Comment by ninjagoo
1 day ago
> we have implemented a tip clearance of only 0.5mm (120mm models) or 0.7mm (140mm models) in order to minimise leak flows through the gap between impeller and frame.
> Achieving such small tip clearances is essentially at the absolute limit of what injection moulding can consistently reproduce
For folks thinking about Lego tolerances [1] that are an order of magnitude tighter at 10 microns or 0.01 mm, it turns out that the largest Lego moving parts are a turntable at 50mm or so, and rotate at an rpm an order of magnitude slower (100 or so rpm vs 1200 rpm), so these tight clearances (not tolerances [2]) are quite tricky to achieve, and more importantly, maintain over the life of the product, apparently.
[1] https://en.wikipedia.org/wiki/Lego?#Design
[2] changed 'tolerances' to 'clearances' per note below
So if I've got the right idea, the clearances harder to achieve for a fan vs a lego piece because you're not just concerned with the static tolerances of the shape of the fan, but also the dynamic forces that will make the blades flex and bend under load.
Clearance in this case is how far away the blades have to be at rest, such that the dynamic forces the blades experience under load won't flex them outwards to the point they scrape against the enclosure. Which I'd assume has far more to do with material properties than it does the raw geometry of the blade.
Now I wish I had a high-speed camera to be able to inspect the dynamic deformation of a noctua fan. I'm curious about how rigidly they behave under load.
The plastic will also shrink and grow depending on its temperature (yes this will have a significant impact over the normal temperature range of the inside of a computer).
0.5mm is the clearance, not the tolerance. Tolerance must be significantly smaller than clearance.
yes, good point. fixed.