Great! Now let's discuss the design of the unit and keep the spirits up :)

The overclocker will probably not be very satisfied with this unit, due to the lack of forced airflow. The mosfets on my board (nf7-s 2.0) stays at approx 1-2 degrees above room temperature when running the CPU at default voltage (but how often does that occur? :D )
My mosfets maxed out at about 120 degrees when using 2.1V, and that was with fans running in the case. However, will this be a problem in this case?
At 2.1V the temperature of the side panels will be ridiculous, if the heatpipes manages to transfer that amount of heat. The southbridge, clockchip and graphic RAM will most likely become very hot if you overclock the system, and won't have a forced airflow to aid in cooling.

However, maybe the natural airflow will be enough to keep the heat from these components away during normal operation and slight overclocking?

The multiple "thermal breaks" may or may not be a problem, we all know that you loose some cooling capacity when having a thermal joint, no matter how good your TIM is. The question is how much. It will be very interesting to see some real performance data on the final unit.

Oops, got to go! I'll check back later!

What about the Easy Bake Oven aspect of the case? I figure since all the sides of the case are going to be actively heated from components, what's going to keep the air in the case from heating up like an oven since there's no fans?

Also... what's to stop high heat loads from the CPU from killing off the cooling of the HD, and infact could reverse the flow of heat back to the HD through the heat pipes. (since heat can move both ways depending which side is cooler and which is hotter). Wont all components on a given heat plane in the case be the same temp at some point due to heat moving back and forth as everything reaches equilibrium?

Ok, so if we don't know the performance how can we discuss how to make it better. For all we know it is perfect as is.

See last statment. All we will be doing is talking shit being we don't know anything about the performance of the thing. How can we talk about inproving it if we don't know how it works. it might not need improving.
How do you know whats limiting the performance if you don't know the performance in the first place?
Can't add or correct being we don't know what the thing is capable of yet.
First off do we even know if any mods are needed yet? No... Second doesn't adding fans, tecs and water cooling gear pretty much destroy any reason to buy this beast in the first place? All that stuff adds noise, which is NOT the reason to buy this expensive unit. I don;t think to many people are interested in spending a lot of cash on a case that promises silence only to have to drop another small fortune for noisy items to make the thing work well. But we don't even know if it works yet so who knows...
Skip the preview and to the review with performance numbers and then we can discuss the situation. :shrug:

Whats the point in specualting what may work better when we don't even know how it works to begine with.

But if specualtion is whats wanted then I don't understand how this company thinks they can make all these heat pipes work without some sort of fan. Even heat pipes need the heat removed. Natural convection isn't going to cut it. Fans will have to be added to make it work. As unregistered just pointed I think you will either have to run a undervolted and underclocked CPU or a Cyrix in order not to expect a melt down.

Also pH pointed out the mosfets and other parts that get extreamly warm on the mobo. Both my ABIT KD7 and ASUS A7V8X have 6 tiny mosfets that there is no way your getting a heat sink on being they are cramed in between caps and they are just to small. They are about the size of the black cap on the end of a ball point BIC pen. Also the mosfets on my two boards get over 60C while running stock. At least thats wht my DMM and thermal coupler tells me. The caps get about 10C over room temps as does some other things.

Maybe this is why they have been working on it for 6+ months and several revisions and NDA's to not say how the thing works.

in the auto world that case would be called a 'concept car'
built for show and its pr value (hello ?)

in this world I believe it will be called vaporware (but I may be wrong)
it is NOT just OCers that may find dissatisfaction with that case
try anyone with a premium CPU

all a heat pipe can do is move the heat,
dissipation is still a function of surface area and airflow velocity

JoeK is on the money, the inside of that case will be an oven

1) You prefer to whine instead of contributing? I was actually glad that you did contribute quite a _lot_ in the rest of your post though.

2) I didn't actually mean to cool the heatpipes, but rather to use the cooling (watercooling or compressor) like usual and instead of getting rid of the heat with a rad + fan, use the passive heatsinks on the TNN500A. Coupled with a quiet compressor or waterpump (if that can be done) that should work better?
I guess I wasn't clear enough, would you like me to try to get more into detail? (And no, I'm serious and not trying to own you =) )


3) Excellent, just what I was after. Thank you for the input and hopefully we'll be able to answer those questions in the review. I won't be able to talk for Martin, but he can do that on his own.

Just a thought, looked at my subwoofer (Logitech Z-680) and thought that it's cooled by rather small (compared to the TNN500A) passive heatsinks. Still, it is putting out 500W RMS (according to Logitech). Does that mean that the Zalman should be able to handle a computer, or that most of the power used by my subwoofer doesn't turn into heat?

A 500W speaker?

Speakers are different: they don't put out 500W constantly, and a lot of that energy is converted to a mechanical action.

Processors and other electronics can put out a fair amount of power, and even with peaks, will produce a significant amount of heat.


Bill's on the right track. Zalman ought to have concentrated on eliminating the heatpipes, but that would have made for a really odd shaped heatsink.

I still say that they ought to have concentrated their efforts on silencing a fan. A few padded baffles (muffler style) ought to do it.

Contribute to what is what I havn't figured out yet. First Ben asks for comments on the preview, then askes for comments on the product, then you ask how to do the preview better, then you ask how to make the product better (without even knowing if it works or not). I am not whining either BTW. These are my opinions.
That will not work. I tested my Little Giant 170GPH pump with no fans turned on on the rad and the water crept up 20C over ambient. And thats just the PUMP, not including any other heat input. There HAS to be airflow to remove heat. Last time I checked air doesn't move to well by itself. :D Could be wrong though. It would take one massive passive heat sink just to keep the water cool from the pump only. A heater core sure didn't do it so I highly doubt the sinks on this thing, especially with the added heat load of whatever the loop is supposed to be cooling, will do it. And there isn't a pump on the planet that runs silent. Even the quietest pumps will be noisy in a spot where there is no other noise to blend into. I think a compressor is completely out of the question. Where to put it and how much more weight does this thing need. Not to mention they are not quiet and also require a fan and rad to work properly.

ok if this was in reference to me.

Also you cannot compare a subwoofer in this situation. A subwoofer does not handle 500W of heat consistantly. If it did it would catch fire.

Go get a 500watt heat lamp and grab it while after it has been on for an hour. Tell me how a little passive sink is going to cool that. (Discalimer: Do not really do this, sever burns can and will happen.) :D

Jaydee I am pretty sure ion was replying to my whining ass, as my post was numbered 1,2,3

Oh you CAN dissipate this heat with passive heatpipes all right. BUT AT WHAT TEMPERATURE DIFFERENTIAL?

This is the same sort of way that HWLabs get rated for heat dissipation I think; huge difference between the water and air. Will be same thing with components in the Zalman most likely.

I have a 1600+ XP that runs with a fan at 5V on it at 75-80C in the summertime. Case temps are low 40s. In that particular case I care more about silence than about cooling. We'll see how the Zalman system compares to just big heatsinks with 5V fans or better yet water cooling with a couple of 5V L1A or Stealth 120s.

Your sub is 188W RMS (checked the PDF, it looked too much for that type of woofer). The combined power with the sats and the woofer does 505W RMS. Very respectable pc sound. I want one :) .
5 cents on the subject:

Anyway. Take a memory module. It can consume up to 30W (usually <25W). It does however do it over all the smaller modules (usually 8x2) and with a large area by comparison to the consumption. It doesnt need extra cooling. The heatspreaders seen are just bling bling. You dont have the same on the zalman case. Your heat is concentrated on the lateral side, which doest have that much surface area , and even less airflow.

( I've had a very interesting problem when , back when i used sdram, the thermaltake heatspreaders on my 256mb module actually heated the module with the hot air from the heatsink. nasty stuff. Without them, it remained a bit cooler. )

In a subwoofer not all the wattage is directly converted to heat due to joule effect , and it's not continuous use. You are cooling mainly the power supply power diodes and/or regulators. Individually they can work happily at 120ºC without interfering too much with the stability of the rails (which matters). And each are spreaded though all the area of those sinks (which is not that small). They dont have to cool that much to maintain operation. They only have to cool enough. And the sinks are outside, in the open.

(http://www.bit-tech.net/review/266/2 , pikie)

In the pc, and referring to the zalman case, you can't have proper operation at 120ºC on any component... The MOSFET's can go to those temps, but the rail starts to go to crap. Same as hot air in the PSU . And inside the case there convection currents will stop at the top of the case (which i would suggest to be open , or at least heavily perfurated). You dont have this problem with the heatsinks on the Sub.

Assuming a very low airflow in a room , the zalman case would just overheat.

JoeK is right about the HDD thou. I had the displeasure of getting one for testing and it's just bad. It removes heat from the sides of the HDD ... to the sides of the HDD. If no air flow, no heat is properly removed, and it can work on reverse. The only thing that helped was when i removed the side panel. Interesting enough, it get the same results with no Heatpipe, and the HDD bolted to the case. Even if i should have larger area for dissipation. Then I tore it apart.

I can confirm the reviews that state that the heatpipes are poorly fitted, and no thermal paste exists between them and the HDD encasing. Just some crap glue. The contact is not very good.
Even with thermal paste (i had some Alumina leftovers) , without airflow, it's a mediocre cooler. Only with a fan it works properly. And with a fan and with and without heatpipes the difference isnt much, i got 2ºC due to larger surface area.

Yeah after I replied I realized he may not be refering to me. Didn't feel like deleting it all though as I feel it was usefull info.

Yes I agree is can be passivly cool. I was just using the unspoken guide lines we wanted our system to be somewhat opperational. T-breds and Bartons are only rated to like 85C now as opposed to 90+C they used to be. Not sure what the Athlon 64's and P4's rating are. I would not feel comfortable running over 60C continuously though. If you buy a freaking expensive case for your computer I would expect it to be able to handle full load temps from games and D.C. projects and multimedia apps.

This setup from Zalaman seems like a run for an hour and shut off for an hour type of thing. Sooner or later the heat is going to overwhelm the passive capabilities and start causing problems unless the room is air conditioned to chilly temp.

But this is specualtion. If we don't know what the thing can do we really can't say much. I assume they are on to something to have spent so much time and money developing this thing. Or maybe they are screwed and can't admit it to the boss. This whole send the prototype out might just be an attempt to keep the investors happy that something is being accomplished with their money. Maybe not. hard to say without some numbers to show it works or not.

hehehe, a Case with a CPU load Duty rating ;) hehehe "100% CPU load duty rating chart" with a curve setup for max watt output from the CPU ;)

Feel free to be an example of ideal forum member behavior.

Popping in here to complain, (Which, from what I've seen, seems to be all you've done lately.) is just contributing additional whining. When is the last time you answered a noob's question?

Seems to me you are just choosing to be 'part of the problem'.

The best i think Zalman could do is having a bigASS radiator on the top of the case, with no pump. Natural convencion would make hot water go to the radiator to be cooled.

I dont know about temps, but surely would be high (well depends on how BIG the radiator gets).

nope
google on 'thermosyphon'
they can be made to work, but never the way you have suggested

The heat pipes are a lot more efficient than a passive water circulation system.

Heat pipes work by phase change, the boiling of the working fluid in the pipe removes a lot of heat. The density of the gas is way lower than that of liquid, so it can carry away heat much faster than liquid only in a passive system.

If they plan the case well, they can get some amount of airflow by convection.
The Apple cube is only 8x8x8 and it dissipates about 45W by convection alone.
Upgraded Cubes do need fans though. If the cube was larger or had better airflow and heat transfer through the case, it could dissipate more heat.
At least that's Powerlogix's plan with thier aluminum case replacement.

The Zalman case is about silence, get the overclocking out of your mind, it wont work for that. If anything you will probably need to UNDER-clock a bit to keep things running well. It wont do Dual Opterons with a 10KRPM RAID array.

There is a good sized market for Quiet over fast and cool. Zalman's niche is in that space. They better get moving on this design before some other case maker beats them to market.

You should write marketing for Zalman Starman:

"My computer crashes even at stock speeds" was successfully spun to:

"If anything you will probably need to UNDER-clock a bit to keep things running well."

Silent, Cool, Fast...
Pick Two..

actually its Silent,Cool, or Cheap

Heh.. but here, none of it's cheap..

Overclocked Opteron SMP with ATI9800XP
the ubercooled phase change systems
or silent $500 zalman case..

If you go in any extreme, the costs go way up. and some times you get 2 of the three qualities, but if you really push, you only get the one .

Cool and fast are closely aligned, but sometimes it seems people's goals are to have the coolest die temp, instead of the fastest machine...

Although cool CPU might override the others for something like a laptop..

But basically the Zalman case is not for people who hang here, they're about speed and cooling systems, damn the noise.
"70dB screaming Delta fans.. bring em on if I can get 1C cooler die temp" seems a common sentiment.

How many articles have been posted here about various waterblocks in development, and with no hard numbers or anything until the thing is ready for final testing. How many months have some of those projects taken to reach readiness?

For me, noise is every bit as valid a spec as flow rate or die temp, I come to watercooling looking for a noise solution primarily. If I had to run my CPU at 60C and non-overclocked and could have a nearly silent system, I'd go for it. So, if the Zalman case is more of a concept case, that's fine. Lets see what they are doing and see if there is a way to do it better. I'd rather see how they did it, rather than just a number, of course having both is better, but it beats having no information.

"70dB screaming Delta fans.. bring em on if I can get 1C cooler die temp" seems a common sentiment.

I doubt that anyone in this forum uses the EHEs for anything other than heatsink testing. Maybe on hardforums or xtremesystems? I was wondering the other day after a week of listening to a 92mm Tornado if anyone actually ran them all the time. I can't imagine that personally.

"If I had to run my CPU at 60C and non-overclocked and could have a nearly silent system, I'd go for it. "

You can get die temperatures well below 60C with overclocked tbreds and bartons using water, an Eheim 1048, a big heatercore, and 2 120mm L1As at 6V. That's "nearly silent" and not decreasing the life of your system either.

And I am glad that you brought up laptops. If in the Zalman case you are restricted in video card and CPU choice by what won't get so hot it melts, then why not just buy a laptop? I have never had any complaints about the noise levels of my thinkpad, and the new t40s have fast CPUs, mobility 9600s and all the trimmings. When you add a $1000CAD case to the cost of a desktop then I suspect higher end notebooks start to become competitive. For that matter the new Apple G5s are pretty quiet and not that much more expensive than PCs...

I had my laptop hooked up to a 19" monitor, standard keyboard, standard mouse, and stuck it on a shelf in my desk. Worked awsome for quiet computer! If I didn't actually use my laptop at work and for my CNC mill it would have stayed that way. :) Good point.

Yea my Thinkpad is my primary work PC: I use dual display to have its 14" screen and a 19" Sony monitor in multimon action. I don't bother with an extra keyboard but I use a MX700 mouse. It's nice to work in silence. With the newer notebooks you can even get by pretty well at LANs and things are only going to continue to improve. Notebooks suck for serious i/o work; wouldn't want to do video editing on one. But they make fine quiet PCs.

Regarding the oven-effect: don't forget the holes in the bottom and on the top. It will probably still get _hot_ (like a monitor?), but you won't be able to bake your bread in the case. =)