Sign in with
Sign up | Sign in
Your question

Todays Hot Mod: Reverse PSU Cooling

Last response: in Systems
Share
September 4, 2007 6:57:06 PM

Some ATX power supplies are designed to pull fresh air from outside the case,draw it across the hot components of the PSU and exhaust it inside the case, this can and most often does increase case temperatures and make CPU cooling more difficult.
FYI: This mod requires voiding the warranty on the PSU....but if your a mod squad member like me then you could really care less.
Disassemble the case to the point that you have access to the fan/s, simply remove, flip over and reattach the fan/s then reassemble the PSU case and reattach the power leads.
Now the PSU will be working in harmony with the rest of the system and drawing cool air from the lower front across the motherboard then up and out the back of the case.
Before performing this mod please ensure that your PSU fan is indeed blowing air into the case and not drawing air out.
Good Luck n Happy Trails..

Side note: Most modern PSU's are designed to exhaust hot air out of the case which is correct.

Folding@Home
September 4, 2007 7:29:23 PM

Umm... every PSU I've seen exhausts the hot air out the back of the case and not into the case. As far as I'm aware it has always been the case where the PSU intakes from the case and exhausts out the back.
I suppose its always possible to get a PSU that was put together wrong and has the fan installed the wrong way, but that should be a very very rare occurance for even the cheapest PSU manufacturers.
September 4, 2007 7:33:07 PM

mad-dog said:
Most ATX power supplies are designed to pull fresh air from outside the case,draw it across the hot components of the PSU and exhaust it inside the case, this can and most often does increase case temperatures and make CPU cooling more difficult.


You got it backwards...wow, you haven't been reading much THG then have you? Most full-sized power supplies pull warm air out of the case, across the PSU components, and expell it out the exhaust. That's why most cases have the power supply mounted above the CPU, to pull hot air away from the CPU. That's also why ATX cases normally have the PSU mounted "upside down" with the lid on the bottom, so that a lid-mounted fan can pull hot air away from the CPU.
Related resources
September 4, 2007 7:36:53 PM

erloas said:
Umm... every PSU I've seen exhausts the hot air out the back of the case and not into the case.


You're right, most do. And that's one more reason why cases with the power supply on the bottom are a bad idea: Power supplies pull out more heat than they put in, and some systems have successfully used the power supply fan as the only exhaust fan.
September 4, 2007 7:37:04 PM

Definitely backwards. (Mad dog has more posts than me???)
September 4, 2007 7:42:23 PM

sirrobin4ever said:
Definitely backwards. (Mad dog has more posts than me???)


Yah, maybe i should edit slightly to reflect newer PSU technology

Folding@Home
September 4, 2007 7:47:14 PM

Most of my AT power supplies also pulled hot air out of the case, and AT predates ATX.
September 4, 2007 8:30:29 PM

Crashman said:
You're right, most do. And that's one more reason why cases with the power supply on the bottom are a bad idea: Power supplies pull out more heat than they put in, and some systems have successfully used the power supply fan as the only exhaust fan.


Sorry Crashman, but if you take a look at most of the high end full tower cases, the PSU is mounted on the bottom of the case. Remember, heat rises. So, it would only make sense that the power supply mounted on the bottom would pull in cooler air than at the top. Thus, cooling the PSU's inner components more efficiently. Take my case for example, the Silverstone TJ09. The PSU is mounted at the bottom. At the top of the case you have two 120mm exhaust fans, and one intake fan directly behind the hard drive cages, and one fan in between the hard drive cages to cool the hard drives. Tell me again why you would want the PSU to pull out already hot air over already hot components?
September 4, 2007 8:35:12 PM

Crashman said:
You're right, most do. And that's one more reason why cases with the power supply on the bottom are a bad idea: Power supplies pull out more heat than they put in, and some systems have successfully used the power supply fan as the only exhaust fan.


The only way that makes sense is if the PSU is the only exhaust in the system, and I would only do that in SFF cases, or HTPC's that put out barely any heat anyway.
September 5, 2007 2:32:54 AM

korbin44 said:
Remember, heat rises. So, it would only make sense that the power supply mounted on the bottom would pull in cooler air than at the top. Thus, cooling the PSU's inner components more efficiently.


Right, and the PSU is supposed to be more rugged than the other components in a system. The main reason companies now make cases with the power supply on the bottom is because their customers are stupid.

I watched the whole thing unfold, from the early customers believing claims that a system would run cooler with the power supply at the bottom, to Cooler Master adding this to its latest design. This is market driven, not engineering driven.

September 5, 2007 2:39:48 AM

korbin44 said:
The only way that makes sense is if the PSU is the only exhaust in the system, and I would only do that in SFF cases, or HTPC's that put out barely any heat anyway.


It worked fine in a bunch of systems before P4 came out. Works fine with A64 singles, plus many X2's and stock-speed Core 2 Duos so long as the graphics card doesn't run too hot and the PSU fan is 120mm.
September 5, 2007 2:45:16 AM

Crashman said:
Right, and the PSU is supposed to be more rugged than the other components in a system. The main reason companies now make cases with the power supply on the bottom is because their customers are stupid.

I watched the whole thing unfold, from the early customers believing claims that a system would run cooler with the power supply at the bottom, to Cooler Master adding this to its latest design. This is market driven, not engineering driven.



Okay, so by your account, all those engineers who are paid to make better products are stupid and don't know what they're doing. Don't be ridiculous, crashman. It's a matter of heat. Once again, heat rises. Heat also degrades your components over time. I want my system to maintain its current performance for a long time. By using the PSU to draw out all the heat in your case, your are destroying the PSU components much more quickly. It only makes sense to have the PSU pull in cooler air to COOL the components, no matter how rugged the PSU is supposed to be. Obviously you are not an enthusiast, or else you wouldn't be talking nonsense.
September 5, 2007 3:43:18 AM

korbin44 said:
Okay, so by your account, all those engineers who are paid to make better products are stupid and don't know what they're doing. Don't be ridiculous, crashman.


I'm not being rediculous and the "engineers" are very smart: They build what people buy. Nothing else matters but selling a product. And BTW, I seriously doubt that many of these cases are "engineered" anyway, they're more likelky designed by mechanical designers like me, who are paid to make whatever the customer demands.

Power supplies are supposed to be made to take the heat. We were using the original ATX design with hot P4's and the good power units lived for many years.
September 5, 2007 4:25:34 AM

It's all my fault, I'll take full irresponsibility..

Folding@Home
September 5, 2007 3:39:38 PM

Honestly Crashman, it only makes sense from a design perspective to put the PSU at the bottom of the case. I have built many computers with many different cases. My Silverstone TJ09 has the best cooling solution of any case I've ever used. One of the reasons is that the PSU is on the bottom. Besides that, I have a PC Power and Cooling Silencer 750 that powers an oc'd QX6700, 4 hard drives, 2 video cards and a 226 watt peltier. Now, why would I want more heat to be pulled in to the PSU with all that hardware already heating up the internals of the power supply? I wouldn't. That's why I chose the case I have. Look, if I had a choice, which I do, to have my PSU cooled by hot air or cool air, I would choose the cool air. Who wouldn't? Obviously you wouldn't, or we would not be having this conversation. But hey, you can continue to 'heat' your power supply while I 'cool' mine. To each his own, right?
a b B Homebuilt system
a c 121 ) Power supply
September 5, 2007 4:21:39 PM

IMHO, a top-mounted PSU is better able to assist with general case cooling, which is why it is usually there. OTOH, in better cases that provide more than enough cooling without that assist, it would be slightly better for the PSU to mount it at the bottom.
It seems to me that in a case with good airflow, the slightly warmer air going through a top-mounted PSU is still going to be sufficient to cool it, and the effect on its lifespan will be insignificant. All bets are off if the case airflow is poor and/or the PSU is a $15 special.
September 5, 2007 4:48:10 PM

Sure, the difference can be very small. But, when dealing with high end systems and extreme overclocking, heat is a killer. Any reduction in the amount of heat moved out of the system, without 'heating up' any additional hardware in the process is worthwhile. Besides all of that, most fans inside a PSU are only 80mm. A case with a bottom mounted PSU and one or two 120mm exhaust ports on the top and/or back of the case will move out more air, and be much quieter than a noisier 80mm PSU fan. It just doesn't make sense to make the PSU do more work than it should.

On the lifespan issue, when dealing with overclocking you run the risk of shortening the life of your hardware considerably. I don't want to take that chance. So any little tweak or improvement I can do to eliminate heat, I will do just that.
September 5, 2007 5:10:56 PM

korbin44 said:
Sure, the difference can be very small. But, when dealing with high end systems and extreme overclocking, heat is a killer. Any reduction in the amount of heat moved out of the system, without 'heating up' any additional hardware in the process is worthwhile. Besides all of that, most fans inside a PSU are only 80mm. A case with a bottom mounted PSU and one or two 120mm exhaust ports on the top and/or back of the case will move out more air, and be much quieter than a noisier 80mm PSU fan. It just doesn't make sense to make the PSU do more work than it should.

On the lifespan issue, when dealing with overclocking you run the risk of shortening the life of your hardware considerably. I don't want to take that chance. So any little tweak or improvement I can do to eliminate heat, I will do just that.

Amen
a b B Homebuilt system
a c 121 ) Power supply
September 5, 2007 5:47:27 PM

Well, someone selling a $200 case is probably betting that the buyer is more likely to be a heat-stressed overclocker than a someone building a business machine.
September 5, 2007 7:17:48 PM

Crashman said:
I'm not being rediculous and the "engineers" are very smart: They build what people buy. Nothing else matters but selling a product. And BTW, I seriously doubt that many of these cases are "engineered" anyway, they're more likelky designed by mechanical designers like me, who are paid to make whatever the customer demands.

Power supplies are supposed to be made to take the heat. We were using the original ATX design with hot P4's and the good power units lived for many years.

RE: Engineers...
I work for a major food manufacturer as a maintenance supervisor. In my opinion, here's how it works: The engineers design something and say: "This will work". Now technically, it SHOULD work, but because of all the practical factors involved with installation or from outside factors, 90% of the time it DOES NOT WORK AS IT'S DESIGNED. Everything that we have "engineered" by engineers gets re-engineered on the floor over and over again by mechanics or our process electonic technicians until it finally is working the way it was supposed to have worked when the engineers designed it. Engineers design it, we (the little people without advanced scientific degrees) make it work.

I think that you all are arguing over something that is not really understood by most of you. The PSU is not made up of fine transitorized circuitry like your CPU. It's made up of more rugged electronic components that do just fine under moderate to high heat conditions. Heat is the enemy of smaller, more micronized IC circuitry as it applies inside a computer case. The CPU, GPU, and Northbridge chips all have 90, 65, or 45nm transistors, making them more prone to failure in high heat conditions. PSU's are rugged devices made to withstand much higher heat thresholds than the more delicate and expensive components of your system. The only downside to passing warmer air over the PSU is that you might lose some efficiency because of the increased thermal load, but relatively, you're not risking burning up your PSU any faster.
September 5, 2007 7:34:02 PM

If it ain't broke, don't fix it. The (first?) ATX spec did call for PSU that intake through the back and exhaust into the system but most manufacturers didn't adhere to that because they knew it was a bad idea, so they stuck with what had proven to work in prior AT designs.

If you happened to come across such an ancient ATX PSU today, it's time to retire it due to old age, no point in flipping the fan. The difference in parts locations to offset some idea of heat rising and causing a chimney effect is minimal, because this is not a passively cooled system. The difference is trival so long as the case has the proper placement of air intakes such that this faster moving stream flows past the hotter parts.

PSU in top or bottom doesn't matter much, if in the bottom it will throttle down the fan some due to being cooler, which allows it to increase in heat slightly. At the same time, moving the PSu exhaust from the hotter zone towards middle and top means the upper rear fans have to move more air themselves, offsetting the potential noise reduction of the slower spinning PSU fan. Either config can work fine, and neither config has enough benefit to place much weight upon instead of the other merits of any particular case.
a b ) Power supply
September 5, 2007 8:37:05 PM

I'm not sure where people get the idea that the ATX spec called for the PSU fan to be an intake fan. Here is the ATX PSU Design Guide:

http://www.formfactors.org/developer%5Cspecs%5CATX12V%2...

If you look at section 4.3 which deals with fan direction/placement it says that in general a rear mounted exhaust fan is used.

The reason this is, is original ATX design specs called for air intake at the bottom front, exhaust at the top back. This spec was originally created before the advent of any other additional cooling (case fans) were used. So the PSU fan was responsible for ejecting all the warm air from the case. Later it became popular (as CPU and graphics cards became hotter) to have a fan in the lower front to assist in drawing cool air in. Of course as extra cooling was needed we seen the advent of side door fans and rear panel fans. This leads to why case makers felt that it was possible to move the PSU to the bottom of the case. Since nearly every new ATX case has at least one rear panel fan (usually 120mm), the PSU was no longer the only form of exhaust for the warm air arriving at the rear of the case. I am sure some case designer seen that it was no longer absolutely necessary to draw the warm air at the rear top of the case into and through the PSU (and thus make the PSU warmer). By placing it at the bottom the PSU gets cool air directly from the front of the case, and the CPU and GPU heat will also get a portion of the same cool air to draw the heat out through the rear panel fans. This should increase the lifespan of the PSU since it should maintain a cooler internal temperature on average when compared to a top mounted PSU. The hotter the PSU is, the shorter the lifespan of the electrolytic capacitors will have. This design change should also have minimal impact on the temperature of the rest of the case, because of the addition of rear panel fans that eject the heat from around the CPU, GPU area.

September 5, 2007 8:49:11 PM

DJ_Jumbles said:
RE: Engineers...
I work for a major food manufacturer as a maintenance supervisor. In my opinion, here's how it works: The engineers design something and say: "This will work". Now technically, it SHOULD work, but because of all the practical factors involved with installation or from outside factors, 90% of the time it DOES NOT WORK AS IT'S DESIGNED. Everything that we have "engineered" by engineers gets re-engineered on the floor over and over again by mechanics or our process electonic technicians until it finally is working the way it was supposed to have worked when the engineers designed it. Engineers design it, we (the little people without advanced scientific degrees) make it work.

I think that you all are arguing over something that is not really understood by most of you. The PSU is not made up of fine transitorized circuitry like your CPU. It's made up of more rugged electronic components that do just fine under moderate to high heat conditions. Heat is the enemy of smaller, more micronized IC circuitry as it applies inside a computer case. The CPU, GPU, and Northbridge chips all have 90, 65, or 45nm transistors, making them more prone to failure in high heat conditions. PSU's are rugged devices made to withstand much higher heat thresholds than the more delicate and expensive components of your system. The only downside to passing warmer air over the PSU is that you might lose some efficiency because of the increased thermal load, but relatively, you're not risking burning up your PSU any faster.


Actually, you can burn up your PSU faster by increasing the heat. Especially if you use a POS PSU made of inferior components. I know that PSU's use more rugged components, but that is not to say they can't be abused. In my case, I'm using a lot of the total watts of my PSU because of the components used in my setup. The PSU I use is a high end power supply, so I'm not that worried about it failing. But, I have done my homework and read tons and tons of reviews on power supplies and computer hardware in general. In a setup like mine, efficiency is key. If my PSU starts to falter and the efficiency goes up and down, I'm subjected to other hardware, like my processor, failing. Some hardware could even short circuit due to the fluctuations in my power supply. So, putting my PSU in the bottom of my case benefits me in more than just one way. I'm also protecting the efficiency of my power supply which in turn makes sure all the other hardware is getting the correct voltages.

Too many people who build their own systems ignore the power supply. Not in terms of forgetting to put it in their systems, but rather the quality of the build they are getting in the PSU they buy. The cheaper you go, the more likely you are to have it fail and/or damage your system. Besides, these days it's all about performance per watt, and total power used. To have an inefficient power supply affects your bottom line(electric bill), especially if your system is always on. So, to increase the heat of the PSU is, i.e., to increase the money you burn.
September 5, 2007 8:56:25 PM

techgeek said:
I'm not sure where people get the idea that the ATX spec called for the PSU fan to be an intake fan. Here is the ATX PSU Design Guide:

http://www.formfactors.org/developer%5Cspecs%5CATX12V%2...

If you look at section 4.3 which deals with fan direction/placement it says that in general a rear mounted exhaust fan is used.

The reason this is, is original ATX design specs called for air intake at the bottom front, exhaust at the top back. This spec was originally created before the advent of any other additional cooling (case fans) were used. So the PSU fan was responsible for ejecting all the warm air from the case. Later it became popular (as CPU and graphics cards became hotter) to have a fan in the lower front to assist in drawing cool air in. Of course as extra cooling was needed we seen the advent of side door fans and rear panel fans. This leads to why case makers felt that it was possible to move the PSU to the bottom of the case. Since nearly every new ATX case has at least one rear panel fan (usually 120mm), the PSU was no longer the only form of exhaust for the warm air arriving at the rear of the case. I am sure some case designer seen that it was no longer absolutely necessary to draw the warm air at the rear top of the case into and through the PSU (and thus make the PSU warmer). By placing it at the bottom the PSU gets cool air directly from the front of the case, and the CPU and GPU heat will also get a portion of the same cool air to draw the heat out through the rear panel fans. This should increase the lifespan of the PSU since it should maintain a cooler internal temperature on average when compared to a top mounted PSU. The hotter the PSU is, the shorter the lifespan of the electrolytic capacitors will have. This design change should also have minimal impact on the temperature of the rest of the case, because of the addition of rear panel fans that eject the heat from around the CPU, GPU area.


Nice find. I'm glad someone else can see it from my perspective and put some hard evidence behind it!
September 5, 2007 9:15:37 PM

Yes, so the power supply should be at the top to assist the exhaust fans because the internal air temperature in a well built system is only around 35C, well below the threshold at which top-end power supply companies test their units.

After all, few people are stupid enough to put a generic PSU in a $200 case.

But removing the PSU from the top removes some case exhaust flow. Because of that, many cases with the power supply mounted at the bottom have a 120mm fan at the top rear.

Because you're now using more fans, your system becomes noisier. Because the fan faces out the top of the case rather than into the underside of the power supply, its noise has a more direct pathway to your ears. So the CPU gets the same cooling effect by adding components and making the system noisier.

Then comes the cable routing nightmare, which can only be solved by designing the case with the motherboard upside-down. And upside-down heatpipes don't work right, so you're more likely to suffer chipset overheating. Etc etc etc.

I'm glad we all agree!
September 5, 2007 9:30:44 PM


I looked at cases with the power supply at the bottom and ultimately decided to get one with it mounted on top.

People keep repeating over and over that heat rises. Yeah, we all get that. We're all over the age of two. Yes, heat does rise, but the power supply itself can get pretty hot under load, even though it has a fan.

So in a case with the power supply on the bottom, down with all the dust and hair, the heat from the power supply is rising, rising, rising up to the video card, the CPU, the mainboard, the RAM, etc. Personally, I'd rather have my $140 PSU run a bit warmer at the top of the case than put it on the bottom and have it heat up my $275 video card, my $275 CPU, etc. Not only that, but maybe all the dust and hair that accumulates at the bottom of a case would be more damaging to the power supply than it running a little warmer at the top.

Just my opinion of course.
September 5, 2007 9:57:58 PM

carver_g said:
I looked at cases with the power supply at the bottom and ultimately decided to get one with it mounted on top.

People keep repeating over and over that heat rises. Yeah, we all get that. We're all over the age of two. Yes, heat does rise, but the power supply itself can get pretty hot under load, even though it has a fan.

So in a case with the power supply on the bottom, down with all the dust and hair, the heat from the power supply is rising, rising, rising up to the video card, the CPU, the mainboard, the RAM, etc. Personally, I'd rather have my $140 PSU run a bit warmer at the top of the case than put it on the bottom and have it heat up my $275 video card, my $275 CPU, etc. Not only that, but maybe all the dust and hair that accumulates at the bottom of a case would be more damaging to the power supply than it running a little cooler down there.

Just my opinion of course.


Actually, the PSU doesn't radiate heat like you are describing. That's why the fan is there. It pulls the heat out of the power supply and out of your case. And all that dust? You need to keep your system cleaner than that!

And, Crashman, you can get ultra quiet 120mm fans that move more air than a little 80mm fan on a PSU does. Scythe has a fan that outputs 8dba, that's almost no noise. But, being an extreme overclocker, I really don't care about noise. I care about my hardware not shorting out.
September 5, 2007 10:17:20 PM

korbin44 said:
Actually, the PSU doesn't radiate heat like you are describing. That's why the fan is there. It pulls the heat out of the power supply and out of your case. And all that dust? You need to keep your system cleaner than that!


The fan helps, but does not eliminate, internal heat in the PSU. When I touch the outside metal on my PSU, it's hot. It is therefore radiating some of its heat exactly the way I described.

The dust issue can be helped by cleaning, but even if you vacuum your case often, gravity is still guaranteeing that you will have a dustier, dirtier power supply if it's on the bottom of the case.
September 5, 2007 10:44:04 PM

My place is pretty dusty, but I don't end up with more dust in my power supply with it at the bottom. Actually, it's the same amount as the rear radiator has that is mounted at the top back of the case. Your PSU case may get hot to the touch, but most of that heat is pulled out by the power supply. All you need is a thermal imaging device to see exactly how the heat is radiated in your case. Maybe you have a crappy PSU. Yeah, you spent $140 or so on it, but that doesn't mean quality. Oh, and for the record, my PSU is on the bottom, with $500 worth of video cards above it and a $1000 processor above them.
September 5, 2007 10:49:59 PM

All my favorite power supplies have a 120mm or larger fan on the lid, which faces downward and pulls air away from the CPU.
September 5, 2007 10:54:01 PM

korbin44 said:
Hmm, looks like HP and it's 'engineers' agree too.

http://www.dailytech.com/article.aspx?newsid=8691

Check out where the power supply is located.


LMAO! You're using the exception to disprove the rule? That's just a modified version of someone else's case, a case that was designed to appease you rather than provide the perfect configuration.

I suppose all the "engineered" computers from other brands don't matter then.
September 5, 2007 10:57:39 PM


Korbie,

Do you live on the international space station? If not, you will have more dust/dirt on the bottom of your case. I can't imagine how/why you would be trying to argue this.

Your PSU is radiating heat too. Obviously the fan is blowing most of the heat out of it, but not all. I can't imagine how/why you would be trying to argue this.

The fact that you spent $1000 on a CPU might have something to do with it...

a b B Homebuilt system
a c 78 ) Power supply
September 5, 2007 11:09:54 PM

Each case tends to have a different air flow path.

Example

Sonata 2 and 3
Air is drawn in at the front bottom and back bottom next to the expansion slots. this allows the cooler hard drives to get the air first then the video card(which gets some fresh air from the back as well). Now the semi warm air moves over the cpu taking the last heat source on its trip out the rear fan and PSU. Since the air at the top of the case is now hot the PSU has to suck hot air, thereby lowering its maximum wattage output(its even worse with cheap ones). This is why many people say the psu is warm or even hot to the touch. it's not making all that heat by itself(but its still makes some. about 20-25percent of its power goes away as heat).

Now you look at a case with a bottom PSU. I am going to use the 900 as a example.

Cool air comes in the bottom front passing the cool hard drives(this depends on how you have it set. mine comes in all over) and in the back near the pci cards again. So now some air goes out the bottom of the case through the psu. This air since is cool to begin with causing the psu to spit out cooler air. In this case my psu air has never been warm since its getting rid of its own heat and not the full systems heat. The rest of the air goes over the normal path of video card(s) over the cpu area and out the rear fan and large top blow hole. The top air is much cooler since has skipped the psu.

Neither item is excessively hot but when all the air goes to one place the heat builds up quicker. So if the psu raises the heat by 3-5 c and the cpu raises it my 2-5 c, together it can be 5-10c hotter then the case air that has already passed over the hard drives and video card(s)

Both ways have a flaw anyway. the massive variation in hardware on the market. The lower rear intakes while good backfire when a video card dumps its heat out the back(or even when the psu does this, but it has less impact since its cooler) of the case since that hot air is drawn back in. This ends up being like a short circuit of air that gets heated by the video card dumped out the back them drawn back in and heated more and it just keep looping. So in some cases sealing off those rear inlets(as long as you have enough other inlets like a side vent or more front vents) forces the computer to get fresh air elsewhere and can lower the internal temperature of your case.

Ohhh and just for a point i had a old HP that ran the psu fan backwards and pressurized the case forcing out hot air through a vent of the front after it passed over the hard drive on its way out. As long as one thinks of airflow any case can be made to work well.
September 5, 2007 11:12:40 PM

You guys are childish. Honestly, how old are you? You must not build your own systems or your ridiculous arguments would have stopped a long time ago. Dust is negligible. The heat radiated is negligible when the fan pulls most of it out. Don't be jealous that I can affoard a $1000 cpu and you can't. Is it wrong for someone to want the best processor out? No. My next cpu will be the penryn extreme processor and I'll drop another grand on that. Heat is a big issue, and obviously a psu mounted on the bottom is the preferred way to go. That's according to Silverstone, Gigabyte, Antec, Voodoo, just to name a few companies that incorporate the bottom mounted PSU into their products. Get with the program. Your arguments are invalid. Mine are based on reviews that I've read from Anandtech, Tomshardware, xtremesystems and HardOCP to name a few. Do your research and come to your senses. Oh, and don't hate on a guy that has better stuff than you.
September 5, 2007 11:18:30 PM

Yeh Korbin, get with the program: Everyone is copying everyone else and VooDoo is just using popular designs from other companies. Lian-Li came out with a power supply at the bottom and a bunch of people like you asked Silverstone if they could have one, so Silverstone made one. Gigabyte said "Let's design a case with all the features that people like Korbin want" so they looked at the Lian-Li and said "People want the power supply at the bottom, let's do that"

It's easy to see that this is all a huge case of Monkey See Monkey Do, which is why Cooler Master waited YEARS before giving up and caving in on it's latest designs.
September 6, 2007 12:13:17 AM

As crashman said a while ago, for the people that this would even remotely concern, their internal case temperatures would most likely be in the low 30's, so the minute difference between having the power supply at the top or the bottom is really a moot point.
As for the power supply being "hot to the touch" in electronics terms that means nothing. Most people exhibit uncomfortable sensations around 43C, and pretty much all electronics can and will run forever at that temperature. Also, if you look at the specs for any power supply, (visit antec's site, for example), they state that the supply can deliver its full rated power 24 hours a day at 50C.
September 6, 2007 7:30:00 AM

Quote:
I'm not sure where people get the idea that the ATX spec called for the PSU fan to be an intake fan. Here is the ATX PSU Design Guide:

http://www.formfactors.org/develop [...] DG2.01.pdf


You're not sure because you're ignoring that this is not the original ATX guide. Think about it, "2.01" is obviously not the first.
September 6, 2007 7:36:00 AM

Heat radiation from a PSU is negligible. Remember that we don't care about trivial differences of a single-digit % or single digit watts. Not in an actively cooled system with thermally adjusted fans that produces at least 200W (below 200W, you can do darn near any config and it'll still stay cool enough so long as the case at least has a basic passive intake of suitable size and the AMD / Intel recommendation for a PSU exhaust plus one smaller fan.

When things get more interesting is when the video card alone produces over 100W, the CPu produces over 70W and you're trying to use the OEM heatsink, or there are so many hard drives in front that it's half blocking the intake airway (of course, ignoring all the really pathetic old case designs out there which were designed for the Pentium 3 era and put mostly stamped out metal around all the fan mounts because all they cared about at the time was EMI).
July 2, 2011 12:24:06 PM

I have a Lian-Li PC-Q07, a teeny-weeny mini-ITX that leaves only about 9-10mm of space between the PSU and LOW-PROFILE CPU COOLER(!), as well as pretty much everything else inside the case, AND there are no room for other system fans at all inside this tiny aluminum case: solution would be as the top of this thread suggests; to REVERSE the PSU fan so that it draws air through the PSU and into the case, and being a regular ATX 400W+ PSU, there no worry about hot air from the PSU blown into the case; the Shuriken(sp?) LO-PRO 32.5CFM CPU cooler (blows down into the CPU/board) would totally get a nice BOOST from the PSU fan; these are the ONLY TWO FANS in this mini-ITX configuration... average IDLE temps were 58C Mobo/62C CPU (ASUS MoBo/PhenomII X3 65W ONLY)... now they're only 42C/45C respectively. You'd have to break the rules when the engineering+styling become tortured as in the case (pun intended) with Lian-Li PC-Q07.

B. Chan
a b B Homebuilt system
a c 243 ) Power supply
July 2, 2011 1:05:13 PM

Wow almost 4 years old
!