Personal computing discussed

Moderators: renee, Starfalcon

Grand Gerbil Poohbah
Topic Author
Posts: 3774
Joined: Wed Jan 26, 2011 12:50 pm
Location: Wisconsin, USA

Optimal Heatsink Fin Density?

Thu Aug 15, 2019 9:38 am

I presume there's a logical answer to this question (and I have my own theories) but: Why is the fin density on water cooling rads so much greater than on "air" coolers?

I envision air/tower coolers' fin density to be pretty optimal. Having enough fins to maximize surface area without getting so dense that even high static pressure fans can't push air through them. On the other hand, water cooling rads seem to all have their fins packed so densely that even a high static pressure fan gets ~50% of it's airflow bounced back. Both types of coolers operate on the same premise, use a fan to blow air across some fins. So surely there's an optimal fin density regardless of the method you're using to get the heat into the fins in the first place....right? If so, then what's the reason for the existing density difference?

The heatsink fin density on graphics cards tends to vary more. I presume that's a direct result of the manufacturer's choice of fan and that fan's static pressure properties. In this case, it would seem that [at least some] GPU manufacturers are actually designing their heatsinks in close relation to the fan that will [more commonly] be forever strapped to them.
Main: i5-3570K, ASRock Z77 Pro4-M, MSI RX480 8G, 500GB Crucial BX100, 2 TB Samsung EcoGreen F4, 16GB 1600MHz G.Skill @1.25V, EVGA 550-G2, Silverstone PS07B
HTPC: A8-5600K, MSI FM2-A75IA-E53, 4TB Seagate SSHD, 8GB 1866MHz G.Skill, Crosley D-25 Case Mod
just brew it!
Posts: 53991
Joined: Tue Aug 20, 2002 10:51 pm
Location: Somewhere, having a beer

Re: Optimal Heatsink Fin Density?

Thu Aug 15, 2019 10:14 am

I don't know the reason for certain, but water cooling is more efficient at removing heat from the CPU than the heatpipes used on air coolers, and can handle higher TDPs; so logically you would need more fin surface area to get rid of that heat?
Nostalgia isn't what it used to be.
Lord High Gerbil
Posts: 8549
Joined: Fri Jun 21, 2002 10:47 pm
Location: Sunny Colorado front range

Re: Optimal Heatsink Fin Density?

Thu Aug 15, 2019 11:22 am

A solid-mass heatsink needs enough metal mass to efficiently move heat from source to extremities. A fluid heat exchanger uses either a manifold to divide the heated fluid among several paths, or else a very long repeating path, so the fins can be made very thin and dense to increase the surface area. That tends to hold true in any sort of fluid cooling loop -- heatpipe designs, pumped water cooling loops, refrigeration cycles, etc.

Static pressure does increase with fin density; but the challenge isn't what heatsink is optimal for the fan. Rather, the fan needs to be optimal for the heatsink. There's also a question of which side it should go on. Nearly every fluid heat exchanger application I can think of that relies on an axial fan, uses the fan on the "pull" side rather than the "push."
Abacus Model 2.5 | Quad-Row FX with 256 Cherry Red Slider Beads | Applewood Frame | Water Cooling by Brita Filtration
Posts: 94
Joined: Tue Feb 14, 2017 11:58 am

Re: Optimal Heatsink Fin Density?

Thu Aug 15, 2019 12:57 pm

DPete27 wrote:
I presume there's a logical answer to this question (and I have my own theories) but: Why is the fin density on water cooling rads so much greater than on "air" coolers?.

There are two reasons.

1. Because, assuming the airflow is identical, the more surface area you have, the greater hear transfer. Radiators have limited depth available near a case air inlet/outlet (unlike tower coolers), so the thinner the rad, the more likely you are to get the sale, and make it fit in your case. They assume you'll deal with the higher noise and higher pressure necessary.

2. Tower coolers use heatpipes to transfer heat from the processor to the air. The higher density of heat transferred by these wonders of physics, combined with the much longer depth of fins available (plus the option for dual fans) means you can more easily maximize airflow, while still delivering more than enough cooling performance from the available fins.
Grand Gerbil Poohbah
Posts: 3346
Joined: Tue Jan 20, 2009 4:14 pm
Location: Los Alamos, NM

Re: Optimal Heatsink Fin Density?

Thu Aug 15, 2019 3:33 pm

It 100% depends on the type of air movement you're driving. Different fans for different purposes. High static pressure can push through tight fins and be more efficient in terms of heat removal. Low density fins help with low static pressure fans and are generally more quiet.

So there's no "right" answer or "optimal", it's totally dependent upon your method of moving air.
Desktop: X570 Gaming X | 3900X | 32 GB | Alphacool Eisblock Radeon VII | Heatkiller R3 | Samsung 4K 40" | 1 TB SX8200 Pro + 2 TB 660p + 2 TB SATA SSD
NAS: 1950X | Designare EX | 32 GB ECC | 7x8 TB RAIDZ2 | 8x2 TB RAID10 | FreeNAS | ZFS | LSI SAS
Gerbil Elite
Posts: 956
Joined: Mon Nov 16, 2015 10:30 am

Re: Optimal Heatsink Fin Density?

Thu Aug 15, 2019 3:56 pm

If you modify fin density and nothing else, you're picking an optimal amount of airflow (and less directly making a tradeoff between noise and peak cooling capacity). If the airflow to fin density ratio is too high, the whole of the fin surfaces see roughly the same low air temperature no matter whether the fan speed is a little too high or way too high, and you just can't squeeze any more cooling capacity out of it (modulo turbulence anyway). If the airflow to fin density ratio is too low, the air reaches the fin's temperature well before exiting the heatsink, and the extra density is just needlessly restricting airflow.

As defaultluser said, fin depth adjusts this. Most PC cooling would be better off being deeper, if the space were available, and liquid cooling in particular doesn't tend to do well on that count (meaning fins need to be spaced closer to get the same optimal airflow).

I would guess liquid coolers do bias towards higher optimal airflow than air towers. There's less potential benefit from passive or near-passive operation, since the pump is making more noise anyway, and there's more potential benefit from noise-be-damned all-out operation since you've got less effective thermal resistance between the CPU and an average point on a fin.
Gerbil Team Leader
Posts: 242
Joined: Tue Mar 01, 2016 9:17 pm

Re: Optimal Heatsink Fin Density?

Fri Aug 16, 2019 12:27 am

The limit to fin density for air cooling is not for performance reasons, as can be seen from the earliest extremely high-fin-density stamped sheet heatsinks like Zalman, which were much denser in places than anything sold today. I absolutely hated those things because while performance was pretty good for the time, at least at first, they quickly filled up with lint so you had to clean them every time temperatures started creeping up. The problem was they were nearly impossible to really clean (at least to as-new performance) short of removing them and washing them.

As radiators are often mounted with one side facing the outside of the case, if you employ draw-through cooling then the dirty side will be very easy to clean so denser is fine. If you use blow-through cooling then intake filters would be recommended, just as they were with Zalmans. That's because you would have to remove the fans from the radiator to clean underneath them.

Who is online

Users browsing this forum: No registered users and 1 guest