Why can't a case replicate a CPU heatsink?

[at2u]

Some cases for lower power CPU's manage this but its very expensive - probably because of all the metalwork involved. I just want the case to act as a heatsink. I could run a very low rpm large fan to cool it that would be completely inaudible. There seems to be nothing out there. Just heatpipes from the CPU to the casing.

 

[UsandThem]

There were a few in the past, but I'd have to dig around to remember their name (I think Zalman made one).

However, they were expensive and I doubt any manufacturer made any return on their investment.

The biggest obstacle they faced was users buying a much cheaper case that had some sound-deadening material, and add a few quiet PWM fans.

 

[UsandThem]

Yup, here's the one I remembered from back in the day: https://www.quietpc.com/tnn500af

 

[aigomorla]

Very costly at marginal performance gains.
This is also why most case makers have even shifted back to steel frame over alu, because of cost metrics, which i feel is a step back in evolution.

Sort of like how you see luxury cars now being made of Alu frame, and not steel.

 

[gorobei]

the physics and economics dont really work.

typicaly secc steel is used in cases in 6 to 8 mm range. the thermal conductivity of steel with zinc galvanized coating with paint is probably way way worse than a basic aluminum fin cooler.
the thermal mass is pretty low and the surface area of a side panel is probably only barely equal to a tower cooler. assuming you routed a heatpipe from the cpu to a side panel you would have to have a tensioning system to make sure the contact between the pipes and the sidepanel remained consistent every time you opened and closed the panel. that system is additional complexity and cost.

most passive cooling cases are made of thick aluminum with fins for additional surface area, so either extruded or milled = massive cost increase in materials alone much less the machining time.

the additional cost difference for a "simple" silent case vs a cheap case with cooler is not something that can be overcome with a niche small volume run. even if it works for most casual cpu usage, it will hit the throttling point way sooner when any extended use case comes up.

not saying it could be done to some acceptable level, but you would first have to convince several million people that such a thing was desirable enough in order to create a market demand. and it would have to be a sustained demand for any case maker to spend the time/money on tooling up for the heatsink-sidepanel and heatpipe engineering.

 

[Fanatical Meat]

As said above zalmen made one, the basic economics of buying a super complicated case that will likely only be used for one build because the next build will have a different CPU mounting configuration makes the idea unappealing to most people.

 

[DrMrLordX]

the next build will have a different CPU mounting configuration

. . . and there's the real problem. Looking at the tnn500af, it does seem possible to make a case that acts as a dissipative surface, but it's insufficiently flexible to work with any hardware you'd want to throw in there. I'm a little surprised they made the heatpipe arrangement as flexible as it is on that old case.

I would rather see a case with a built-in pump/res setup that pipes heat to a massive rad built into the case itself. Then you could swap out the block based on your socket config, and maybe add a GPU or two to the mix. Maybe something like that already exists . . .

edit: I do see some 280mm and 360mm rads integrated into cases already, but those are small potatoes. It would be more fun to have something like a MO-RA3 or larger built into the side of the case exterior.

 

[artvscommerce]

I've noticed the Streacom products have become lower in price over time. Have you checked their stuff out?

 

[mindless1]

There is no benefit in doing so unless you are developing a low power portable system where the case and mainboard are integrated to the point of marriage.

The issue with heatsinking being the case is you don't just need surface area, you need conduction rate which means the thickness of the metal. It is much more cost effective to move the heat away from the component with the majority of the mass right at the heat source then fins radiating outward from that.

So the question you asked is invalid. It's not that a case can't be a heatsink, but rather that it is just some novelty idea to try to do it with no practical benefit for a typical PC. If you were talking about some special avionics or military module that needs to be sealed then it will be in a thick metal case designed specifically for that as I mentioned initially, completely integrated not a standard ATX consumer grade etc.

I could run a very low rpm large fan to cool it that would be completely inaudible.

That has nothing to do with whether the case or a separate heatsink is used. You can literally slap a big ole audio amp heatsink on your case panel if you are just too bored to use proven working solutions to cooling, or remote mount a water pump where you can't hear it, to a large enough radiator that you don't even need a fan. However, I have built many systems with heatpipe sinks, including overclocked, where I never hear the 120mm heatsink fans. It's usually going to be a case wall mounted fan or PSU or especially video card fan that makes the most noise that escapes the case.

There are all kinds of wacky ways to make life hard on oneself instead of just using what everyone else does. Besides, if you make it too quiet then you'll be cursing at crickets for disturbing that.