In Cooling Performance Investigation #1 I have found that, PSU fan facing the motherboard is more efficient in terms of CPU cooling than PSU fan facing the side panel. At the same I had a concern that, with both the CPU fan and PSU fan intake air from the same space but blow in opposite direction, they might end up fighting for air. It turned out to be either a non-issue, or its effects are small enough to not make a difference.

Today I want to find out that, if I provide a supplementary source of air, will it make cooling performance even better?

PC-Q07 is not designed to have a case fan. Neither am I going to drill holes on the aluminum panel just to fit a fan. My work around is pretty simple: put a fan in the ODD bay blowing downward to the gap between motherboard and PSU. Let’s call it the top fan. Yes, it’s a very awkward configuration, but I still wanted to give it a go because I think by provide more cool (supposedly cool because there is vent hole at the top of the case) air to both PSU fan and CPU fan, there is a chance that their temperature might drop a bit.

The problem is no one sells an ODD bay fan on the market. I had to make my own and it took me a while. My solution is these:IMG_1485– a 3.5” to 5.25” adapter

IMG_1489– and Scythe Ita Kaze Hard Disk Cooler. It is meant to be attached to a hard drive body (notice the four holes matching those on the bottom of a hard drive), but Scythe kindly provides holes on the side too, meaning  you can treat it simply like another 3.5” hard drive:IMG_1490

The end result is this, fitting perfectly in the ODD bay:IMG_1502

And after putting it in place: IMG_1523This picture should explain the whole idea. I expect this fan absorbs fresh/cool air from the top and feed to the CPU cooler and PSU fan (not shown here).

Test method is the same as before. I run Furmark to fully load GPU and wPrime to fully load CPU, and then record various temperature at the end of a wPrime 1024M run. Independent variable is whether the top fan is on or off.

I also added a second independent variable: whether the CPU fan is on or off. I have an obsession with silence so I try to do everything I can to make my computer quieter. If the system temperature is acceptable when fanless, then I’ll run fanless.

Top Fan Off, CPU Fan Off:Top Fan Off, CPU Fan 2000 rpm:Top Fan On, CPU Fan Off:Top Fan On, CPU Fan 2000 rpm:

The results are shown in the table below.

CPU/GPU temperature Fan Off Fan On 2000 rpm
Top Fan Off 90/78 68/74
Top Fan On 91/77 71/76

Clearly, the top fan is useless. Honestly I was really expecting it to make a difference. Well, it did. Just in a negative way. The only improvement shows up when the CPU fan is OFF, and GPU temperature drops by an insignificant 1 degree.

The other goal was to find out whether this system can run fanless. It turned out to be yes and no. I don’t think the processor will burn itself at 90 degree, but I don’t think it’s a temperature you want to see very often either. With these said, it’s worth mentioning that 90 degree is after running both the CPU and GPU fully loaded for about 10 minutes (~590 seconds). For text editting, web browsing, watching movies, it gets nowhere near that. Only under very demanding games (not only GPU-wise but it must be able to populate 4 threads, too) might I see 90 degree showing up. I don’t think my NBA 2K falls in that category.

It is also interesting to note that higher CPU fan speed helps GPU stay a little bit cooler, too.

Just as a sanity check, I performed the test when PSU faces outward, with CPU Fan Off and Top Fan On.

I had to stop the test midway because CPU temperature is soaring to 100 degree, 320 seconds or ~5 minutes after wPrime 1024M starts.