Thread: Some interesting air flow measurements

Well now...those bend results are quite interesting.

The article makes the excellent recommendation that

A reminder that dust extractors with coarse bags should be located outside the workspace or in a cabinet vented to the outside to prevent the dangerous fine air from escaping and being drawn back towards you by the vacuum created at your machine.

but it also suggests that

we'd like at least 350cfm in a 100mm duct system, but levels as low as 250cfm will still work for light dust. We should aim for 400cfm.

These two recommendations are somewhat contradictory since 400 cfm is totally inadequate for grabbing fine dust at source of the dust production. Flow rates of 800 to 1000 cfm are needed to do this.

350 cfm should be the minimum flow rate in a 4" duct since that is where the linear air speed reaches 4000 fpm, below that the dust will not easily stay in suspension and will clog ducts. At 250 cfm the linear air speed is now 2800 fpm which is now well below the 4000 fpm limit.

I contacted them and have had several email exchanges with the guy (Terry) that did the measurements. Terry is very technically minded (especially in electronics) and has been most helpful in expanding on what he did. He has tried to be as careful as possible and knows generally there are limitations with his measuring method but was not aware of the extent of his over estimation of the air flows.

The main problem comes about because he is using a fan type air flow meter that is 3" in diameter. When the 3" fan is placed in from of a 4" duct it partially blocks the duct which increases the air speed into the duct and through the fan. This then over estimates the airflow.
The other problem with a 3" fan is it cannot get close enough to the wall of the duct to accurately measure the air speed at these points in the duct. This also over estimates the speed of the air flow.
He is also not using a proper test duct.

I don't want to be critical of Terry as he clearly has done his best but these results are a common problem with DIY air flow measurements. Even worse is DIY particle measurements where variable backgrounds are not taken into consideration and can really mess up the measurements.

This is a common problem everywhere.

Good data is rare. Crappy data is common. Even good data have limited information content. It is necessary to understand how the data were collected so we know its shortcomings ... all data sets have shortcomings. Another trick is understanding what info is not in the data as well as that info which is in the data.

Epistemology should have been one of the most important courses of study we took ... but most people don't take it.