• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Finished My Mineral Oil Cooled PC

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Direct link to update with pictures: http://forums.bit-tech.net/sho...?p=1922914#post1922914

Was bored tonight, so I decided to do a little soldering...

http://i28.photobucket.com/alb...log/log_progress85.jpg

http://i28.photobucket.com/alb...log/log_progress86.jpg
4 LEDs in series with a 22ohm resistor for about 2.07v drop across each diode. I have the positive lead connected to the 12v line and the negative lead is connected to the 3.3v rail. Do the math -> 8.7v. Each is held in place with a generous glob of hot glue.

http://i28.photobucket.com/alb...log/log_progress84.jpg

http://i28.photobucket.com/alb...log/log_progress83.jpg

http://i28.photobucket.com/alb...log/log_progress82.jpg

http://i28.photobucket.com/alb...log/log_progress87.jpg
Not overly bright so they won't get annoying like blue LEDs. Looks pretty decent.
 
At first I thought this was purely a practical rig. Now I see it's going to have a little style! I like very much, I'm a chemical engineer who is a total noob around machine shops.

Good luck, looks like fun!


 
Nice work :thumbsup:

Keep us updated; I would like to see the finished product, and see what your temps are like.
 
Originally posted by: 2dt Drifter
The amount of patience you have working on this is mind boggling. Good luck
Click to expand...

i second this.

:X

trying to keep my comments out of this, because its very dangerous to do this. But it seems like you did all your research.
 
I'm curious about something.

so the entire acrylic/plexiglass section is going to have the mineral oil, correct? If so, at the bottom of that section it will have a lot of static pressure from the weight of the oil.

density * gravity * height = pressure

~800 kg / m^3 * 9.81 m / s^2 * height of the oil in meters = pressure in Pascals

I'm not trying to rain on your parade or tell you it won't work. Just wanted to make sure you had thought of this. Since I have absolutely no idea how much pressure a piece of plexiglass can hold.

How are your seals going to work? Leakage would be the suxorz.



Other question: how do you know if a mobo fan header can handle X fans in parallel ?? Just last weekend I was thinking about doing exactly what you did for 2 fans, but got scared that they might max out the header.


Still looking forward to results!



Edit... just had a thought which you probably were going to do anyway - but since water is heavier than mineral oil while being less viscous, you should just do a pressure/leak test with water and if that holds then mineral oil should hold for sure.
 
Some good questions!
Originally posted by: wired247
I'm curious about something.

so the entire acrylic/plexiglass section is going to have the mineral oil, correct? If so, at the bottom of that section it will have a lot of static pressure from the weight of the oil.

density * gravity * height = pressure

~800 kg / m^3 * 9.81 m / s^2 * height of the oil in meters = pressure in Pascals

I'm not trying to rain on your parade or tell you it won't work. Just wanted to make sure you had thought of this. Since I have absolutely no idea how much pressure a piece of plexiglass can hold.
Click to expand...

The tank itself will be glued together with a chemical called methyl chloride. I bought a commercial version of the stuff called Weld-On #4. It basically melts the edges of the joint between the two pieces of acrylic and dries clear. Then for water tightness, I'll seal all of the joints with silicone. The corners are reinforced with aluminum angle. This pic: http://i28.photobucket.com/alb...log/log_CaseMkIV_2.jpg shows what I mean. And in this pic: http://i28.photobucket.com/alb...log/log_progress78.jpg you can see how the angle is attached to the rest of the frame.

How are your seals going to work? Leakage would be the suxorz.
Click to expand...

I mentioned the corners will be sealed with silicone. The 3 bulkhead fittings that will go through the bottom sheet of acrylic have seals and I will also seal around them with more silicone. The rest of the tubing will be secured with the usual water cooling fittings and clamps.

Other question: how do you know if a mobo fan header can handle X fans in parallel ?? Just last weekend I was thinking about doing exactly what you did for 2 fans, but got scared that they might max out the header.
Click to expand...

Don't worry, I checked beforehand 😉 My fans will be plugged into this: <a target=_blank class=ftalternatingbarlinklarge href="https://www.crystalfontz.com/product/SCAB.html">https://www.crystalfontz.com/product/SCAB.html</a> It has 4 headers. I'm going to wire each set of 3 radiator fans together and plug them into 1 header. Each NF-P12 is 1.08w, 3 x 1.08 = 3.24w. I'll take the RPM monitoring from 1 fan so I can still get an RPM reading too. I ordered a super ridiculous high speed 120x38mm fan: http://www.petrastechshop.com/12x38scdfuka2.html for moving oil over the CPU heatsink. I will probably control this fan with the SCAB as well. This fan uses 7.2w.

The SCAB is rated for 1.5A per fan header with the total of the 4 headers not to exceed 4A. So that's 18w per channel, not to exceed 48w total. At max I will be using 13.7w.

This post shows the adapter I made to connect 3 fans together in parallel for the SCAB: http://forums.bit-tech.net/sho...=1931216&postcount=167

Still looking forward to results!

Edit... just had a thought which you probably were going to do anyway - but since water is heavier than mineral oil while being less viscous, you should just do a pressure/leak test with water and if that holds then mineral oil should hold for sure.
Click to expand...

Yes, I will definately leak test with water first 😉 Wouldn't want expensive transformer oil leaking all over the place.

Thanks for the questions and your interest!
 
lol @ the "a chemical called methyl chloride"

I'm a chemical engineer and it just so happens my boss patented a process to convert methane to methyl chloride, which was bought by BP... the royalties of which have gotten him a very nice house in Los Angeles. 🙂

Anyway, something else sounds a little funny to me. Fans are specifically meant for moving air... not oil. To move oil you literally need a propeller!! (Not exactly a fan... they are designed differently!)

To push through high viscosity oil I wonder how fans will like that kind of heavy loading. We know that fans are not very high torque devices because we can stop them with literally something like a piece of paper or a finger, without catastrophic results to the paper or finger.

I'm not sure if (a) the fans will have a hard time starting up in oil, or (b) if they will die a bit quicker from the constant, heavy load since they are not meant for high torque applications or (c) both or (d) neither.

I mean pumps are high torque, low speed... fans (or blowers) are low torque, high speed.

In fact... in most high precision turbines (compressor, or generator) if you even get a little liquid in there, the thing is finished. Just like if you get air in your pump, it can hardly do jack shit...

I would think you want something more torquey for your CPU "fan". Substituting high RPM/low torque where what you need is high torque/lowRPM could potentially be a problem (and even if it ends up working out... you're still using the wrong "tool" for the job.)

Then again, I'm guessing if you're trying this then that means someone else has tried it before you?


Sorry for FUDding in your thread 🙁

Don't forget rule #0 of being an engineer. If it works, don't fix it.

 
Originally posted by: wired247
lol @ the "a chemical called methyl chloride"

I'm a chemical engineer and it just so happens my boss patented a process to convert methane to methyl chloride, which was bought by BP... the royalties of which have gotten him a very nice house in Los Angeles. 🙂

Anyway, something else sounds a little funny to me. Fans are specifically meant for moving air... not oil. To move oil you literally need a propeller!! (Not exactly a fan... they are designed differently!)

To push through high viscosity oil I wonder how fans will like that kind of heavy loading. We know that fans are not very high torque devices because we can stop them with literally something like a piece of paper or a finger, without catastrophic results to the paper or finger.

I'm not sure if (a) the fans will have a hard time starting up in oil, or (b) if they will die a bit quicker from the constant, heavy load since they are not meant for high torque applications or (c) both or (d) neither.

I mean pumps are high torque, low speed... fans (or blowers) are low torque, high speed.

In fact... in most high precision turbines (compressor, or generator) if you even get a little liquid in there, the thing is finished. Just like if you get air in your pump, it can hardly do jack shit...

I would think you want something more torquey for your CPU "fan". Substituting high RPM/low torque where what you need is high torque/lowRPM could potentially be a problem (and even if it ends up working out... you're still using the wrong "tool" for the job.)

Then again, I'm guessing if you're trying this then that means someone else has tried it before you?


Sorry for FUDding in your thread 🙁

Don't forget rule #0 of being an engineer. If it works, don't fix it.
Click to expand...

You bring up a good point. I had the same concern about the fan. (Only that Scythe fan will be in the oil). The folks at Puget Systems have this to say about it:

# Won't the mineral oil put too much load on the fans, causing them to fail?

Answer: Even if they did, it wouldn't matter. We left them running just because they were fun to look at, and to answer just this question! The answer is no -- after two months, the fans are spinning exactly like they did at the beginning. While it remains to be seen if they will burn out more quickly, they definitely have lasted these two months with zero loss of speed.
Click to expand...

and further more:

Update After One Year:

* The fans continue to run strong. We're amazed by the number of people that continue to be concerned that the fans will burn out. It is now very clear that the fans have absolutely no problem running in this environment.
Click to expand...

They don't seem to have a problem with the fan they're using so I hope I won't either 😉
 
Looking good mate, keep the updates coming. btw my hat comes off to you in terms of your skill and patience. Well done.

Just out of interest, how much has it cost you in materials so far for the case, electrical components and oil gubbins (rads, fans, tubing, fittings, switches, etc.).
 
Originally posted by: daw123
Looking good mate, keep the updates coming. btw my hat comes off to you in terms of your skill and patience. Well done.

Just out of interest, how much has it cost you in materials so far for the case, electrical components and oil gubbins (rads, fans, tubing, fittings, switches, etc.).
Click to expand...

Thanks! $980.04 🙁 And I haven't bought any big PC hardware either...
 
Originally posted by: legoman666
Originally posted by: daw123
Looking good mate, keep the updates coming. btw my hat comes off to you in terms of your skill and patience. Well done.

Just out of interest, how much has it cost you in materials so far for the case, electrical components and oil gubbins (rads, fans, tubing, fittings, switches, etc.).
Click to expand...

Thanks! $980.04 🙁 And I haven't bought any big PC hardware either...
Click to expand...


holy hell. I'll stick to homebrewing as a hobby. :beer: :beer: :beer:
 
@Legoman666. You;re almost finished, well done.

Are you creating your own MB tray or did the backplate come with its own tray?

If the former is the case, how are you going to fab it; I can see the main problems being:
1. Getting the stand off mounting holes in exactly the right locations for an ATX or mATX MB (not sure which type of MB you are using).
2. The stand offs will have to screw into the tray some how. This can be done by using thick aluminium for the tray and tapping the stand off screw holes (weight is a problem with this one), or drilling holes in a thinner sheet of aluminium and welding nuts on to the backside of the holes or using these snap-in plastic stand offs. Usually, the MB tray in a case comes with threaded metal inserts for the stand offs.

Alternatively, you could buy a cheap aluminium ATX case, such as this one and cut out the MB tray and back plate. You could also scavenge the drive bays as well.

Just my thoughts on the matter. I'm sure you've already thought about it (as you have with everything else on the build) and decided what to do.
 
I am a retired machinist who made part for the GPS satellites so I can see how much work you have done...

Well gone...

Myself I am happy to mod a store case 🙂
http://home.comcast.net/~rarebearz/rarebearcomputer/

One thing about the Mineral Oil..
The thinner the liquid the faster the heat transfer, I would try antifreeze as it has properties the absorb heat faster..

In fact maybe Alcohol might work better

EDIT: I read more and I guess the whole mobo is covered in Mineral Oil?? I have never seen this before other than a transformer 🙂
Very Cool.....


As for solid square stock aluminum I would use extruded stock from a company like 80/20

They have a store at ebay for over runs and returns
http://stores.shop.ebay.com/80...age-Sale__W0QQ_armrsZ1

I cant wait to see finished computer...

Thanks Much for Sharing..
 
Originally posted by: daw123
@Legoman666. You;re almost finished, well done.

Are you creating your own MB tray or did the backplate come with its own tray?

If the former is the case, how are you going to fab it; I can see the main problems being:
1. Getting the stand off mounting holes in exactly the right locations for an ATX or mATX MB (not sure which type of MB you are using).
2. The stand offs will have to screw into the tray some how. This can be done by using thick aluminium for the tray and tapping the stand off screw holes (weight is a problem with this one), or drilling holes in a thinner sheet of aluminium and welding nuts on to the backside of the holes or using these snap-in plastic stand offs. Usually, the MB tray in a case comes with threaded metal inserts for the stand offs.

Alternatively, you could buy a cheap aluminium ATX case, such as this one and cut out the MB tray and back plate. You could also scavenge the drive bays as well.

Just my thoughts on the matter. I'm sure you've already thought about it (as you have with everything else on the build) and decided what to do.
Click to expand...

Or this case for $30 + shipping. Solid aluminium construction.
 
You must log in or register to reply here.

TRENDING THREADS

Back
Top