My cooling towers(pics), not finished.
 

http://filebox.vt.edu/users/jtorres/...C/PICT0003.JPG
http://filebox.vt.edu/users/jtorres/...C/PICT0006.JPG
http://filebox.vt.edu/users/jtorres/...C/PICT0008.JPG
http://filebox.vt.edu/users/jtorres/...C/PICT0009.JPG

Pump and block should be in soon and I'll post results. Just wanted to share my progress :D

that is pretty nice. i recommend running them in seriies not parralle though. i hope you use alot of anticorrosive additive to stop all that aluminum from corroding the copper

Yah that aluminum gives me the willies! Looks really good though.

I'm going to try parrallel first then series to see which gives better temps. I already ordered a bunch of hydrx, will come with my mcw6000.

wow.. Looking very nice.. can't wait to see the finished setup.. keep us posted! :dome:

It's the copper that corrodes the aluminum, not vice-versa, but yeah good advice.

where did you get the towers?
they look cool :)

jman

also nice job cutting the fans to fit didn't realize how it was done till i examined the close up

12" transmission coolers off of ebay.

The pump and the block w/ hydrx should all get here this week. I have to build a box to mount the towers, hold the pump, ac/dc converter, switch, rheostat as it will all be external. But if I get the stuff before friday I'll have results and final pics up then.

Very interesting. I'd be interested in knowing how loud it is, as well as how well it performs in cooling the system. People make musical instruments out of tubing like that because that shape resonates well. :D

Yeah it definitely resonates at full blast when the tubes arent on the coolers. I turned the fan on to test the mounting once and was like "awww crap" because it was pretty loud. But once the tubes are on the coolers and the fans are turned down a little you dont notice it.

I do want this to perform the best it can but dont forget Im putting this in a living room by the projector so I need it to be queit =)

You, sir, must have been reading my mind. :D I've been trying to figure out a way of making a more vertically-oriented radiator setup (compared to most of the horizontal box setups).. originally entertained the idea of making a semi-passive setup similar to Xine cooler system (rather than one continuous loop but tank to tank parallel tube system) utilizing a "cross-flow fan" and was struggling with the idea of having to solder all those tubes. I also saw the radiator setups from another company but they wanted obscene amounts of cash for their passive radiator setups.. Now you show this.. After seeing your work, now I'm inspired to build one.

Question for you... the interior of the radiators, is there a mechanism (ie - like blades or vanes) inside the radiator to increase turbulence inside and get more of the water in contact with the radiator surface? It's not much, but I think it's one of those "last 5%" type of modifications to get the most performance out of it.

Ozymand

The inside is finned. Its almost like the fins from the outside protrude inside. Course the cylinder part is all one chunk of aluminum. The fins alternated between two heights. Big was about half an inch and little was about a quarter of an inch. Would have liked the internal fins to be larger but as its been stated before this isnt made to cool water.

Im curious to see what the interior of the dual pass ones look like... Might be a lot more intricate. Course it might be a lot worse too =)

Ah... nice then.. better than a smooth surface (what I was more worried about)... the fittings on the ends of them.. are they normal NPT fittings?

Definitely not. Im sure it varies from cooler to cooler, and this may be why I got mine for so cheap, but I have no idea what the threads were on these things. I tried everything that was close to 3/8" at Home Depot and none really fit. I went tiwh 3/8" npt nylon fittings. Putting teflon tape over the fittings actually helped a little but I still had to give a great deal of torque to get them all the way down. Pretty sure I destroyed the fittings by putting them in there. They are very secure and water tight but I doubt I'll even be able to get them off.

My new suggestion is to avoid teflon tape and go to RTV. I have seen bits of teflon tape floating in one of my new systems and it didn't make me happy. With a bit of clear RTV on the threads, screwed a very snug hand-tight, it doesn't leak and it doesn't put bits of tape into the coolant.

As to the cooling effectiveness of the vertical solution, it seems that it shouldn't be as effective per unit noise as a solution which is thin (air moves through wide area but thin distance). After all, a longer travel through the resistance should be more restrictive and it should be picking up less heat as it stays in contact with the radiator. That's my off the navel theory.

Yes?

Agree with you on the tape but lost you on the second part of the thermal dynamics :D

What is the outside diameter of that trans cooler (fin to fin)? I see a couple places online that sell em at different legths but they dont post the diameter.

could you be so kind to give us a picture of the fins inside.

The very outer diameter of mine was 2.9785" Perfect for the 3" ID tubing :D

Wish I could man but theres no way I can get the barbs off now. A few others have asked, wish I took pics before I put it together.

Yeah that is perfect.

NPT is a tapered thread. It should start sloppy but tighten and eventually seize up as you screw the parts together. Normally, some thread will still be visible when the joint is done.

EDIT: If you really want to get the barbs off, put the joint in the freezer. Both metals will contract away from each other.

James-

Is there any update as to the status of your cooling tower? I'm thinking about doing the exact same system as yours to cool my HTPC that drives my projector (damn noisy 120mm fan even undervolted). I love the idea of the WACC (Wet & Chilly Chips) cooling tower, but they are simply TOO expensive to get shipped to the US.

There are currently two possible choices for my setup using a dual pass 18" cooler or using a single pass 24" (or possibly 30"). The only problem with the dual pass is that an airpocket will form at top of the cooler and will result in an imbalance of fluid levels. I think that some sort of fill point will need to be inserted at the top of the dual pass cooler for simplicity. I have also hunted quite hard for a drawing of the inside of the dual pass, but have not found one.

I would LOVE to passively cool my Athlon64 3000+ (not overclocked), but depending how well it does, i may have to have some sort of very very very quiet fan to aid in cooling and permit 24/7 operation (w/SETI@Home). IF this system works very well, I may build a second one for an upcoming build (Athlon 3500(90nm), NB, and 6600GT).

Any Ideas or tips for building the system?

-DrPyro

Hi DrPyro,
Im pretty much done with this project. I wanted to paint the mdf box they are mounted on (especially since my lazy butt put a cup on the wood while I was watching a movie and it left a nice little ring in the wood :D ) but I dont have a place to paint it.

I can take pics when I get home. I'm waiting for my projector so this box is sort of just hanging out in the middle of the room.

I got rid of the fans all together. a 2500+ barton at 12 x 200 and an nforce2 (which gets very hot, dont care what all the cooling-the-nb-is-a-waste people say) are in the loop. Was going to add a 9600xt but I didnt have the time to make another top for my maze4 GPU block so it could be mounted with the barbs pointing towards the end of the card, rather than towards the top. My microatx case gives literally about 7mm of room between the top of the AGP card to the top of the case so I couldnt bend tubes over the card, I wanted to bend them around the end and to the nb. Sorry little OT :D But with the barton and nforce2 on it temps were right at 50c and 30c with a 25c ambient with a full load 24/7 (FaD running 24/7). Thats completely passive on the towers. I have 3 60mm fans undervolted to 5v in the case (2 in the PSU and one exhausting). I doubt a 24" or larger one would have much trouble cooling your system completely passively. These are much more efficient than the reserators, the ID is much smaller but the actual surface area the water goes over is much much more because of the internal fins.

Using the towers is pretty straight forward. The mistakes I made were trying to pump the water down the coolers which trapped huge amounts of air, and not really deciding on what pumps I wanted to use. originally I tried a mag3 but it was too loud, then I tried a single 1048, that was too weak, then I tried 2 1048s which was perfect but you guys are gonna laugh when you see the ClusterF* of the routing job I did inside the MDF box to fit the two pumps in there.

Someone mentioned the dual pass Tranny coolers before, I have never had one in my hands so Im not sure about their design. I dont know if they are the same as a single pass (Just a tube with internal and external fins) or they actually have a divider in the middle so it forces the water all the way up then all the way down. It might be ghetto like the reserator where the inlet and outlet of the cooler are just right next to each other. If its a seperate path that'd be awesome, else I wouldnt touch them.

I thought of putting my T line some how at the top of the towers. It would be much much easier that way. With just the T line at the inlet of the first pump it takes forever to fill it. I have to fill the T line, move the box around to get water in both pumps and in some of the tubes, turn it on, wait for the pumps to cavitate too much, fill the T line, tilt the box around, etc etc etc until the towers eventually get full and spill over then its easy. I didnt put the T line at the top though just for aesthetics.

I'll be sure to post final pics when I get home tonight.

I forgot to reply to this!

That was just bad writing, sorry! What I mean so say is that a thin but wide radiator is better because air can more easily travel through it. So, a cooling tower that has many shorter fins should perform better than a tower with fewer but longer fins. On the other hand, you get better convection currents with a taller unit (think chimney).

Truthfully, I don't know which effect will dominate, so I'll TRY to keep my mouth shut. :D

If the aluminum were anodized would that stop the dissimilar metal corrosion?
Lucky