Slightly OT - pipework for air.

You appear to have answered most of your questions yourself, as regards the throttle, restrict the output, because the fan is designed to give a particular output at a known speed so if you slow the motor the fans characteristics will change but not relative to fan speed.

Martin P

The blower motors on blacksmiths forges were always variable speed. Mark.

I managed to find an answer to q.1 here:

I think my main question, really, is: will I damage my blower by reducing the pipework so drastically?

Unless it's feeding the forges of (insert deity)'s hells, the blower won't be big enough to be damaged by stalling the airflow. You'd notice if things were turning bad because it'd get very noisy. Most likely scenario is that the motor load will reduce somewhat due to the reduced airflow (same reason as the Hoover speeds up when you choke it and stall the fan, less air==less load)

Mark Rand RTFM

Not really. Counterintutively perhaps, for a reduction it does not matter much but an expansion does. You could use a steeply angled cone with a good radius where it joins the smaller pipe will be the most efficient. Going from small to large is quite different, and a 'diffuser' with an included cone angle of about 6deg is the most efficient (about 18 inches long in this case).

Faster & lower pressure in the 2". Flow will be reduced because the 2" pipe will have more resistance than the 4", so ideally you should do the reduction near the downstream end. How much the volumetric flow rate will be reduced depends on the flow v. pressure curve for the blower.

If it were a liquid pump then you would not throttle the inlet as that could cause cavitation. With air I would say again on the outlet for the same reasons as talking about expansions & contractions. If you can control the motor that is probably the most efficient way to do it.

HTH.

Why is that I wonder? I'd speculate it might be something to do with vortex generation around a sharp trailing edge at a step or too steep change in diameter, which would cause a transition to a more turbulent flow which I would guess must be less efficient than a laminar flow. Is that reasonable?

Nobby

For the contraction, as Charles says, the only critical bit is the 'corner' going into the small pipe. If it's sharp then the streamlines 'overshoot' and breakaway at the corner. This leads to a reduction in the apparent useful flow area, smaller than the small pipe. Do a search for 'vena contracta' for more.

In a similar way, going from small to large, the streamlines can break away from the corner, exiting the small pipe and then as you suggest cause turbulence which causes much more drag in the (larger) pipe. Again a radius at the transition from the small pipe to the taper is far more critical than at the big end of the taper. The figure of 5~7 deg is an empirical one, CFD can be applied to individual circumstances if you have the inclination/money to do so.

For a 'nice' and reasonably simple design consider using a NACA duct profile, but note that for a circular pipe you need to consider the cross sectional area and thus modify the NACA duct coordinates appropriately. When I say 'simple' I mean from the design perspective, tin-bashing a HVAC duct to a NACA shape might be a bit of a challenge.

Richard

Well, you're supposed to learn a new thing every day, I can stop now for today!

Thanks, Nobby

You're lucky, I am several years behind target :P

bigegg submitted this idea :

If the motor is a none synchronous type, throttling the input can cause the internal fans blades to run in a partial vacuum and with a reduction of the load, the motor speeds up.