The stack effect — also known as the chimney effect — is the basic principle that hot air not only rises, but if guided through a tube, the rising hot air will create more pressure that will effectively draw air more effectively into the tube. This is not only great for a chimney, but also a useful principle if you seek to cool something like a PC in a more passive manner. The main question is of course how much of a ‘chimney’ you need to see real effects for something like a typical water-cooled CPU’s radiator, as demonstrated by [der8auer] in a recent video.
Although there’s a lot of fun physics behind the stack effect that you can run the numbers on, the more practical demonstration here using 3D printed funnel segments for the radiator and various thermometers provides a very hands-on feeling for what you can expect from this approach.
With just a single segment stacked there is already a clearly noticeable temperature change, with the second segment creating a draft as visualized by the smoke machine. After this he goes for broke with the full stack and a resulting 19°C temperature drop on the CPU. While impressive, at this point the required funnel gets a bit silly, though the same principle has been applied to computer cases before, including the passively-cooled Power Mac G4 Cube and the 90-degrees-rotated SilverStone Raven series of cases, like the RV02.
The basic idea of making use of the fact that hot air rises, and maybe also banking on the stack effect for some free passive cooling, clearly isn’t so crazy.
hmmm… suddenly a steampunklike smokestack on my PC seems justifiable.
This design isn’t new, I’ve seen them in 150 years old barn.
I need to see if it can work on PC case with large top like Thermaltake Tower 500
It is called a chimney, and that is indeed not new. 😄
Please put the stereotypical gigantic Simpsons-style nuclear cooling tower chimneys on all the datacenters. It would be funny
I assume that the “19°C” are a temperature difference? If so, then shouldn’t it be stated in Kelvin (K)?
temperature difference with/without the chimney, in any case, any fans would be even better than this.
Interesting demo for the effect, little use in a PC… through I wonder if the big hunk of metal that is that noctua passive cpu couler would do with just a chimney and no fan
You would only need to use Kelvin when using percentage reduction or “halfed my temp”. People will say I got a 25% temperature reduced from 100c to 75c which is inaccurate (it is 6.7% reduction).
I’ll just stop here recommend “Hot Air Rises and Heat Sinks”, by Tony Kordyban. It’s by far the best book on cooling electronics I have found.
I know nothing about the contents, but good title!
The chimney should be insulated to minimize heat loss and maximize air flow.
Interesting. How much would an uninsulated wall reduce the net cooling effect?
How much would the reduced cross section of the duct (due to the thickness of insulation) reduce the net cooling effect?
You only insulate the outside. You want the heat to stay in the air inside.
If it heats the chimney, the chimney will cool from the outside. That is what you want to stop.
Natrural ventilatrion is driven by density differences, and you can calculate the pressure differences easily by this formula:
dp = (rho2-rho1)gH
Here, rho2 is density of air inside the tower, rho1 the density of the ambient air. g is the acceleration due to gravity and H the height of the tower. As a rule of thumb, the buoyancy pressure at room temperature is about 1 Pa/m for a temperature difference of 25 K and it is changing linearely with height and temperature.
So, you can easily predict the effects. The driving force of the flow is P = V * dp in [m³/s] and [Pa]. For a small tower of 1m height at 25K overtemperature, the velocity will be u = sqrt(2/rho * dP) = 1,3 m/s, which at a duct section of 0.1m² gives only driving power of 0.13 W, which ist much lower than any fan. But it is quiet. And if youo do the same calculation for large buildings like power stations with a height of 120 m, the driving power of natural ventilation may be very impressive.
best regards, Eckehard
It is well known that the classic hyperboloid shape of cooling towers is driven mainly by structural strength considerations. But I have always wondered how much the flaring at the top, and consequent reduction in flow velocity changes the pressure differential and flow rate.
There’s also the effect of introduction of cooling water vapour, dramatically decreasing the density of the air within the tower, increasing its buoyancy.
Fan will move more air and take less space, but it is hard to beat the quiet operation of the chimney.
Been seeing a few WiFi AP/routers built in the smokestack format for seemingly the same reason, and yeah, the hot air definitely rises out of them, and it’s a good form factor for concealing a real antenna too.
As far as the experiment goes, it’s pretty clear that the length and aspect ratio is important. It may be worth tapering the stack to get the velocity up since the air is cooling and taking up less volume as it moves up the tower, although the ∆T is pretty small.
Somebody buy this fella a hot wire anemometer 😁
The overtemperature will drop the moment the air starts moving. Cooler ambient air will be drawn through the radiator and the faster it goes the less time it has to heat up, so the rising air becomes closer to ambient temp.
Right, but the assumption clearly is the air column is maintained at that temperature.
Which implies in this case a continuous power input of around 4 kW, which intuitively feels a little high, so feel free to do your own arithmetic.
I have not done experiments. I have not used liquid coolers. I have not built pcs in a long while. I have seen quite a few builders trying to use many fans to push more air into a tightly closed chassis. My thinking (maybe wrong) is that creates back pressure, especially (maybe) compounded by having a radiator at the top possibly blocking air going out. I have seen some system with the liquid radiator outboard of the chassis which permitted better airflow out of the top. I was thinking to try some high flow fans at the top with unimpeded intake. I understand that add in cards and cpu heatsink will cause turbulence. I had also thought to test completely crazy and “deconstruct” the computer(s) onto a shelf, in a server cabinet, etc., but “box in” the entire cabinet with a powerful exhaust fan at the top with a cold air supply at the front/ bottom.
Arguments about optimal fan and case design have been raging online since the early days of DIY PC building, and it’s a frustrating space to wade into because there are so many people with strong opinions, lots of circumstantial evidence and misinterpretation of science, and comparatively little real-world testing (although of course there are some people, like [der8auer], that make a decent attempt)
That said, there is at least one advantage to having a system with a bit of back pressure (net positive pressure) rather than neutral or negative pressure- dust control. Positive pressure keeps dust from sneaking in through every open hole in the case, and you can put a filter on the intake fans to catch a lot of the remainder.
Putting a radiator outboard of the case can improve cooling efficiency, especially if allows you to use a bigger radiator, but this is mostly an advantage if you want to build a quieter PC, as honestly, most modern multi-fan cooling setups can move plenty of air to keep the system cool, even if their position is far from optimal.
(Incidentally, some modern PC cases have internal dividers and multiple intakes to try and isolate the airflow between various components, so your radiator, motherboard, GPU/PCI cards, and power supply can all have access to fresh room air – this seems intuitively better for cooling, but again, it’s hard to say whether it yields much of a real-world benefit – it depends so much on what you put in the case and how you cool it)
Even termites build chimneys to increase flow. They let the sun heat the chimneys to boost it.
Wait til he learn about adiabatic cooling.
This reminds me of a thought I’ve had since power usage of GPUs spiked.
I wonder if there is a good way to cool a PC that directly dumps the heat outdoors. Having a space heater fight your A/C in the summer doesn’t feel like a fun idea.
I’ve thought of things like through window tubing with an external radiator and “just putting the PC outside like”. Now I wonder if a nice chimney could work out.
Your dryer probably already does this.
The dryer also sucks cool air from the house and blows it outside, which is handy for drying clothes, but probably still a waste of energy when it comes to PC cooling (unless you need your PC to be below outdoor ambient temps).
If you want to dump the heat from your PC outside then ducting and an extra fan or two is the simplest solution.
The other option is that your AC is the thing that moves it
Feels like there’s potential for using a heat pump – maybe an RV-sized one? Or I suppose you could stuff your home server into the ducts with your home’s heat pump…
that’s still functionally “having a space heater fight your a/c”, and not necessarily much better than just running the computer inside your air conditioned space.
You don’t need to do anything that fancy.
If your PC is air-cooled, you can move it to a ventilated rainproof enclosure outside, or keep it inside and enclose it so that intake and exhaust air flow to/from the outdoors.
Easy to run long USB and display cables back to a dock in your air-conditioned space if it’s a desktop machine.
If your PC is water-cooled, you can just run pipes outside and have the radiator outside.
With any of these options, if you want to experiment with a giant chimney rather than using fans, it’s easy to do so.
Connect the waterblock to mains water and send the return directly down the drain.
I’ve been doing similar in the recent UK heatwaves but for cooling the room instead: mains water goes through a 360mm radiator and chills the air travelling through it. When ambient air temp was 36C and water temp 24C, I was getting a nice 26C breeze. Actually had to insulate the pipes because they were accumulating a lot of condensation!
Vertical rotation wasn’t a Raven series thing, it was specifically the RV02 (which wasa more plasticky version of the FT02).
Even way-back-when the Duron 600 ran my main PC, I saw that running it with the side off was better for temperatures. I ran it with the tower lying on it’s side with some mesh to prevent bits and bobs falling in.
I ended up with a Thermaltake DH-102 with the cover off and liquid cooling running through an outboard radiator. Same case for 15+ years now with a 14th gen intel inside.
I did play with the chimney effect but it looked rubbish – I’d have been better served just moving the PC in to the disused chimney!
“…and the 90-degrees-rotated SilverStone Raven series of cases, like the RV02.”
Rocking one of these, but it’s downside isn’t cooling but power and reset button location.
Am I the only one who remembers the MacChimney? Which attached to the top of early compact Macs such as the Plus and, as the Macintosh Bible put it, made your Mac look like the Tin Woodman?
https://www.reddit.com/r/VintageApple/comments/ekc9gc/macchimney_cool_your_mac_and_make_it_look_like_a/
I didn’t. Thanks.
But I cannot believe that I am the only one mentioning the “Trashcan Mac Pro” which worked the very same way in matters of cooling? Though that one used to have a fan at the bottom.
“that will effectively draw air more effectively”
U wot mate??