How do you calibrate a flow bench

[ole4]

It is straight glass tube.

DSC01071.JPG

DSC01069.JPG

Also the internal orifice on my bench is not a calibrated orifice. It is sized depending on the max flow I expect to get at 100% of the scale which is 36 inches long. I use the PTS calculator to size the internal orifice so it will supply 16" H2o at a rating slightly above the max flow I calibrate to at 100 percent of the scale. The critical part is the calibration orifice that is taped down the settling chamber in my case it is a 112CFM 1.25 diameter sharp edged orifice. The flow at 28" is 112CFM and I adjust the manometer angle to get it to 100% of the scale at 28" of depression. The most these little heads will flow is slightly below 100 CFM.

[gnash]

That is interesting and follows my testing/calibration observations that within any given test orifice range selection, after calibration by square root of the sum of squares, (How I define my flow test plate CD) my low end flow always indicate higher flow than the calibration plate and my top end flow always indicate lower. I guess I have to wonder why this isn't addressed by the DM software. Does anyone use software with a single plate system that addresses the CD change with rising test pressure across the flow element?

I use Excel for doing conversions and stuff like port energy. Some of the electronics available have some kind of capability to send your testing data to Excel spreadsheet. I use the FP1 from Flow Performance. It does a good job working alongside Excel to do my conversions when I am using my flow bench.

[302ford]

I would love to hear from Erland on this subject

[Erland Cox]

I have 4 different calibration plates 67,7 - 191,1- 276,3 - 395,3.
That is what they should flow cinsidering their diameter.
My flow bench has a big settling chamber and after that a tube with pitot pipes in it.
I have tested the plates with the biggest flow cylinder that fitted the plates and sealed over and under the plates.
Then I have tested them in both directions.
After that I changed the flow constant of my bench a little to agree better with the plates, I lowered flow a bit if I remember correctly.
All plates are within less than 1% if their value in both directions.
All heads that I have tested flow less than claimed except MBE:s P/S head for Chevrolet which flowed exactely what MBE claims.

Erland