Caught this on Slashdot, seems like SPCR's kind of bag:
http://inventgeek.com/Projects/IonCooler/Overview.aspx

Ionic Breeze crammed into full tower case with passive PSU and GPU using RAM disks (not the newer flash disks, but the old-style "stick some ram on a battery powered card and plug it in to SATA" sort), he claims complete silence. He's got pretty detailed instructions, it's a very interesting concept. He's claiming 325 CFM off his ion rig, which is pretty substantial. What intrigues me about this isn't really the possibility of using the giant room-purposed ionic fans, but what about the "bathroom" ion fans? They're much smaller and easier to handle, though they're obviously designed to run off AC. Wonder if you could use, say, two of them to replace a dual-80mm column in the rear of a case? Or perhaps strap them to a big ass tower heatsink, like a Ninja?

Thoughts a-brewing.

Very interesting DIY project, but I call BS on the CFM claim. Firstly he never tells you how he measured it (or whether he simply read it off the packaging of the ion generator), secondly even industrial fans running off 24V don't produce that kind of airflow. Thirdly if there really was 325CFM going through his rig there would be lots of turbulence noise etc, so not 0dB. I have no doubt that the technology works, so there's no need to exaggerate the CFM.

I like to see this sort of project......but. It looks dangerous and way too difficult to construct. The CFM claim is bogus, maybe a mis-print. And I bet this system could double as a space heater in the winter.

Interesting idea, but terrible write-up of it. Just an odd selection of details: he explains exactly what type of hot-glue to use, but declines to give any temperature results? The text implies that there was several iterations of the final design that were tested, it just seems odd that you would go to those lengths and then not post actual temps. Color me suspicious. (I entertained the idea that the whole thing was a Tt viral marketing campaign, but there's other content on the site)

If you did real testing I'd bet money on the difference in temperatures between having the ionic filter on and off would be pretty small. Doesn't appear that he even thought of trying that as a test. With the exception of the CPU, its a pretty modest system that wouldn't be tough to run passively. For all we know the CPU could be thermally throttled 90% of the time.

The 325CFm figure is obviously pie-in-the-sky. The text is chock-full of errors and inaccuracies, so I would not be surprised to learn that he arrived at that number via rectal extraction.


But I like the idea...I'm just frustrated by the fact that a site that aims itself at "inventive geeks" skimps on the exact kinds of details that us geeks relish.

And what's with their forums? The site has articles going back 2 years, the forums have 170 members, and there are zero posts?

Somoene on the SPCR staff should replicate this DIY project, and do a real review on how well it works.

Doesn't have to be a staff member...this would be a perfect article for a reader submission. Just take their existing concept and apply some better scientific standards to its presentation.

Then you can add "published author" to your resume.

I'd love to tackle this project myself, but the investment is a bit high for pure science content. I'd say the thing to do would be to just get someone to create a temp shroud for their room-sized ionic fan and use an anemometer to measure the cfm and report back. If the properly assembled unit isn't moving 325CFM, then the hacked up unit probably isn't either.

I have to agree though, from what I've read about them, 325CFM sounds quite high for the flow rate. However, if a wall-socket sized one could generate 10-20CFM, it could conceivably be used on a megatower heatsink, providing air movement without additional noise. Maybe. Worth a shot, at least. If I had one, I'd give it a go.

I caught it on engadget, interesting build but not enough screenshots for the details. CFM claim is a bit bogus because if the air is flowing at that speed through the small front side vents/chinks, it will probably sound like a beast sucking in air through its teeth.

Be interesting to see the time frame before the optical drive starts to fail, as well as any degradation on the less inert plastic components. Probably won't be included in the project documentation.

There's temperature readings now on the last page. If those numbers are correct, I'm impressed.

A 3 degree swing between stock HSF with stock case fans and Tt passive heatsink and zero case fans?


Sorry, I call BS on that....not possible.

I will happily eat my words if proven wrong, but until I see someone else try this I think there's something fishy. I hereby resurrect my theory of it being a Tt viral marketing campaign.

Major errors, problems, but an intresting problem.
TT would NOT be my first choice for a case for those purposes. The Aluminum they use in the cases isn't as good for silencing as some other cases, the open front makes noise more percieved, and a lot of other stuff.

Assuming the cooler really DOES the claimed CFM, at 0 db (bs since there's turbulance noise), there's also the whine of electrical components, among other sources of noise. Not exactly 0db.
A prescott isn't the best choice. AMD does offer some thermal throttling (shutoff) now, and dissipates much less heat. A Conroe or Allendale would have been a better choice for passive.
Ramsinks aren't as useful, and they block valuable space.
A 20-24 pin adaptor isn't needed. Just plug the cable straight in leaving 4 pins open.

I'd think a nice well designed case, with well thought out airflow, and various methods of construction, will be as quiet as or even quieter than this PC. Notebook drives suspended in the lower chamber of a p180, all the fans replace with undervolted nexus' or yate loon's, some damping, a scythe ninja, and a single yate / nexus in the upper chamber exhaust, aerocase condor, thermalright hr05, and it'll be dead silent.

59С under load with a Sonic Tower and no fans? No way, Pentium D is too hot even for a Ninja.

In my opinion, the statement "The lack of internal harddrives really helped in power conservation" is a major flow in the documentation of the project and shows a deep unknowning of the power consumption of computer components... This leads me to conclude that the guy who did the whole thing doesn't really know what it's all about. I think that a good understanding of the power needs of every component in the system is essential when building a silent computer.

As previous tests at SPCR have shown, a 80GB hard drive would not consume much than 10W, even during extensive seeks... On the third page of the documentation, a 8GB RAM disk is introduced, which would doubtfully consume less than a ordinary hard drive. I think the use of RAM disk is pointless. It would look great in the eyes of some guys who know nothing about the power consumption of a hard drive, but in my eyes this looks like regular marketing nonsense. Or maybe the presence of a motor impies a 100W power draw for some people?!

The second weak point that attracted my attention is the choice of the Thermaltake fanless power supply... I think it's just the worse fanless PSU one would like to buy when starting such project - regular readers wouldn't ask why.

The third major issue - the choice of a Pentium D CPU. The autor stated that "The main reason we chose the Intel CPU is if for some reason there was a thermal issue it will self-protect the CPU unlike AMD's". I may be wrong, but for me the main reason for choosing a CPU to be cooled passively, with a modest performer like the TT Sonic Tower, would be the TDP! For the Pentium D 940 B1, it is 130W - the highest value for any Pentium CPU, pretty high to be cooled passively by any heatsink I've heard of...

And just one more consideration - the 0db issue. So far, I havent heard of any real environment which would allow the measurement of such value. If this is "above ambient" value, I wonder how the other 38db system would sounl like in the same environment...

I didn't want to bother with the other nonsense in the documentation, it's just too much for this project to be considered serious.

I read this review more thoroughly than before. I think it is an interesting concept. Clearly the cfm figure is out of the ball park.

The rest of the figures though dont seem quite as unrealistic.

Basicly looking at the table, the author is reporting a 7c drop using the ioniser. This is compared to running the system passive. It would only take a small amount of airflow to achieve this.

Also the heat is being dumped outside already on the gpu and the power supply. I dont have experiance of the power supply, but the schooner is a pretty effective product. My brother used one for a while, and the external heatsink does make a difference. A very unscientific holding your hand near it test demonstrates this.

A very small amount of airflow could make this difference, and that i think is what the author has achieved.

As for the extensive use of tt parts... Well the author did mention that TT had gifted the PSU. He also mentions that his work use tt parts extensively. Its not unusual for manufacturers to product place via donations.

So if you cut through the glaring errors in the write up, this is an interesting article. I think that someone here should give it a try. I am kinda tempted, but I got other stuff on the workbench just now..

I've tried both the Schooner and the BeQuiet! Polar freezer on my GFX. I got similar temps, the Polar freezer a degree or two better.

The external part of the Schooner isn't really helping much. The fins are way to narrow apart to allow any convectional cooling.

What is most interesting though is that his temps, with all the passive cooling and NO airflow, is perfectly acceptable. Why bother then?

Just an example - Titan offers some high-performance fans (over 100CFM). You may imagine that at this airflow, the noise character of the fan is dominated by air turbulence, and not by motor noise - and it is well beyond anything that one would call even quiet. Now we have 325CFM - hence, three such fans... and 0db. One must be very convincing to make me believe that. And this project isn't.

I just wonder how such CFM value can be achieved with no fans - and how they were measured.

Beside this, none of the CPU temps exceed 60 degrees celsius. I agree that these temperatures are acceptable - even more than acceptable.

What I argue is that they can not be achieved in a pure passively cooled system, which stands in the table.

well, we have a homemade electrosatic precipitator in our lab, and considering how cheap it was to make, I might consider trying this. WalMart sells ionic pro ioinzers, one is only 8'' tall. Not sure how much of a wind it can generate, but I'm going over there tomorrow. Hopefully they'll have a demo unit, if not, I'll buy it, if it sucks, I'll return it. If it draws a decent amount of air, then for $20, I'll invest. I've got plenty of lab equipment at my disposal, the only thing stopping me is time. They also sell an $80 version, it's a bit bigger, even though they call it the "compact" model. Regardless, I don't have a heck of a lot of room in this case, so, the smaller the better, that is, if it works.

obviously, he meant 325 cubic feet per month

I have an ionic air filter, it makes very lound nose when it is time to be cleaned.
And it does make an electrical noise that I can hear on a very quite day from 10 feet away.
the filter does move alot of air, but it is at zero pressure diff. so moving air from an intake to a case would be different.

FWIW......Have you ever seen the TV comparison between this device and a standard small portable air cleaner (an Orek). They drop a balloon over each device whlie they are running. Over the Orek the ballon goes up. Over the Breeze device the balloon just falls down.......not enough airflow to move a balloon. The ad says "If you don't move the air, you can't clean the air". So much for massive airflow.

Every consumer report on these sort of devices gives them a bad score as an air cleaner. "Don't waste your money" is the conclusion.

you're talking about passive ionic coolers, right? because they use these things in industrial processes to clean flue gas, not as a passive cleaner, there's always an induced flow rate from a process or pumps, but they're extremely effective, but the intention in these cases isn't to move and recirculate the air.

Just because they may be poor air cleaners doesnt mean they would be poor fan replacments.

Even if the airflow is minimum, it doesnt need to be that much. SPCR has proven that even minumal airflow is greatly more effecient at removing heat from heatsinks than no airflow.

the fact of the matter is its no moving parts, completely eliminating motor noises, and wear. even if its only as much airflow as a nexus fan, it would be good enough for most SPCR users.

I just wish they would make these devices in a standard case fan form factor so you could just pop them in as replacements for standard fans.

as far as the claimed CFM, i have a feeling he just put a decimal point in the wrong place. people make mistakes, dont make such a big deal out of it.

It may also be worth noting that the two gases that make up the major part of the air (about 90%) - oxygen and nitrogen - consist under normal conditions of molecules with balanced electric charge, that is, they appear electricaly neutral to the environment. Moving these gases via electric charges generated by a ionic cleaner is a tough job, not to say impossible.

Ionic cleaners do their job by attracting charged molecules to them, not by moving any air.

Having said this, I wonder if it's possible to generate an airflow even close to a small fan.

I agree that sometimes little airflow may do a great job - but consider cooling a 130W rated CPU with this little airflow, and obtaining 60 degrees celsius...

I'm with Npp on this one.

Here's a first-person parallel example:

I have a tower system that I've been tinkering with that uses a passively cooled cpu. It's got a Katana-CU heatsink, which is probably as good or better than the Tt Sonic Tower for low airflow/passive cooling. And this system is being cooled by a single 5v Nexus 120mm fan, which is probably pretty close or better than the actual airflow generated by the ionic breeze. The fan is not ducted to the heatsink, so the airflow pattern is similar to the ionic case too.

At load, my CPU tops out at 54°




My CPU is a 14 watt NX1750.

It's definitely not impossible, see http://en.wikipedia.org/wiki/Ionocraft , the main question is whether it is a practical means of achieving a more silent PC.

Good example Rusty! If a 14W CPU hits 54C, there's no way a 130W one can be in the same temperature ballpark.

I've also read some projects offering space ships, accelerated by ultra-high energy laser beams, pointed at them... So you're right, that such absurd-at-first-sight results are not impossible - but just a little quote from Wikipedia: "When the ionocraft is turned on, the corona wire becomes charged with high voltage, usually between 20 and 50 kV." Highly improbable may be better suited to explain the situation.

Let's just say that moving even small (I mean really small) ammounts of air - and thus removing heat from any component in a PC - using the described technique ist just beyond the capabilities of the low-energy environment of our desktops...

The 14W CPU example really shows the point...

Remember that big voltages do not necessarily mean high energy consumption. Every CRT display contains similar voltages and still as the current is fairly low, it does not consume *so* much energy.

they induce what has been called by my chemical engineering department's technician a "corona wind". Think about it, you charge the air/particulates in the air with -20kv, and they're going to move pretty quickly to anything at a lower potential. Give some large plates a positive potential, and they'll go even faster. It will generate a wind, the issue is if you have enough small wires to ionize a large volume of air. Then you have the issue of charge air flying around. In a pc case, i see this as a bad idea. Especially in the winter months with all the dry air. The "inventor" of this project puts screen around the aparatus and grounds it. I'm assuming he's doing this to attempt to neutralize the gas he just ionized so that the static buildup doesn't short to one of the pc's components. Still, the gaps in the screen are huge compared to air molecules. To effectively neutralize all the air, you need a fine mesh, and then airflow becomes restrictive.

I'm not sure why he only grounds his cpu hsf, and none of the others. This could work, and work well, but the risk of static discharge is one that I wouldn't want to take.