I did my GCSE physics project on Reynolds' number, many moons ago. The actual experiment itself was simple, but the hard bit was creating several-meter long hol

I did my GCSE physics project on Reynolds' number, many moons ago. The actual experiment itself was simple, but the hard bit was creating several-meter long hollow glass syringes in order to inject die into the flow tube well into its flow, in order to avoid turbulence where the header emptied into it... ended up dangling borax glass tubes off the roof with a 1kg weight lashed onto the end, and then blasting the centre with a blowtorch - serious fun. Anyway, long story short, it's a really interesting experiment to run, Reynolds number (for water, at least) can be deduced by a teenager with about £100 of equipment, and the boundary between laminar and turbulent flow is a mysterious beast indeed. One interesting artefact was inducing turbulent flow in speeds "too slow" for turbulence, by increasing flow to the point where turbulence triggered, then decreasing flow to the laminar realm again, yet retaining the turbulence indefinitely. Interestingly, this low-speed turbulent domain required a smaller throttle on the header tank to maintain the same flow rate - i.e. the turbulence decreases resistance to flow in the tube. Something that pipeline engineers could/should/maybe do employ. Had to pay for a new floor at the end of it all, as there's no tidy way of running this one.