This came up on another forum that I am a member of. I thought it was very interesting.
"Following up on the CFM measurement conversation in the FB510 thread, but I thought I'd start a separate thread..
5HP Clearvue with 6" PVC. Measured 35' from collector, with three 90° bends and two 22° bends in the path (plus a couple wyes in the path that are closed off). I know there are few joint leaks, but I don't think there's a lot of leakage. Maybe later I'll try to tighten things up and see if affects readings.
I see ~950 CFM (4750 fpm) at the 6" port.
When I restrict that to 4" by adding a 6"x4" reducer, I get ~700FCM (8000 fpm).
When I add a 6' length of 4" flex, I get 685 CFM (7800 fpm)
I am using a $30 Pyle PMA82 Anemometer, which I "calibrated" by holding out the window of my car. At various speeds it is consistently ~5% under the car speedo reading, so I think it's reasonably accurate. My duct readings are a rough average of about +/- 15% variation I get depending on exactly where I hold the anemometer.
My system is surely not CFM optimized, with 35' and 320° of bends before the first port, but it's space and convenience optimized -- the collector is tucked away in a corner and the ductwork is underground."
I don't know how accurate these results are, but this experiment shows 26% reduction in CFM by a single 6" to 4" reducer 35 feet from the collector with multiple bends and turns. I am currently looking into getting a gauge/meter to make some measurements myself.