A "silent" case design / mod idea (with pictures)

[sipitai]

Hi everyone,

I'm planning on making a few modifications to an old midi tower ATX case, in order to build a new "silent" PC (using all passive components and two 120mm case fans).

My plan so far is as follows:

• Create a metal panel that separates the case into an upper chamber (where the PSU, DVD and FD are located) and a lower chamber (where the MB, CPU, RAM, NB, GPU, and HD are located).
• Cut two 120mm holes in the separator panel, and mount two 120mm fans, blowing upwards, on the "lower chamber" side.
• Cut a small hole in the separator panel for the PSU, DVD and FD cables to pass through.
• Cut two 120mm intake holes in the bottom of the case, directly below the two holes in the separator panel.
• Replace the standard feet with 50mm high wheels.
• Seal all the holes, gaps, etc. in the case where air can flow in or out, or where air can flow between the upper and lower chambers, including the gaps around the DVD and FD, the cables passing through the separator panel, and the front bezel cables passing through the front of the case, but excluding (obviously) the newly cut 120mm holes, and the hole where the PSU is mounted.

Here is a diagram to help visualise what I'm on about:

*** Please note that there is an updated version of this diagram further down the page ***



The list of components I'm planning on using is as follows:

*** Please note that there is an updated version of this component list further down the page ***

• PSU : Antec Phantom 350W (passive)
  MB : Gigabyte GA-K8NPRO-SLI (passive NB)
  CPU : AMD Athlon64 3000+ S939 (Venice) + Scythe NCU-2005 CPU cooler (passive)
  GPU : Gigabyte GV-NX66T128VP Nvidia GeForce 6600GT (passive)
  HD : Samsung SpinPoint P120 Series - 250GB / SATA 3.0Gb/s / 7200rpm
  Fans : 2 x Nexus 120mm Real Silent case fans

My objectives are as follows:

• Optimise airflow / cooling:
  
  The idea being that by modifying the case, air would enter the case via the intake holes at the bottom, flow through the lower chamber (negative pressure), pass through from the lower chamber to the upper chamber via the fans, flow through the upper chamber (positive pressure) and exit the case via the PSU, the air would be moved through the case by two 120mm fans in parallel (approx total of 74CFM), and the PSU would be located in a separate chamber. I am assuming this will result in better airflow / cooling and therefore reduce the temperatures of the MB, CPU, RAM, NB, GPU and HD.

• Reduce noise:
  
  The idea being that by modifying the case, the two fans will be located in the middle of the case, instead of on the edges, and all unnecessary holes, gaps, etc. in the case, especially those at the front, will be sealed. I am assuming this will reduce the amount of noise that can escape from the case.

But having said all that, I'm not entirely sure if I'm on the right track here or not...

Here are the questions I have so far:

• Are the assumptions I've made above in regards to optimising airflow / cooling and reducing noise correct?
• Will it actually work, i.e. will all of the components stay within a reasonable temperature range?
• Will it run as silently as possible given the selected components (i.e. a 3000+ CPU and a 6600GT graphics card)?
• If you answered no to any of those questions, what would you recommend I do differently?
• Can I get away with less airflow, or will I require more airflow?
• Would I be able to get away with even hotter components (e.g. a 3500+ CPU and a 7800GT graphics card)?
• What size, shape, etc. should I make the intake holes in the bottom of the case? By default I'm just planning to cut two 120mm holes (to mirror the holes for the fans), but I'm open to suggestion here.
• Given the intake holes are on the bottom of the case, instead of the front, will dust be more of a problem than it would be normally?
• Seeing as the NCU-2005 weighs a reasonable amount, should I look into creating some sort of support for it (especially seeing as I plan on moving the PC around, taking it to LAN's, etc.), or would that be totally unnecessary?
• Would the whole thing work just as well without the separator panel and with the fans mounted, blowing upwards, at the bottom of the case (i.e. is my current plan needlessly complex)?

Any help, information, etc. that anyone can offer will be greatly appreciated.

Or if anyone has any further questions, please feel free to ask.

Thanks!

[datapappan]

Interesting setup, kind of echoes Bluefront's one fan PSU mod.

I have a similar setup, w 2x120 mm fans in the case bottom, fitted w filters. I tried sealing the case and have the fans push through the PSU, but cracks in the I/O-panel seemed to leak enough to steal air from the PSU, making it overheat. The VGA is "passive" though, one of the fans are blowing right at it, retrofitted w a Titan HS. CPU is fan cooled as per my sig, PSU as well.

On to your setup. I have no experience with passive VGA coolers as such, my observation is that it might benefit from a directed air flow (read: place fan under it). If you place fans on the bottom, leakage may cause the PSU to not get enough air, as above. Dust seems to make its way in to a case any way, but then filters take a lot of fan pressure making more noise. Make sure your divider is sturdy, or the fans will make it hum.

All for now. I'd say go for it, it has potential.

/datapappan

PS After checking your setup once more, I propose changing the setup to PSU w bottom 120 fan + 120 exhaust fan in the back - the PSU is quiet restrictive and you might need more air to cool the Scythe.

[Mutie]

I am only a beginner myself so take my comentary as such.

I suspect that you may be asking to much of your psu to be the fan out for all your airflow, and that may be a choke point. I geuss that depends on what kind of cfm you psu is capable of but personaly i have been trying to get my psu fan to run as slowly as possible as it is the loudest part in my machine.

My other concerns are about airflow obstruction which may or may not exist and is mostly a question of how accurate the scale of the diagram is.

Hole 1 in the bottom will definatly help with the gpu cooling.
Hole 2 looks obstructed by you hdd maybe that would improve if you mounted the hdd higher , there is also limited airflow over the top of the hdd and it may run a bit hotter, maybe mount hdd verticaly?
Fan 1 looks obstructed by the psu and Fan 2 looks obstructed by the fd, a single fan in the middle might actually turn out to be just as effective.

That having been said I like your ideas and would love to hear what you finaly decide and how it works for you.

[datapappan]

Mutie, Welcome to SPCR!

/datapappan

[justblair]

Nice diagram!!

* Are the assumptions I've made above in regards to optimising
airflow / cooling and reducing noise correct?

Not sure, the fans look like they will be impeded by the psu, FD and DVD, I think that there are better ways to doing it. Bluefronts designs are pretty good to look at for examples of well thought out air flow. Or have a look at this case

http://www.nexustek.nl/breeze500.htm

My brother has one, and I can confirm that it is a pretty dam quiet box. He has only mild silentitius, contracted from hanging about the same flat as me. Its got the same fans in it you are talking about, even at full speed (he wont let me near them to drop them down to 800rpm,)

I'd go with that design, choose a nexus power supply as well (same uber quiet fan, duct the psu to the power supply fan. Drop the fan speeds

* Will it actually work, i.e. will all of the components stay within a reasonable temperature range?

Yes I dont doubt for a second it will work.

* Will it run as silently as possible given the selected components (i.e. a 3000+ CPU and a 6600GT graphics card)?

Not sure, think you will get some wind resistance off the impeeded fan

* If you answered no to any of those questions, what would you recommend I do differently?

Yep, first things first I would have a look at some of Bluefronts creations.

* Can I get away with less airflow, or will I require more airflow?

Your using a pretty efficient processor, I dont think you need more airflow. But I think that it would be better in the way I suggested above, two fans in parrallel will move air, but a push pull arrangement from what I have read here (dont flame me if I am wrong, I dont use push pull) will produce more "torque" which is better at getting round corners or throug ninjas

* Would I be able to get away with even hotter components (e.g. a 3500+ CPU and a 7800GT graphics card)?

Probably. My brother is going the 64bit route soon, but for the meanwhile he is socket a athlon 3000xp no probs. He is coincidently using the sata spinpoints, just got them (I'm slightly jealous)

* What size, shape, etc. should I make the intake holes in the bottom of the case? By default I'm just planning to cut two 120mm holes (to mirror the holes for the fans), but I'm open to suggestion here.

I would make sure that they are roughly the same surface area as your outflow. But I would only have one. I am making a custom case at the moment, with lots of holes cut for mostly passive operation. I have cut the holes, this week though I am adding alu braces all over. I was suprised by how much a rigidity you lose when you start cutting. this makes reverb from the case an issue when you are using fans.

* Given the intake holes are on the bottom of the case, instead of the front, will dust be more of a problem than it would be normally?

Yep, The nexus has a filter in place, I can confirm that it does gather quite a bit of fluff.

* Seeing as the NCU-2005 weighs a reasonable amount, should I look into creating some sort of support for it (especially seeing as I plan on moving the PC around, taking it to LAN's, etc.), or would that be totally unnecessary?

Dont know about this, see a lot of people fretting about heavy heatsinks. I use (and my brother two) a big typhoon that is no supermodel when it comes to weight. I dont have any fears about using it, the mobo seems to take the strain without any difficulty. I have a 6cm copper heatsing on the Northbridge, a 4cm alu on the southbridge, Mobo's are held on with multiple screws that spread the weight around. If it really keeps you up at night, build the duct I suggest out of 6mm steel and brace the sink with it

* Would the whole thing work just as well without the separator panel and with the fans mounted, blowing upwards, at the bottom of the case (i.e. is my current plan needlessly complex)?

Yes it would, but not sure about running the cpu passive then, all the examples I have seen use the CPU fan to draw air through the thing. Air will otherwise take the path of least resistance.

Hope that helps...

Best advice, have a look at bluefronts designs. There is some in the how to section.

[MoJo-chan]

A bit OT, but sipitai, I'd just like to ask what program you used to make your diagram?

[ronrem]

A VERY sound concept,you may need to tweak it a bit but you pressurize the upper chamber and that air will move out via the PSU. Chech the PSE for restrictive vents,but a passive usually is designed to have minimal restriction. 2 120 mm holes (round) or equivalent (square) is ample intake area even if you add a light,low restriction filter,like a thin foam,cloth,fibre material. This being experimental,I suggest a panel with good temp monitoring and fan speed controls. Obviously if those fans spin at 1500 each,you have enough cooling,but the goal would be to see how things go with them each at 750 rpm or so.. I really like the whole Idea of sticking the fans in the middle of the box. Excellent. A bit of acoustic padding should make this nice and quiet,I assume you are hip to HDD silencing too.

More potent Vid cards do tend to be hotter. I have not seen much on the heat creation of the various Venice chips,but power consumption on the 3000,3200,3500 is listed as the same,and so the heat output of the 3500 probably is only a little more than a 3000. AMD just lowered the 3200 price-and dropped the 3000-though they can still be had. If you are thinking about a bit more power, the 3200 seems to be real close in both heat + price now.

Look at the Coolmax CR450B PSU. Why? It is less than half the price of any passive PSU and has a 140mm 3 speed fan - min speed 1000 rpm. Since you are otherwise sticking a 120 in that location, You'd get as much airflow at way less bucks.

[ronrem]

relized that the suggested Coolmax PSU only intakes through the bottom mount 140 fan-so you would need to make some other outlets for the air pushed by the other 120,such as a slot above the psu. With the 140 moving a pretty nice amount of air the other fan gets to run a bit slower.
I know Coolermaster has a 120 that stock,12v,is 720 rpm,and so I assume it can undervolt to 500 rpm. A Nexus has a higher 12v speed but they undervolt very well and may do as well at low rpm.

[MikeC]

Interesting try. It may work in a cool climate (modest ambient temp) and a modest system, but that's all. I don't think much of the Nexus Breeze design, either.

The main issue is this: Why try to pump the heat of the whole system through that PSU? (actually any PSU). There's so much impedance at the exhaust end of your setup, adequate airflow will definitely be a problem.

This is a fundamental flaw.

You'll probably have to run the fans pretty fast to keep the all the components from overheating -- and there goes any potential noise advantage of the internal fan position.

If you have to run the 2 Nexus 120s at 12V, it will be 24~25 dBA @ 1m. That's not really quiet, and the case will not reduce the noise much -- maybe 2 dB tops.

There a PC in the lab using a much more conventional setup with 2 Nexus120 fans running at something like 5.5V. One in the PSU, one on the back panel. Big pass HS -- can;t recall which one. It's an XP2500+, which is not much different from the A64-3000+. It is really inaudible -- except after you turn it off, you know there was something before. Measures ~18 dBA @ 1m. Ambient is 20~24C most of the time, and the thing is rock stable, 24/7.

Does having the fans on the inside makes them any less audible? I am not sure. Maybe. Air cavity resonance effects will probably counterbalance whatever gains you make by reducing direct sound path from the fan to your ears. But the main thing is that I don't think you can afford to run the fans slowly enough for this to be a quiet system -- except in very cool weather/ ambient.

Also, how would this compare in noise to using the bottom intake + back panel fan + the fanless PSU? This would have much less impedance in the airflow path and you would eliminate one fan. You would probably be able to run that fan slower, too.

If that seems too farfetched, how about a quiet 120mm fan PSU and a 120mm back exhaust fan? No question you could get adequate cooling for your CPU w/ that Scythe w/ such a setup Cooling-wise, it would be superior for sure.

Anyway, if you're bent on experimentation, have fun but watch the temps.

[Bluefront]

That design is similar to the setups I've been building for the last few years. The problem is the airflow through a std PSU.......too restrictive. I've eliminated that problem by creating an upper chamber the PSU sits in. Two sides of the PSU are removed, so the air can flow through the PSU easier......it still enters through a bottom 120mm hole.

One-fan computer

D-Box

AMS

The AMS is slightly different in that the PSU does not handle any of the CPU heat. I'm still using the same design though, with all air blowing upward and exiting at a back/upper-rear opening. This last case uses a P4 3.2 over-clocked slightly....still never goes over 48C no matter what.

[MoJo-chan]

I suggest you take a look at http://www.geocities.co.jp/SiliconValley-Oakland/5070/ (there is an English section, but the Japanese section is sometimes more up to date and in depth if you can read any of it).

This guy is the master of passive cooling. In particular, look at how he moves hot air around the case with cardboard covers, for example moving heat from the GPU out the back and away from the CPU. He also tends to replace stock heatsinks with larger ones. Putting the HDs at the top is another good idea.

One thing he doesn't do, which might be worth considering, is using the case itself as a heatsink. Like the Zalman fanless case design, some heatpipes could make use of that large surface area, or at least external heatsinks for cooling.

[sipitai]

Hi everyone,

Thanks for all the helpful comments, suggestions, etc. so far.

It seems the biggest problem with my plan at the moment is that having the PSU as the only exhaust vent will result in too much airflow impedance, which in turn will make it harder to "silently" cool the components. Although I've given this all a bit more thought and I believe this problem can be solved.

My proposed solution:

• Cut an additional exhaust vent in the top rear of the case that opens up the area around the PSU. The vent would be cut directly above and to either side of the PSU and would measure 140mm wide x 80mm high + 180mm deep + 80mm high.
• Cut a single large intake vent, instead of two smaller ones, measuring 260mm wide x 120mm deep.

Here are some diagrams to help visualise what I'm on about:



Its worth noting that according to my calculations the total area (in mm²) of this additional exhaust vent, excluding the area that would be impeded by the PSU, would roughly match the total area of the intake vent, as well as the total area of the two 120mm fans.

Its also worth noting that the vents will be covered with a low restriction grill. Most likely something along the lines of Modder's Mesh. I have increased the total area of the vents to compensate for this.

I am assuming here that equalising the total area of the intake and exhaust vents will remove the airflow impedance and thereby solve the problem. But is this assumption correct, or have I got it wrong?

Here are a few more questions:

• Will this new plan solve all, or at least most, of the airflow impedance / cooling problems mentioned in the posts so far (especially those pointed out by MikeC)?
• Will this new plan have a detrimental effect on the PSU temperature?
• Should the total area (in mm²) of the exhaust vents be larger, smaller, or equal to that of the intake vents?
• Should the total area (in mm²) of the intake and exhaust vents be larger, smaller, or equal to that of the fans?
• Does anyone know, in millimetres, how narrow a gap between two surfaces can be before airflow begins to become impeded?
• Even with the new plan, would a more conventional "two fan" set up (e.g. standard case, PSU fan, and rear case fan, etc.) still be a better solution?

Again, any help, information, etc. that anyone can offer will be greatly appreciated.

Thanks!

P.S. I'm not overly bent on experimentation, I just want to try and set up a PC that is going to run as silently as possible given the components I've selected (which I'm hoping my current plan will achieve). I'm also definitely not located in a cool climate (around 30 - 40°C in summer), so I want to make doubly sure that I don't stuff the whole process up

P.P.S. I made the diagrams using Photoshop, and they are in a scale of 1 pixel : 1mm.

[MoJo-chan]

Ah, Photoshop, okay... very neat

As you your case, I'd consider having all the vents at the top of the case, and no vent at the back. If you replace the entire top of the case with a grill of some kind (wire mesh should be fine if it's not too dusty) and then replace most of the PSU case with grills too, air would be able to simply enter at the bottom of the case, heat up and rise out of the top. Negative pressure would bring air in at the bottom, and the fans would help move things along too.

Stick a shroud over one fan and the CPU heatsink to force some more air over it. I'd also consider moving the HD to the top of the case where it won't be releasing heat around the motherboard.

Also, consider where fan 2 is. If you put it too close to the FD and DVD-ROM, it won't work too well and might make more noise. A better position might be at the bottom of the case, under the GPU, or under the NB and RAM.

A cardboard divider between the GPU and CPU might be worth trying to. No point having heat from the GPU rise up to the CPU when it could be exhaused out the back.

[sipitai]

Hi everyone,

Following on from my previous post, here is a response to all the other points, suggestions, etc. in regards to my plan so far.

• "Using a PSU with a 120mm fan and a 120mm rear case fan, instead of the current fan set up, would be a better solution."
  I would disagree. My reasoning here is that I cant see any advantage to using this set up.
  
  In fact I see the following disadvantages:
  
  • I cant select the model of fan used in the PSU without voiding the warranty.
  • I cant use foam, elastic, etc. to mount the fan used in the PSU.
  • Heat from the PSU has the potential to increase the case temperature.
  • Given the rear case fan is mounted on the edge of the case, there is a direct path for noise to escape (although having said that I'm not sure if this would noticeably increase noise levels or not).
  • The airflow path (bottom to top / rear vs. bottom to top) appears to be slightly less optimal (although having said that I'm not sure if this would have any effect on cooling).
• "Using a single fan, or two fans in push / pull (i.e. Nexus Breeze case), would be a better solution."
  I would disagree. My reasoning here is that I would assume that in either case there just wouldn't be enough airflow to cool the components effectively.
  
  The push / pull set up would provide more torque than the current set up, but given I am setting up a low impedance case, I would assume that more torque and less airflow would be a step in the wrong direction.
  
  
• "Replacing the entire top of the case, as well as the PSU housing, with grills might help with airflow / cooling."
  Its definitely an interesting idea, and it definitely would allow for more unrestricted airflow. Although at the same time I would assume that there would be little or no airflow through or around the PSU, and having the entire top of the case open would create a direct path for noise to escape.
  
  Its also worth noting that I would be voiding the PSU warranty.
  
  
• "Airflow into the upper chamber will be obstructed by the PSU / FD."
  I agree. But will this be a problem?
  
  Given the fans are just "pressurising" the upper chamber in order to force air out via the PSU / vents, does it matter that there are a few obstructions in the way?
  
  I would assume not, but I'm really just guessing.
  
  Having said that, its also worth noting that I may remove the FD from the plan.
  
  
• "Using a PSU with a 120mm fan, instead of the Antec Phantom 350W (passive), would be a better solution."
  I would disagree. My reasoning here is that I would assume, given all of the air exhausted from the case either flows through or around the PSU, an actively cooled PSU shouldn't be necessary.
  
  I would prefer to avoid adding any extra fans to the plan unless I have to, given that adding extra fans also adds extra noise.
  
  
• "Some directed airflow onto the CPU / GPU might be a good idea."
  I agree. But is this 100% necessary?
  
  Again, I would prefer to avoid adding any extra fans to the plan unless I have to, and I'm assuming here that the current fan set up should be capable of adequately cooling the components.
  
  
• "Ducting the CPU heatsink to one of the fans, or placing a divider between the GPU and CPU might help improve case temperatures."
  I agree. Although I will wait and see what the temperatures are like to begin with, and then take these steps if necessary.
  
  
• "The hard drive would be easier to cool, and would no longer obstruct airflow, if mounted vertically."
  I agree. Although will mounting a hard drive vertically (i.e. top to bottom, I'd have the back of the drive facing up and the front of the drive facing down) cause any performance, stability, etc. problems?
  
  
• "Moving the hard drive closer to the top of the case might be a good idea."
  I figure out of everything in the case, the hard drive is the component that I want to keep the coolest. So I'd rather have it give off some heat that effects the other components, than the other way around.
  
  
• "Taking steps to silence the HD might be a good idea."
  I agree. I will be either suspending the HD with elastic, or placing a thick layer of foam between the HD and the case.
  
  
• "Reducing the speed of the fans would be a good idea."
  I agree. I'm actually planning on using some sort of fan controller so that they can run at low speeds most of the time, but speed up when they need to. Although I still need to do more research on this one.
  
  
• "Using intake filters might be a good idea."
  I have given this a bit of thought, but I think this time round I will leave filters out of the equation. Only because I think using them takes the whole project to the next level of difficulty, and I want to try and play it safe in terms of airflow.
  
  
• "Using acoustic padding might be a good idea."
  I'll be honest and say I have no idea on this one. I figure it could either make a noticeable difference, or make almost no difference at all.
  
  What does everyone else think?
  
  
• "I should take a look at Bluefront's case designs."
  I definitely agree with that, and I have.
  
  His ideas, along with a whole load of others in these forums, helped me bring my plan to the stage it is currently at.
  
  Although if you still think my plan needs a bit more Bluefront influence, give me the details and I'll take another look.

Having said all that, its worth noting these are all just assumptions on my behalf. So if anything I've said here is incorrect, please let me know.

Again, any help, information, etc. that anyone can offer will be greatly appreciated.

Thanks!

[MikeC]

Its worth noting that according to my calculations the total area (in mm²) of this additional exhaust vent, excluding the area that would be impeded by the PSU, would roughly match the total area of the two intake vents, as well as the total area of the two 120mm fans.

I am assuming here that equalising the total area of the intake and exhaust vents will remove the airflow impedance and thereby solve the problem. But is this assumption correct, or have I got it wrong?

Here are a few more questions:

• Will this new plan solve all, or at least most, of the airflow impedance / cooling problems mentioned in the posts so far (especially those pointed out by MikeC)?
• Will this new plan have a detrimental effect on the PSU temperature?
• Should the total area (in mm²) of the exhaust vents be larger, smaller, or equal to that of the intake vents?
• Should the total area (in mm²) of the intake and exhaust vents be larger, smaller, or equal to that of the fans?
• Does anyone know, in millimetres, how narrow a gap between two surfaces can be before airflow begins to become impeded?
• Even with the new plan, would a more conventional "two fan" set up (e.g. standard case, PSU fan, and rear case fan, etc.) still be a better solution?

Again, any help, information, etc. that anyone can offer will be greatly appreciated.

1. No, it should improve PSU temp -- more airflow overall through/around it; the external casing of the Pahntom IS the heatsink, really.
2. Equal is about right; it is what I always shoot for.
3. I suggest equal size.
4. Generally, smaller than the area of the blades.
5. Hard to say. There is so much in the details; execution will count for a lot. IE, a better idea executed more poorly is often not better than a poorer idea executed well. The main thing is that the conventional method is easy to try, and cheaper. You still need to take care of basics in either case: Low restriction vents, clear airflow path, sturdy/damped fixings for fans/HDDs.

A couple of specific caveat:
a. Avoid putting the fan(S) too close to the PSU or the drives in the upper chamber. Your pic shows the fans half covered -- this will not be good for either airflow or turbulence noise.
b. Dust could be a problem if you're not a good housekeeper (or don't have one), have pets, or live in a dry area. But you're right about filters being a pain; even if you get them as nicely done as Bluefront, you still have to clean them out regularly anyway. I have a very furry cat and find that a vacuum at the vents and insides about once every 2 months good enough. There's usually not more than a fairly light dusting of cat fur.

[MoJo-chan]

A few quick comments:

[1] I cant select the model of fan used in the PSU without voiding the warranty.

[2] "The hard drive would be easier to cool, and would no longer obstruct airflow, if mounted vertically."
I agree. Although will mounting a hard drive vertically (i.e. top to bottom, I'd have the back of the drive facing up and the front of the drive facing down) cause any performance, stability, etc. problems?

[3] "Reducing the speed of the fans would be a good idea"
I agree. I'm actually planning on using some sort of fan controller so that they can run at low speeds most of the time, but speed up when they need to. Although I still need to do more research on this one.

1) I have a Nexus 4090 which can be opened without breaking any stickers etc. It makes sense to me, after all anyone serious about PCs will want to open their case and add hardware, or exchange fans, so why not in the PSU too. It also makes it much easier to clean. Having said that, if you get a fanless PSU or a Seasonic, you might not need to mod to get the noise levels you want.

2) Mounting vertically has no negative impact on reliability or performance. Many Dell and HP PCs have vertically mounted HDs.

3) I'd try building your own if you are confident enough. Commercial controllers often don't seem to go down to low enough voltages. For example, I run Panaflo 92mm fans at 4.5V. Some start reliably at under 4V.

[Bluefront]

Here's the deal.....when you are into modding computers, get the word "warranty" out of your head. It is too restrictive. Make sure the thing works, then take it apart. I test PSUs with a little test unit before I use them. After that I'm on my own.....

IMHO....avoid exhaust openings on top of the case. Too noisy. Intakes and exhausts should be on the bottom or rear only for maximum silence.

I filter all my setups....see those photos. The AMS has the easiest setup, using two identical filters. Each just drops out for cleaning.....no sweat at all, or taking panels off either.

Take a look at this setup. I moved the PSU to the bottom in a wood case, and use the original space where the PSU sat, for an intake/filter housing. You could modify the setup and use the top space for an exhaust fan....put the intake somewhere else.

Here's some photos of the PSU case

There is a lot of room for new case ideas......I'm not waiting for Antec to do it.

[sipitai]

Hi everyone,

Thanks again for all the helpful comments, suggestions, etc.

In response to the most recent posts, and after giving this all a bit more thought, here are a few more updates to my plan:

• Modify the separator panel so that the area where the fans are mounted is 15mm lower than the rest of the panel (to increase the distance between the fans and the components in the upper chamber). This would increase the gap from 10mm to 25mm.
• Mount HD vertically instead of horizontally (my current plan is to have the front facing up, and the bottom facing down).
• Remove the FD (to make more room in the upper chamber and simply the construction process).

Its also worth noting that I will be mounting the HD, as well as the fans, using foam and elastic (the idea being that the a layer of foam would be placed in between the component and the case, and elastic would be used to mount / tie the component to the case).

Here is an updated version of the original diagram to help visualise what I'm on about:



The current list of components I'm planning on using is as follows:

• PSU : Antec Phantom 350W (passive)
  MB : Gigabyte GA-K8NPRO-SLI (passive NB)
  CPU : AMD Athlon64 3200+ S939 (Venice) + Scythe NCU-2005 CPU cooler (passive)
  TIM: Arctic Silver 5 Thermal Compound
  RAM: Corsair Value Select PC3200 1Gb Kit (VS1GBKIT400)
  GPU : Gigabyte GV-NX66T128VP Nvidia GeForce 6600GT (passive)
  HD : Samsung SpinPoint P120 Series - 250GB / SATA 3.0Gb/s / 7200rpm
  DVD: BenQ DW1640
  Fans : 2 x Nexus 120mm Real Silent case fans

I will also be adding a Fan Controller to list when I get a chance to do a bit more research.

My only remaining question is...

Am I ready to make a start, or are there still some problems that need to be dealt with?

Thanks!

[MoJo-chan]

HD : Samsung SpinPoint P120 Series - 250GB / SATA 3.0Gb/s / 7200rpm

Fans : 2 x Nexus 120mm Real Silent case fans

Finding Samsungs with NIDECs is hard. Let us know if you find a good source!

As for the fans, are the Nexus ones rebranded Yate Loon? If not, you can get orange Yate Loons from eBay quite cheaply. They are second to none for noise IMHO.

[iakovl]

sipitai, if u don't mind
i'll use your idea for mine own mod. k?

[sipitai]

Ok,

I've modified the case, bought the parts and put it all together. Although I still need to fine tune it and do a bit testing.

As soon as I have I'll post the details, along with some photos, in the general gallery forum.

Finding Samsungs with NIDECs is hard. Let us know if you find a good source!

At the moment, any retailer that gets their Samsung SpinPoint's from Altech (distributor) should have the drives with the Nidec bearings. At least in Perth, Australia that is.

I bought a 250GB and a 200GB (for a friend) and they both had Nidec's.

sipitai, if u don't mind
i'll use your idea for mine own mod. k?

Go for it.

The only thing I ask in return is that you post back to let me know how it all goes

In regards to my design, its worth noting that if I wasn't modifying an existing case, I would increase the height of the upper chamber and move the exhaust vent from the rear top / sides to the upper back of the case (in a circle / square around the PSU).

[Qwertyiopisme]

Do note that I have not read all of this thread, only the most recent entries

Wouldn't it be more effective to place the fans at the very bottom, for two reasons.

For the first you'll have turbulent airflow though the case, which is *very* good to have, the way you have it set up now most of the air wont bother going through a heatsink, it will just go past, whereas if you have the fans at the bottom they'll stir the air around quite a bit more.

Secondly you'll be able to remove the bottleneck in the PSU/DVD area.

The only downside to this would be that more sound would leak out, however I'm not sure if the lower voltage you would be able to run your fans at would offset this factor.

EDIT: and now that I look at your image a bit more, I seem to feel that your harddrive wont get enough cooling. Maybe placing it more in the direct path of the air would be good?

[andywww]

Mind posting a link to your project post?

[serenityNOW]

HD : Samsung SpinPoint P120 Series - 250GB / SATA 3.0Gb/s / 7200rpm

Finding Samsungs with NIDECs is hard. Let us know if you find a good source!

How can one tell if it is a NIDEC?

[jaganath]

How can one tell if it is a NIDEC?

It is printed on the rotor hub of the hard drive:

Link to image of SP0802N with Nidec stamp

[serenityNOW]

Thanks!

I am also thinking to change case feet, will those work fine for a silent case?

Is there a recommended type?

[jaganath]

I am also thinking to change case feet, will those work fine for a silent case?

Is there a recommended type?

That site that you linked appears to sell castor wheels. If you want to prevent vibration being transmitted from the case to the floor (or whatever surface your PC is resting on) you need the case feet to be something soft, like silicone rubber or even foam (it would need quite a lot of foam though). Generally silicone rubber is best.

[serenityNOW]

Not sure I know what the foam wheels are..., but conserning silicone rubber, do you mean like dose used on roller blades?

[jaganath]

conserning silicone rubber, do you mean like dose used on roller blades?

Yes; ideally they would be even softer than roller blade wheels but it might be difficult to find such soft wheels. Essentially there must be some kind of shock absorbing material between the floor and the PC, that's all.