• 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.

Help about a new case project and cooling

whitewarrior11

Senior member
Hello everyone!

I thought of building a new case for my computer, built out of wood maybe (I know it's not new but anyway, it's the only material I know I can handle). I don't know how to bend or weld any kind of metal (aluminium or steel) and I don't know either how to build a case out of acrylic (besides buying one already precut).

So the thing is, I read on the Internet about wooden cases and I have a couple of questions for you guys to help start this project:

1) What is the thing about Electromagnetic Interference? Can I just solve the problem by applying, say, aluminium inside the case?

2) I know wood is a pretty good INSULATOR, meaning heat won't be dissipated through the case itself. I also heard this isn't much of a problem since airflow is far more important in getting the case and the inside cool. Is this right?

3) I currently have a pretty recent case with a temperature sensor on the front panel. What is the maximum temperature my CPU should run at before, say, potential problems? It is currently at 110 F with substantial load (video encoding) and usually around 85-90 F when idle; it's a Pentium 4 3.0 GHz, socket 478 (I don't know much of the other specs but I do know it was the "last" 3.0 GHz Socket 478 P4 to be offered before Intel started to use LGA775).

4) Any design ideas about the inside of the case (like how the components should be placed and things like that) would be really appreciated.

5) Finally, I saw an idea on another forum about silencing a fan inside a custom-built wooden case and I would like your input on this. The intake 80mm fan was on the bottom of the case (that is, facing the table you put the case on) and taking air from the bottom inside the case through a sort of small duct. To be somewhat clearer, the fan didn't take the air directly from outside the case: it was taking it from a small duct below it which in turn was connected to the outside (a sort of sound-dampening duct). Could this work?

Thanks in advance for your input!
 
Acrilic cases are a thing of the past...Hopefully. They build static charge and do not age well.
Dirty inside and out. New ones can be picked up real cheap now for these reasons.

Wellcome to Anandtech, ww11. I hope your chair has arm rests as you will need them to hang on for the ride 😛

Wood construction of a PC case has been done. I saw pics recently of a PC inside a deep desk drawer.

Check over at silentpcreview.com , search cases and damping forum for "wood".
You won't find much wood here 😉 that's a joke 😛

Come back to us and post what you find, if you will. Lots of decent folks here.

There is one member here that is just compleating a wooden case project. If I can remember who it is I will PM him.


GY
 
Thank you! I will check silentpcreview.com and post here what I find. Maybe I will help someone else..😛..although wood doesn't seem to be a very popular material 🙂
 
Acrylic is pretty easy to use but difficult to keep nice looking (as GalvanizedYankee said). You can easily cut it with a jigsaw and glue it together with epoxy. You wouldn't want to weld the thin metal that you would use for a case, instead use rivets. You drill holes, then use a special tool - squeeze the handle - done!

1) EMI - Yes it is real. It may or may not matter. I have noticed EMI interfering with television antenna reception plus sometimes speakers will pick up the noise. Any metal layer will block EMI, so yes you can line the case with aluminum foil.

2) Yes. The hottest components (CPU, video, power supply) rely on fans to keep cool because the heatsinks on/in them do not even touch the case. The only part of your system that uses the case for cooling is the hard drives. You would want to direct a fan onto it for airflow.

3) The temperature sensor is probably reading incorrectly, indeed those temperatures sound too LOW for a modern processor unless you were using water cooling (or a super air cooler like the XP-120 with a high RPM fan). Many of us around here like to trade temperature information in Celcius for various reasons, such as it being internationally used (as is the internet) and the fact that CPU manufacturers give temperature guidelines in Celcius. Doing some guesstimath in my head, your CPU is running in the upper 30's Celcius, which is running nice and cool. Remember that unless you use abnormal cooling, your CPU will never run cooler than room temperature. Indeed Intel and AMD are confident that their CPUs will keep running properly up to 60ºC and not die up to 90ºC. Note that these numbers are moving targets because different cores have different numbers. For instance, the highest MHz models may have lower numbers. Also, the Prescott CPUs are rated higher (probably because they run so hot).

4) Use your imagination. If you were going to build a case that basically is designed just like a store bought case, may as well save yourself the trouble and just go buy a case. If you build a case, think on what aspects you want to improve. For instance, I wouldn't design in as many drive bays because, face it, most people don't have that many drives. I would probably do a compartmentalized pseudo-BTX design, kinda like the Silverstone Temjin/Antec P180/Lian Li V1000 designs, except I wouldn't put the power supply opposite the CPU area.

5) How that idea works is due to the notion that sound waves travel in a straight line. If you do not have a line of sight with the source of the noise, you don't hear it. Well, actually you do because sound gets reflected and the "harder" a surface is the more sound it will reflect. Ideally, you have some kind of chamber covering the fan so that there is no direct path for sound to travel, then line the chamber walls with sound dampening foam (not vibration dampening stuff). Here's an idea for you that I've seen in rackmount cases. In those circumstances the idea is airflow without taking up front/rear panel space, however it can work for noise attenuation. The idea is placing fans inside a case. Think of it this way, take your normal tower case and imagine a line going from top to bottom, between the motherboard and drive bays. Then, imagine this line being filled with 120mm low noise/RPM fans. Perhaps it will work, perhaps not, but the idea has merit.
 
Consider doing as follows. I have a penchant for resurrecting old ATX tower cases from highly-rated OEM machines produced in the late 1990s.

With a pop-rivet tool (probably less than $15 at an auto parts store and a few dollars for an assortment of pop-rivets), a dremel tool and/or a Radio-Shack "nibbler," you can modify that case for fan deployment. You can add, modify or rearrange drive cages (from other old computers -- or buy some aluminum flashing at Home Depot and build your own -- again using pop-rivets).

That will keep you in the FCC "A" or "B" rating range for EMF and related problems.

By drilling a few extra holes in the case, selecting the appropriate round-head wood screws and doing your "fine cabinet-work" to fit wooden panels (rosewood, teak, honduran mahogany, or even oak would be nice) to the case exterior dimensions, you can have the aesthetic of fine furniture together with either cooling or FCC-rating of metal.

For dampening the noise inside the case, strategically placed "Spire" rubber-noise-deadening pads are one way to go. If you use foam-art-board to duct the motherboard for more effective cooling of the CPU, Mosfets, Northbridge and memory modules, the foam board itself will significantly do what the "Spire" will do, and cost significantly less.

 
1. Devices using frequencies that are coming out of the case will be messed up. Whether it is a serious problem is luck of the draw. For me, I can use cordless phones just fine, but it is a known problem in many situations. Don't expect FM/AM radios or various cordless digital communications devices to work (but they might...). Sticking some metal panels in there won't help unless you take the design and implementation time to get them just right. Chances are you'll just have to live with any EMI coming out.

2. Yes, it's no big deal. HDDs use the case for cooling, but nothing else. Have HDDs near cooling fans, or set up w/ heatsinks (Zalman and Nexus have nice ones). As far as custom mounting goes, a HDD-cooler will offer a nice way to cool and decouple drives--you can use the fins or heatpipes to hang it 🙂.

3. The P4 should throttle. If it is reading from an internal diode (check and see if you can get an internal reading w/ the mobo you have), anything under 70C is probably fine. If it's not internal, then there is no absolute temp watch for.

5. It could, but wouldn't make that much of a difference, I don't think. For one, most wood will deflect pretty well. Going with nice quiet parts and a non-PWM fan controller will have a bigger difference, unless you are also going to shell out the money for sound dampening material.

4. Go tall, not wide! See my SPCR thread for a bit more on that. Initially it may seem like wide is easy, but it ends up not being. Whether it is specifically a tower or not, have airflow going towards to the top-rear. Also, don't use plywood, under any circumstances.

If possible, get the CPU and video card ducted (plastic report covers can be creased easily, allowing for nice cheap ducts). They will need the most air moving away from them. Also, make sure everything that can be decoupled, is.
 
You can check out my version of a wood case at PC Mechanic forums. Look in case modding & then custom case work, it's on the first page. I'm working on an exhaust cooling setup for the chipset next.
 
Originally posted by: vanvock
You can check out my version of a wood case at PC Mechanic forums. Look in case modding & then custom case work, it's on the first page. I'm working on an exhaust cooling setup for the chipset next.

Or for those who are lazy like me here is the direct link.

 
I'd start with the frame from an atx case, that way you have a place to mount all the parts. From there you can do anything, make it wider and taller for water cooling, add air ducts, whatever you want. The trick is to "think out of the box".
My case sits on the floor on my left, I've been thinking of building a case with the cd, dvd, fan controller and jacks on the side instead of the front.
 
Well, I do have many ideas for my future case. In fact, I am hesitating between three main designs, two of them having a duct below the case itself where one of the front fan can take its air. Do you think isolating the hard drives in a small box with an intake and an outake fans could lead to potential problems? I'm still trying to figure out how to cool them down as well as reducing their noise (instead of suspending them). Maybe an isolated box could do the trick..?
 
Not really. There is no 'instead of' suspending them. Trust me. If you can't do grommet-mounting, suspension is the way to go. Also, they don't need isolation as far as air flow. As long as the air is not stagnant, HDDs can be cooled. Suspension really is as good as the folks at SPCR make it out to be (er, I'm now one of those people 🙂), it can just be a PITA to get done the first time.
 
Originally posted by: Cerb
the folks at SPCR make it out to be (er, I'm now one of those people 🙂)

AIIEEEEK, run away!!!

I just had a brainstorm on an idea for a silent system with custom case. There are products to make your computer quieter. The more extreme products are designed to run passively cooled, meaning there is no fan-force air cooling. I was thinking how difficult passive cooling was to pull off, then started thinking of WHY it was difficult. What if air moved over the passive heatsinks, but not by using a fan. That is possible because hot air rises, right? Well then, why doesn't it work too well in a conventional computer?

Three reasons
-First is too many obstructions in the case (I/O cards, drives in the way).
-Second is airflow not properly tuned for rising heat (most cases are designed for front to back airflow).
-Third is conventional cases are too enclosed, thus depending on fans for airflow.

In the simplest term, my idea is this: Rotate the case 90º so that it sits on the face, then cut off the top and bottom (formerly face and back) and finally raise it off the ground. What this will do is:
-First, HDDs and I/O cards such as the video card will be oriented vertically, allowing air to rise unobstructed from being heated, coincidentally working in conjunction with the designs of most passive heatpipe video card coolers.
-Second, hot air is allowed to rise straight upwards and not forced to rise diagonally and encounter horizontal obstructions.
-Third, the case itself doesn't obstruct airflow. No full case of air trying to squeeze out of one or two 80mm holes, instead the whole volume of air can rise.

Of course there are some design issues such as placement of externally accessible drives and I/O ports. One potential solution is to use a USB hub. One cable going back into the computer can hook up keyboard, mouse, printer, scanner, card readers, flash drives. Also, optical drives can be hooked up (though with a loss in performance) using external drive bays.

You heard it here first.

©2005 Michael Liu AKA Zap
 
There is also the problem of covnection being too slow. Look up, on SPCR, the math (somewhere is cases and damping, I think). For most PCs, it would take a chimney of at least several feet high, and that's the good ones.

The best way is simply going to be for most manufacturers to get low-power parts out, and bigger, badder, heatsinks. The TR XP-120 and Scythe Ninja can both do close to passive cooling of A64s. Chances are, this situation will only get better w/ Intel coming out w/ P-M based desktop chips (using a P-M poses issues w/ heatsinks, ATM), and Asus having a nice heatpipe VGA cooler.
 
Vanvock --

For a finishing touch, dress it up like a pre-WWII radio, then find some way to make one of the radio-knobs play a WAV file of a fireside chat from Franklin Roosevelt.
 
Originally posted by: BonzaiDuck
Vanvock --

For a finishing touch, dress it up like a pre-WWII radio, then find some way to make one of the radio-knobs play a WAV file of a fireside chat from Franklin Roosevelt.

🙂
 
Back
Top