A kitchen hood, an industrial fume extractor or a laboratory fume cupboard — the question in all of them is the same: is the contaminant captured before it leaves the source and mixes into the room?
The reach of a suction opening is very short
A basic asymmetry of fluid mechanics: a blowing jet travels far, a suction does not. The influence of a suction opening decays quickly — at a few opening diameters away the suction velocity is negligible.
The practical consequence: bringing the source closer to the opening is far more effective than increasing the flow rate. Lowering a hood by 10 cm does more than going up a fan size.
Riding the thermal plume
Above a hot source there is a rising thermal plume, and it already carries the contaminant upward. A good hood design tries to meet that plume, not to work against it.
The plume's flow rate grows as it rises (it entrains surrounding air). If the hood is high, the flow that has to be captured is larger. That is the real link between height and required flow rate. See the natural convection note.
The real enemy: cross draughts
What most degrades capture efficiency is not a lack of flow but lateral air movement: an open door, a nearby supply diffuser, a person walking past, a window. A small lateral velocity is enough to bend the rising plume out of the hood.
For this reason the model boundary in a hood analysis cannot be the hood itself; where the supply air comes from and the general flow pattern of the room have to be in the model.
Modelling
Track the contaminant. Capture efficiency is measured by how much of a tracer released at the source leaves through the exhaust — judging by eye from the velocity field misleads.
Transient scenarios. Door openings and person movement are transient events and produce the worst case. See the moving boundaries note.
References
- ISO 14644-1:2015. Cleanrooms and associated controlled environments — Part 1: Classification of air cleanliness by particle concentration. ISO, Geneva.
- Churchill, S. W. & Chu, H. H. S. (1975). Correlating equations for laminar and turbulent free convection from a vertical plate. International Journal of Heat and Mass Transfer, 18(11), 1323–1329. DOI: 10.1016/0017-9310(75)90243-4