side of the unit so you can see the water level at all times. it's also cool to see the water swirl about
the original design was pretty much a rectangle with 2 fans, that's one of the things i didn't like about it. the new version is completely different
re performance: yes it should kick the s*&T out of a dual-reserator. probably more.
re availability: i've set target dates in the past but due to things that have come up (revues, uni, work etc) i think it's safe to say that it'll be a good week or two between significant progress stages. that said, i want to have photos of a working prototype by the end of this month at the latest.
let me stress that i'm not trying to string you guys along here. i'm talking about my idea and how i think i'm going to implement it, the point of this thread is to get you guys thinking so that you can also throw ideas around (which you're doing - great!) and possibly even create your own watercooling inventions. don't be afraid to steal the thread, if you have an interesting idea, throw it on here.
brb lunch, will post rest of this later
mpteach: a "conversation starter" look - possibly. i think anything external to the computer that has tubing running out of it is a conversation starter
i don't like using a clear casing because it makes dust all that more visible. my favourite option atm is to use lacquered pine because it's cheap but also looks very very cool and it's something i can machine myself in my own workshop, can't do that for plastics/metal.the reservoir will be visible from the front/side of the unit so you can see the water level at all times. it's also cool to see the water swirl about
the original design was pretty much a rectangle with 2 fans, that's one of the things i didn't like about it. the new version is completely different
re performance: yes it should kick the s*&T out of a dual-reserator. probably more.
re availability: i've set target dates in the past but due to things that have come up (revues, uni, work etc) i think it's safe to say that it'll be a good week or two between significant progress stages. that said, i want to have photos of a working prototype by the end of this month at the latest.
let me stress that i'm not trying to string you guys along here. i'm talking about my idea and how i think i'm going to implement it, the point of this thread is to get you guys thinking so that you can also throw ideas around (which you're doing - great!) and possibly even create your own watercooling inventions. don't be afraid to steal the thread, if you have an interesting idea, throw it on here.
brb lunch, will post rest of this later
re pumps: i've contacted c-systems about them but they haven't replied yet which is a shame. it depends on cost though, if it turns out to be more expensive than a 1250 then i'll go with the 1250.
i'm not sure about the numbers of the system just yet, although one of the things i plan to do when it's finished it publish flow rate / pressure head / c/w performance figures for the box as a singular unit. i'm not smart enough to take a guess at the c/w but it'll be in the sort of region that will allow an xp2500 to be overclocked to over 2.5ghz easily. (depending on the cpu, of course)
i'm not sure about the numbers of the system just yet, although one of the things i plan to do when it's finished it publish flow rate / pressure head / c/w performance figures for the box as a singular unit. i'm not smart enough to take a guess at the c/w but it'll be in the sort of region that will allow an xp2500 to be overclocked to over 2.5ghz easily. (depending on the cpu, of course)
Just to add a few things which might help:
1. COST - you gave a figure below 250 british pounds but didn't mention exactly where. Thats almost AU$700. You will have to bring the price down to AU$400-500 for it to be competitive with the other WC kits out there. Anything more expensive and the average Joe/Jill will scrap the purchase.
Although you could justify yours is the best, etc. etc. - the n00b consumer won't care - they want to see in a second which is worth to buy. Most computer folks don't like to research deeply into subjects especially one like watercooling. My countless weeks of posting in forums, asking questions and browsing the net alone for information is worth a lot more than $400.
2. Do the Maths - have you performed a heat transfer analysis if what you envisioned will work? I am unfortunately just out of my chemical engineering degree when I learned watercooling so I doubt my old heat transfer lecturer would help me :p
By doing the maths first - it will save you considerable time, money and frustration from constructing a prototype.
3. Cool the hottest - I know it is not an easy task to cool a hot CPU & GPU + twin HDDs (forget the NBridge, it doesn't really need cooling) - but if you mention you want to make a "kick arse" WC system - yours must be able to handle the fastest/hottest rig out there. You could possibly add a AU$100-$200 premium here. Anyone that even considers to use watercooling in their PC are more often than not hardcore PC users/gamers.
4. Results - once you have completed the prototype stage and if it turns out to be economically feasible - make sure you don't forget to compare your results to other brands - also do not forget to have identical system setups to be able to obtain correct data.
5. Peace of mind - what all watercooling systems, units, etc. lacks until today is the guarantee if it will work. Everything is pretty much guesswork. Nobody publishes numbers that could be compared and the fact that there is limitless combination of pumps/waterblocks makes the choice even harder. Your unit must guarantee it could cool a fast/hot system and always give realistic conditions. If your test conditions are AMD XP1800+ CPU with ambient room temperature of 10C is basically useless. What happens at summer when the temperatures hit 35C? You can't use the computer! If you can, make it able to work all year around.
1. COST - you gave a figure below 250 british pounds but didn't mention exactly where. Thats almost AU$700. You will have to bring the price down to AU$400-500 for it to be competitive with the other WC kits out there. Anything more expensive and the average Joe/Jill will scrap the purchase.
Although you could justify yours is the best, etc. etc. - the n00b consumer won't care - they want to see in a second which is worth to buy. Most computer folks don't like to research deeply into subjects especially one like watercooling. My countless weeks of posting in forums, asking questions and browsing the net alone for information is worth a lot more than $400.
2. Do the Maths - have you performed a heat transfer analysis if what you envisioned will work? I am unfortunately just out of my chemical engineering degree when I learned watercooling so I doubt my old heat transfer lecturer would help me :p
By doing the maths first - it will save you considerable time, money and frustration from constructing a prototype.
3. Cool the hottest - I know it is not an easy task to cool a hot CPU & GPU + twin HDDs (forget the NBridge, it doesn't really need cooling) - but if you mention you want to make a "kick arse" WC system - yours must be able to handle the fastest/hottest rig out there. You could possibly add a AU$100-$200 premium here. Anyone that even considers to use watercooling in their PC are more often than not hardcore PC users/gamers.
4. Results - once you have completed the prototype stage and if it turns out to be economically feasible - make sure you don't forget to compare your results to other brands - also do not forget to have identical system setups to be able to obtain correct data.
5. Peace of mind - what all watercooling systems, units, etc. lacks until today is the guarantee if it will work. Everything is pretty much guesswork. Nobody publishes numbers that could be compared and the fact that there is limitless combination of pumps/waterblocks makes the choice even harder. Your unit must guarantee it could cool a fast/hot system and always give realistic conditions. If your test conditions are AMD XP1800+ CPU with ambient room temperature of 10C is basically useless. What happens at summer when the temperatures hit 35C? You can't use the computer! If you can, make it able to work all year around.
hiya,
haven't had time to do anything with the project because i've been bogged down with uni work, thesis, paid work and film directing.
i have given it some more thought though and by the looks of things it isn't good. technology these days is all about mobility, wireless this, wireless that, passive cooling this, convenience that. watercooling simply does not fit in to the picture. that said, at the present time i can't justify the cost of investing in a technology that i suspect could very well be redundant in a number of years' time.
haven't had time to do anything with the project because i've been bogged down with uni work, thesis, paid work and film directing.
i have given it some more thought though and by the looks of things it isn't good. technology these days is all about mobility, wireless this, wireless that, passive cooling this, convenience that. watercooling simply does not fit in to the picture. that said, at the present time i can't justify the cost of investing in a technology that i suspect could very well be redundant in a number of years' time.
passive watercooling will fit into the picture in a year or two when passive heatsinks (or even normal HSFs, if we're unlucky) for any modern CPU/GPU are so big you can't get air through the case any more.chylld wrote:technology these days is all about mobility, wireless this, wireless that, passive cooling this, convenience that. watercooling simply does not fit in to the picture
also, with all this talk of water additives and such, anyone have any idea what solvents are safe with what hose materials (or exactly what is in most additives, besides a surfactant)? Would be a lot more convenient to just use a few drops of chloroform as a biocide. I'm familiar with some microbes that live on iron, not sure about copper or aluminum though.
These people make some neat hoses, lucky their online ordering is broken though, probably don't want to see the price... then you'd just need blocks and a res all made out of the same metal and plain distilled water should be ok.
http://www.spssales.com/