I have been beavering away but have so far not gained one iota after 8 mm lift it seems. Pretty frustrating all in all. That coupled to another suspiciously good result after some minor modifications, causing me to disassemble the entire bench for a checkup and finding a surprising amount of dust bunnies between the flow straighteners. Adding a pipe more straight pipe before and after the measurement tube, as you are supposed to have 30 cm up and downstream in addition the amount build in the flow element. While recalibrating I was getting pretty confused as where to source the depression. Audietech are of the opinion that you can use the normal vacuum takeoff in the plenum or the one on the calibration orifice. My main confusion is that when you calibrate on the test orifice you get a different result than when you take the depression in the plenum ( in my case situated quite near the top, flush with the 125mm pipe wall).
The test orifice is basically a big washer in a pipe. It has a flow straightener and a take off point.
What is the most sensible point to take the depression. According to what I have read in the industry standards white-papers the take of for a orifice in a pipe when used for differential measurement should be either very close or 2D away from the orifice. Audietech seem to have opted for the latter.
For now I have stuck to the manufacturers brief and left it where it is normally connected and calibrated to 80.9 CFM.
To be quite honest the stability of the readings are pretty much the same as before with the added pipe. My room is rather warm ( 29 deg C) I am pretty much at sea level (or slightly below), humidity is 65%, 1015 mbar… so on with the testing
But still no cigar.. despite having worked the siamese section a bit more and moving the port roof up a tiny bit and some reshaping some other areas I gained a bit of mid lift flow, but always hit the same wall after 7 to 8mm lift .. Then i thought.. maybe the manifold is the culprit. I changed the manifold and mounted a MG item I have .. that behaved weirdly as well.. it gains quite a lot when you Don’t add a radius entry to the intake where the carb normally is… okay you can now shoot me..
I decided to take to whole lot off and place my 12.7mm carb spacer in lieu of a radius entry and retest.
The ”absolute” CFM values being lower are due to the recalibration not being correct before and you have to keep in mind that this number is very dependent on a number of factors. Pitot benches are susceptible to weather and temperature changes thus you have to calibrate before each test session. Anyway these test where done 5 minutes after calibrating and directly following each other. It seems that the manifold ( minispares manifold medium size) and the head do not get along now.. or do they?
Then I looked at other data taken directly after calibration and the figures where about 5% down even after a fresh calibration ..MONSTER GAIN!.. Oh no ..PANIC! There is no way you gain that much in this port changing this much so something must be wrong.
I stared at a forest (50 +) of curves and and difference plots and it just struck me ( after hours of fiddling , rebuilding, retestingand getting generally confused ) that it was too linear.. it looked like some factor was off, as some curves just seemed to be offset. Then it dawned on me.. USER ERROR. I vaguely remember being disappointed as my numbers are way worse than published numbers done on a SF600 by Keith Calver (mini magazine expert series special edition) and wanted to retest at 25 inch depression (which would make it even worse)..And then I found the calculate one depression to another depression option and tried it out, but did not include a note in the test that I had done so on to tests . The factor from 25 to 28 inch H20 is 1,0583.
So what happens when you apply this factor..you get almost perfectly overlapping calibration curves. .. GOOD.. I feared that the electronics had started to develop a magical baseline drift
the gains I hoped to have produced are very minimal and after studying the difference in calibration … wait for it …..NON-EXISTENT.. hurrray !
What have we learned children.. well nothing I did not know before.. but just something you know you should do, but in practice I never seem to manage.. backup computers diligently, pay fines directly, never ever let stacks develop on any level surface. And start off with a standardised system of naming tests and adding what you actually did. Lucky for me I did add some other critical info so I could at least reconstruct where I had gone astray.
But airflow gain ? no.. not really.