Flow bench calibration..with PTS orifice plates

A 50 and 100 CFM orifice.. and of course duct tape.. Essential, as turns out that a small leak affects the readings a lot.

The Audie Tech FlowQuik is pretty decent in terms of repeatability and as it turns out it is not halfway bad for ”absolute” values.

The original calibration orifice that comes with the FQ is a simple ”square”  edge orifice in a pipe with a 2D depression take off. If you use a different point in the system to establish test depression, you get a different depression value and hence a different CFM/CMM number. My numbers where consistently low at higher lift even though I was running out of ideas to improve it.

Just to have a more founded idea to see how far off  my absolute flow numbers are off the pace, I decided to buy a set of calibration plates. While it is possible to buy a set with a  NIST/DIN/MNi/ISO traceable standard, the are a little out of budget ( a several thousand euro’s). Therefore I got the next best thing. PTS plates. For me this maybe even better as it relates more to a few other head porters who did a lot of work getting the number right , among others a bench at Ford ( and I think they have NIST standard venturi for their calibration) as well as Reher Morrison. Bruce machined a set of plates to flow 50 , 100, 200 CFM @28 inches depression. (The original PAP plates where 100,200,300 but as an a-series is limited to 140 odd CFM if you are really lucky I opted for a 50-100-200 set as my bench will not flow that much)

The flowquik can be adjusted easily using a potentiometer.  So if your numbers don’t match your expectations it is very easy to tweak them to your liking.. you will of course be lying to yourself.. but still numbers don’t lie right..

Question is.. what actually happens when you twirl this dial?

In my setup I notice the following. As you can only shift the curve by means of the potentio meter, you are stuck with whatever it does. How it shifts (just up down or does it change the slope as well?) is not documented.

Ideally the bench should read the orifices as 50-100-200 cfm. Of course it does not.

If you calibrate the FQ with it own orifice its is actually almost spot on on the 100 CFM plate (100,1 +/-0,1) , the 50 cfm plate is then read higher and the 200 CFM plate is about 8-9 CFM low.

Ill have to retest that.. I actually put the plate on without recalibrating first..

 

If you set the whole thing to match the 50 CFM plate it  reads the 100 cfm plate as 94 (~6% off), and the 200 as 184.. ~(9% off and unacceptable ). Not too good then and well outside to specs  for the FQ.

Same but now with the 200 as a set point is gets quirky: the FQ has a build in correction to allow you to test without a fixed depression. The actual depression on my bench with the 200 CFM plate is not 28′ but only 16.5′ , not ideal but it is what is it for now. When the curve is shifted so that it matches the 200 plate it measures the 100 plate as 102,4 (~2,4%)and the 50 plate as 53.5 (7%).

As a best match the bench now reads as follows: 50=53.2  100=100.6 200=199.6.

The orifice in a pipe as supplied with the FQ ( which should be 80.9 according to audietech) measures 83,2 CFM when using the normal takeoff and 80,0 CFM using the takeoff on the pipe.

If you plot measured vs target , ideally the R2 would be 1 and y=1x. that would mean that the sensor is 100% linear and has no baseline deviation

 

 

 

If you look at just the upper three values, the sensor is actually very linear and absolute values are nearly perfect

I think I will have to do this for all set points and see which produces the best results. The upper values are very linear and nigh on perfect for intercept.

.

Tagged , , , , , , ,

2 thoughts on “Flow bench calibration..with PTS orifice plates

  1. great ,so my numbers are a bit low anywhere other than 80cfm to 100cfm ! well done joost , brilliant experiment .

    • mowog says:

      I’d say the low lift numbers are a tad high (about 0-60) and to high lift (70-200) numbers are two tads low if you calibrate with the vacuum on the plenum instead of the one the pipe. I should re do the whole lot and plot it for each set point (50, 80.9,100,200) and see what is the most linear and has the lowest intercept..

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: