I put up the 1/4 middle plate for sale: I'll order 1/4" by 1/4" pieces, with some 1/4" by 1/2". This will save me a lot of cutting.

I went to Lowes, and picked up a 3/8 to 1/2 adapter, as well as a 1/2 threaded adapter, into which my 1/2" polypropylene barb fits very nicely. I also picked up flux, and 1/2 lb of acid core solder.

Why the adapter?

Well, "1/2" copper piping isn't exactly 1/2, in fact, it's more like 5/8, on the outside. "3/8" is actually 1/2 on the outside. Since I got fittings, those real measurements are on the inside, so they're all even bigger!

The plan is to solder the "3/8" adapter right on the edge of the block.

I also need about 4" total of 1/2" copper piping, to mate the joint between the adapters.


Does anyone have any extra 1/2" copper piping?

Ok, I picked up a 5 foot section of "1/2" copper tubing. I'll use the rest in the watercooled PSU array :D .


I need some opinions: I can get a propane torch for ~$25, or the MAPP gas torch, for $40. (add tax).


Can I get away with the propane torch here?

[edit] It seems that BladeRunner used a butane microtorch for his block: I'm thinking that the propane torch will work pretty well!

Well actually the statement is true. There is more to that face of the HD than just the chips. The 10000 rpm motor is mounted inside the drive on the same side as the controller board.

Put your finger on one of those chips, the motor on my GPX60 gets warm(though it's 7.5k not 10), the chips get hot. I can only go by my experience though :) ...

Any torch should do BB, I use an old Primus 'pump up' parafin(Kerosene to you?) torch for my soldering, works admirably :D ...

here's another very simple way to cool your HD,
found this over at Overclockers forum
its made by oz2cpu

http://www.webx.dk/oz2cpu/pcmod/w-hd-d.jpg

here are some of his temp results, Scroll down though http://www.webx.dk/oz2cpu/pcmod/water-results.htm#HD

Nice. I've seen it before, but I was concentrating on the PSU at the time.

He used a 2mm baseplate, where I've got a 3.2mm, but I have more surface area for the water, plus the heat won't have to travel through a solder joint.

I've also thought about using grease between the HDD and this block. If the HDD is painted, I'll have to lap it :D

I like the paint job. I also like the fact that he should have no problem mounting these drives in a standard drive cage.

dont know how, but i keep coming up on all sorts of hd blocks since i've read your post :)

here's another one, http://cgi.ebay.com/ws/eBayISAPI.dll...&category=3673
and only $30!!!!! lol
:D

Thanks MadDog, the propane torch is working out just fine! (as I light a cigarette with it :cool: )

http://www.bernzomatic.com/Images/TS4000KC.jpg

just thought i'd chime in here with my design for my rotor style (plagarism?? where?) hd block, originally the inlet barb was positioned on the top flowing freely into the chamber but my bench drill isn't up to it (size restrictions). Its built out of mammoth 3/4" copper with a 2mm sheet soldered on to top it off, if i had thought harder b4 ordering the copper (ordered it several blocks ago) i prolly woulda gone for 6mm plates instead.. ah well live and learn.

Any comments critisicms welcome, but leave my spelling alone :)

One note: wouldn't you rather have the barbs come out the back, rather than the side? A typical case is only 8 inches wide!

the two barbs at the bottom 'facing' u, join together behind the drives, like the block posted in this thread if thats the ones u mean? the other two i'm either gonna stick an elbow on it, or a straight barb facing down, just depends on where i decide to stick it in the case

i was thinking of something similar, but mine encased the whole hdd, it had 1 channel on the side, and those channels branched off into many more that went across the surface of the HDD (both sides), the many channels would then empty into 1 in which it would then leave the block.

Erm... doesn't most of the heat come off the top/bottom of the drive so the coolers at either side won't do much good.

The only drop you would see in temperatures is placing them with air spaces between them in 5 1/4" bays.

~ Boli

(just a thought)

i figure the added mass of the block would quiet the drive some...

Actually, the sides is probably one of the best places to cool a hard drive from, since it is flat and there is a direct conducting path to the major heat producing components.

The large flat metal "lid" which most people initially thing of watercooling, is actually insulated from the rest of the block by an o-ring.

And cooling the base with the pcb and motor spindle is tough.

The sides is the way to go.

8-ball

Thanks 8-ball - first real post I have found stating sides is better than top.

I plan on doing the tubing-on-plate approach, except without the awful 90o turns. Not enough room for 45o turns, so I've got my 90o sweeps which will have to be good enough.

However, I've stopped for more thinking.

See, seeing as the drive is water cooled, I should think about a nice enclosure to quieten them down. This is where I am at now - what to make it out of, how to "pack" it to deaden drive noise.

Funny thing about being fanless - the drives are SOOO noisy, where as before water cooling, I assumed they were silent coz I never heard them....

*grumbles*

That part is SO true... the worst part about loud drives is it is the high pitched whine that drives you nuts when you try to sleep in the same room as your computer.

I think that was Volenti.

Indeed, and I believe it's linked to, earlier in this thread.

Does anyone know where I can find 3/8 copper street elbows?

Good thread, my thoughts and questions.
 

This is an interesting thread, I just started a different thread on the same subject because I didn't know this one existed... (Hadn't read back far enough) http://forums.procooling.com/vbb/sho...&threadid=6995

The problem I have with most of the designs mentioned so far is that I want to put my drives in the 3.5" internal bays, which means side cooling isn't really an option. I suppose one *could* design a block that attached outside the bay, but I really doubt that it would work well as there would be to many thermal interfaces.

I have been thinking in terms of soldering coper tube to a flat plate, and putting that on the top (cover) side of the drive. I wasn't planning to do anything to the controller side, as I didn't see how to get good thermal contact with the drive.

However I noticed the block that was being sold on e-bay mentioned by lukasz70 (it didn't sell BTW, got one bid of $5 and didn't make reserve) did bolt onto the bottom of the drive, and presumably cleared everything. I may take a closer look once I actually get the drives in hand. Do you think there would be much benefit from putting a cool plate near a drive with some airspace in between? It would at least keep the air next to the drive cool...

I'm not sure about this 8-ball, where do you get your info on the heat sources in hard drives? According to at least one article I've read (I think it was on Storage Review.com, but don't quote me on that) the drive controller components and spindle motor don't make all that much heat. Most of the heat from a fast drive is generated inside the case by air friction of spinning the platters. If that is the case, then cooling the lid might be the optimal approach since that surface is the one with the most contact with the air (and it's a safe bet that there's LOTS of turbulence inside a drive case :dome: )

I do know that I've seen at least a couple of mentions of people just slapping a CPU or GPU WB on the lid and getting lower temperatures on the entire drive.

Gooserider

think you'll find that the HD manufacturers design the drives these days to sink the heat to the sides, use the case as a heatsink if they can. _i think_!

how many cases u seen that have a plate on the top or bottom of a 3.5" hd?

Where do all the watts go?
 

This is an interesting question, I might try giving Seagate tech support a call on it and see what the experts have to say...

My take on it is that the sides don't seem all that great as a heat dissipator. The drive manufacturer doesn't know what the mounting conditions for the drive will be. Best case, the drive is screwed to a steel cage, with no thermal interface material. Not a great heat sink at best. But I've seen cases plastic with plastic cages, and / or plastic drive rails which would be even worse! There is more and more of a move to getting rid of rails, and screwing the drive to a steel cage, but my understanding is that this is driven more by manufacturing costs and a desire to improve chassis electrical grounding than to solve heat problems.

I haven't seen many cases of ANY kind of liquid cooling on drives, but I would say I've seen about equal numbers of side coolers and top/bottom coolers. The air cooling solutions I've seen appear to me like they are trying to do more top/bottom cooling than side cooling.

Aside from the question of hot components on the controller board, the two biggest heat sources I see in a hard drive are the spindle motor and air friction heating done by the spinning drive platters Both of those sources are inside the drive, and can't be cooled directly. The sides are relatively thick, and thus slow to transfer heat, they are also fairly small in surface area. The top is thin, thus will transfer heat quickly, and has a lot of surface area. My understanding of the conditions inside the running drive is that there is more than enough air circulation to ensure that if any part of the drive shows a temperature gradient, the air will rapidly convey any heat from the hotter areas to the coolest one. The lack of a clear metal thermal path shouldn't make much difference as I see it.

I also mention yet again that I want to put the drives in my 3.5" internal bays. Edge coolers prevent that, but it should be quite easy with top/bottom coolers.

Gooserider

Re: Where do all the watts go?
 

You might consider applying Volenti's solution, and weld copper tubes to the sides of that drive cage.

Instead of soldering the entire pipe to the plate so the contact surface is minimal. Wouldn't it be better to rip the pipe in half and then solder it on. So there is a more direct path for the heat to take.

Secondly. After about a month of planning and prepping I silenced my PC. The first night with the "silent" comp I almost went nuts because of the HD noise. My solution was hooking up a ramdisk and running SETI and AIM of it. It didn't solve the problem 100% but it sure did cut down on the number of times when the HD had to spin up.

Does anyone have any numbers on how quiet you can get a HD if you water cool it surround it in foam?

It's my plan to do just this. I have an LRR, so one outlet will go to GPU and the other outlet to NB. Then both loops will proceed to either side of the drive cage for a little loop. They will only be combined when back at the reservoir for a submerged pump. Pump goes to radiator, and then to CPU and off we go again.

I figure I can balance the flows in each side as some runs and bits will be longer and more curvey then others, so I will need to mix and match until its balanced.

The plan is to take to copper piping with the angle grinder, but I am yet to start....