[SFF Network] SFF Network's Thermal Test Rig

confusis

John Morrison. Founder and Team Leader of SFF.N
Original poster
SFF Network
SFF Workshop
SFFn Staff
Jun 19, 2015
4,129
7,057
sff.network
Ever since we started reviewing CPU coolers, we've had a dirty (not-so-much-of-a) secret - namely, that our testing has been done on an ATX board. The reason for this was practical enough: the board itself is a part of my main workstation, and I've been the one responsible for cooler reviews at SFF Network! However, as we grow into a fully-fledged technology and review site, we decided it was time to build something dedicated, and more fitting to our niche of SFF systems. Thus, in today's article, we will detail the new system, the build process, and the retesting of all our previously-reviewed coolers.

Read more here.
 

Phuncz

Lord of the Boards
SFFn Staff
May 9, 2015
5,827
4,902
Crazy thought, but if SFFN opened a seed money / donation thingie, you'd probably get enough funds for an Intel board as well. Since SFF is >90% Intel due to the lack of choice in AMD mITX boards. If shipping was cheaper to NZ, I would have sent you some hardware already XD

A Z87/Z97 socket 1150 board with an Intel G3258 would have been another great choice, since it's low cost and as I've proved in my overclocking endeavours it can make a Noctua NH-L12 seem overkill and inadequate.
 

confusis

John Morrison. Founder and Team Leader of SFF.N
Original poster
SFF Network
SFF Workshop
SFFn Staff
Jun 19, 2015
4,129
7,057
sff.network
The second choice of platform was indeed a Z87 board and a G3258 - the cost was that much higher thoughy :/ AMD APU was easiest way to get the 65w heat for base level testing :)
 

iFreilicht

FlexATX Authority
Feb 28, 2015
3,243
2,361
freilite.com
Initially I felt like letting the PSU face the CPU cooler was a bad decision, but as quite a few cases require that, it might be very interesting for people to see whether the CPU fan is actually strong enough to work in that environment.
 

confusis

John Morrison. Founder and Team Leader of SFF.N
Original poster
SFF Network
SFF Workshop
SFFn Staff
Jun 19, 2015
4,129
7,057
sff.network
Thats just a test setup - the coolers will be tested open air to remove any chassis variations, as well as height limitations.
 

confusis

John Morrison. Founder and Team Leader of SFF.N
Original poster
SFF Network
SFF Workshop
SFFn Staff
Jun 19, 2015
4,129
7,057
sff.network
The decision was also made when i cut myself numerous times building the demo system.. The Jonsbo C2 is as sharp as a straight razor in some places :/
 

EdZ

Virtual Realist
May 11, 2015
1,578
2,107
Thats just a test setup - the coolers will be tested open air to remove any chassis variations, as well as height limitations.
Given the normal use-case of SFF, a 'restricted test' rig may be a good option. Something like a set of walls around the ITX board perimeter, possibly excluding the PCIe slot (to give a nice fixed minimum design-case), with a sliding perforated/mesh 'lid' that can be raised and lowered over the test bench. This lets you test coolers with a variety of clearance heights between the fan top and the case wall, and accommodate coolers of different heights that could not be done with a full case.
This would primarily be a test for top-flow coolers. Cross-flow is fairly rare for SFF due to the greater size, and usually entails some case-fans to provide airflow, or at least custom ducting, so performance is going to be more case-dependant anyway.
 

confusis

John Morrison. Founder and Team Leader of SFF.N
Original poster
SFF Network
SFF Workshop
SFFn Staff
Jun 19, 2015
4,129
7,057
sff.network
Given the normal use-case of SFF, a 'restricted test' rig may be a good option. Something like a set of walls around the ITX board perimeter, possibly excluding the PCIe slot (to give a nice fixed minimum design-case), with a sliding perforated/mesh 'lid' that can be raised and lowered over the test bench. This lets you test coolers with a variety of clearance heights between the fan top and the case wall, and accommodate coolers of different heights that could not be done with a full case.
This would primarily be a test for top-flow coolers. Cross-flow is fairly rare for SFF due to the greater size, and usually entails some case-fans to provide airflow, or at least custom ducting, so performance is going to be more case-dependant anyway.

That sounds rather complicated to engineer reliably but it is something I will consider. In the meantime, I will be testing in a 28c ambient (thanks summer -_-). The problem I see with an adjustable box is the changing dynamics of airflow inside the 'case' - the air inside a 5L 'case' will heat up much faster than a 10L 'case', which would skew results in favour of the bigger cooler even further - which would lead to somewhat unreliable comparisons if one chose to use, say, a NH-L9i/a inside a Coolermaster Elite 110.

I could test a cooler at all panel heights available to the adjustable design, but that may lead to an article with so many data points that it is difficult to discern what results to base an opinion on when looking to purchase a cooler.
 

NFSxperts

SFF Lingo Aficionado
Aug 7, 2015
112
53
The decision was also made when i cut myself numerous times building the demo system.. The Jonsbo C2 is as sharp as a straight razor in some places :/

You have to wipe a cloth over all the edges before working on it to dull the edges.
This case is good for testing thermal limits because if you have gpu in that case, the heat generated will be trapped inside the case.