1
$\begingroup$

This source describes in detail a formula used to calculate the airflow requirement in order to cool a computer system.

$$Q = \dfrac{P_{loss}}{c_p \cdot \rho \cdot \Delta T}\cdot k$$

Clearly, the higher the packing density (k) the higher the airflow requirement.

From the same source:

The k constant describing the packing density (concentration) of the components that prevent the free flow of air (k=80-95 rare placement, k=60 dense components)

k = 60 for high component density (ultra small pc)    

k = 85 for low component density (spacious full tower)

Meaning a small crammed PC case full of high-end components needs less airflow to be cooled than a spacious tower with the same high-end components. At this point, things make no sense. It should be like this:

k = 60 for low component density (spacious full tower)

k = 85 for high component density (ultra small pc)

60 = low, 85 = high

Do you think it's a typo?

EDIT

In this chart from the same source the airflow is lower with higher density components:

enter image description here

$\endgroup$
  • $\begingroup$ I think considering the mass flow instead of volumetric flow might be helpful. If the high density component build requires a higher inlet air pressure, the corresponding density of the inlet stream is higher - thus more mass enters the chassis. If we remember Q_dot = m_dotCpdt, then the higher m_dot value in this case means more heat can be absorbed from the electric components. $\endgroup$ – J. Ari May 10 at 12:49
  • $\begingroup$ Can you make this more clear? $\endgroup$ – Vulkan May 10 at 13:23
  • $\begingroup$ If you could clarify what is confusing, I will try to re-write my comment accordingly. $\endgroup$ – J. Ari May 11 at 10:06
2
$\begingroup$

Good question. It looks like you've found the linked article useful.

I don't have an answer for you other than to suggest that at lower component density some of the air doesn't flow past any hot components and so doesn't take away any heat. To compensate a higher air flow is required so k is higher for low density layouts.

The article is only 15 months old so I suspect that if you contact Endrich and ask for your email to be forwarded to the authors Claudius Klose and Zoltán Kiss that you might get a good response. Reference the questions here as they may be interested to know that they have a wider audience than the original publication.

| improve this answer | |
$\endgroup$
  • $\begingroup$ Well, this means ultra small PCs need less airflow... but what about obstructions in the airflow path? In console-like PCs the fan hits something within the first few centimeters. $\endgroup$ – Vulkan May 10 at 0:40

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.