Why We Didn't Put Any Fans In It
Share
Shipping a PC case with no case fans sounds like cost-cutting. It's the sort of thing a brand does to hit a price point and then describes in the marketing copy as "airflow-optimised."
So let's be clear about what happened: we tested the SuperAir extensively with case fans fitted, and the results were worse. Not marginally, not within noise — worse on average, and louder. So we stopped putting them in.
Here's why that happens, because the reasoning generalises well beyond our case.
Your build already has four fans in it
Before we talk about case fans, count what's in a typical SuperAir build.
A tower CPU cooler contributes one 120mm fan, sometimes two. The graphics card contributes two. The SFX power supply contributes one more. That's four to five fans already spinning inside 12 litres, each of them sized, positioned, and speed-controlled by the manufacturer specifically to cool the component it's bolted to.
The question was never "should this case have airflow." It obviously has airflow. The question is whether adding a fifth or sixth fan, positioned by us rather than by the component engineers, improves anything.
In a 40-litre tower, the answer is usually yes, because there's enough distance between components for a case fan to establish a coherent front-to-back current that everything else sits inside. In 12 litres there is no such distance. A case fan is a few centimetres from a component fan, and what happens next depends entirely on whether they agree with each other.
What goes wrong when they disagree
Three failure modes, all of which we hit during testing.
Pressure conflict
Two fans close together and not perfectly aligned don't add up. If a case fan is pressurising a region that a GPU fan is trying to draw from, the GPU fan's effective static pressure drops and it moves less air than it would have with no case fan at all. You've spent 3 watts and added a noise source to make cooling slightly worse.
This is easy to miss on a spec sheet and obvious the moment you measure it, because it doesn't show up as "the fan isn't working" — it shows up as a component running two degrees hotter than it did before you helpfully added airflow.
Restriction
A fan can only move air out of a case if the air has somewhere to go. In a small enclosure with limited open panel area, the restriction is the panel, not the fan. Past a certain point, adding fan power to a restricted case produces noise and turbulence rather than flow — the fan simply works harder against a wall.
This is why open panel area matters more than fan count in SFF, and why we perforated almost every surface of the SuperAir rather than cutting two fan mounts into an otherwise solid box.
Recirculation
This is the real one, and it's the failure that case fans cannot fix.
The dominant thermal problem in a small case isn't insufficient air movement. It's the same air being used twice. The GPU exhausts hot air into the space the CPU cooler is drawing from; the CPU cooler exhausts into the region the PSU intakes from; each component is inhaling its neighbour's waste heat, and the whole system's temperature climbs together.
Adding a case fan moves that hot air around faster. It doesn't stop the components from sharing it. The only thing that stops it is layout.
What we did instead
Two things, neither of which involves a fan.
We perforated nearly everything. Hundreds of ventilation points across the panels, positioned so that every component's own fan has cool ambient air available directly adjacent to it, and so hot air has an unobstructed path upward and out. If the restriction is the panel, remove the panel.
We separated the thermal zones. The CPU, GPU, and PSU each occupy a region with its own intake path and its own exhaust route. None of them is positioned to inhale another's output. This is the part that actually does the work, and it's a layout decision — no amount of fan tuning substitutes for it.
The result is that heat leaves the case the way heat prefers to leave anything: it rises, and there's an opening above it. The component fans handle getting heat off the heatsinks. The case handles getting it out of the box. Neither is fighting the other, and the only noise in the system comes from fans that were going to be there anyway.
In our testing this configuration runs cooler on average than a number of actively cooled cases in the same size class, and it does it in near silence.
The honest caveats
This approach works because of constraints we imposed elsewhere, and it's worth being explicit about them.
It works because the SuperAir takes a full-size tower cooler. A 120mm fan turning slowly moves plenty of air quietly, and it does the CPU's cooling work without help. A case built around a low-profile cooler has a much harder job and might genuinely benefit from assistance.
It works because we limited the case to dual-slot graphics cards. A card designed to dump 350 watts into its surroundings changes the problem substantially.
And it works because the panels are open. If you put the SuperAir in a cabinet, push it flush against a wall, or stack something on top of it, you have re-introduced the restriction we spent the whole design removing. Give it a few centimetres of clearance and it will look after itself.
The general lesson
The instinct in PC building is that more fans means better cooling, and in a large case that's often true enough to be a useful heuristic. In a small case it stops being true, and the thing that replaces it is layout — where the heat is made, and whether it has a path out that doesn't run through something else.
We didn't leave the fans out to save money. We left them out because once we'd got the layout right, they had nothing left to do.
The Magnova SuperAir is available in matte black. See the full specifications →