Porting School #4 - Budget Bench Electronics

[FlowSpecialist]

Originally Posted by DavidVizard-GFN

From emails I have had and from what has been said/asked in various posts on GFN it seems that some people have found out for themselves that the size or even presence of a dummy cylinder bore affect readings of the calibration plate.
DV

The exact details of the setup affect flow much more than people think. For instance the bore diameter and its ratio to the orifice plate diameter is an integral term in the flow equations which determine the Flow Coefficient of the plate.

As the bore diameter gets closer to the orifice diameter the orifice plate flows more and more until if you think about it if the bore and orifice diameter were the same it wouldn't be an orifice plate anymore, it would be a straight tube and those flow at near 100% efficiency rather than the nominal 60% efficiency of an orifice plate.

To illustrate here are the results of the full flow calculations including Reynold's Number and every other term for a 40mm orifice plate at 25" pressure drop in different bore sizes.

80mm - 164.8 CFM
90mm - 162.3 CFM
100mm - 160.8 CFM
120mm - 159.2 CFM
150mm - 158.2 CFM
1000mm - 157.0 CFM

In a very large bore the Flow Coefficient has dropped to 0.596 which is the minimum Cd for a square edged orifice plate. In an 80mm bore it's risen to 0.626, a 5% difference.

These are not small differences and there will be similar effects when flowing other shapes or cylinder heads themselves. Perhaps not to such a degree but it shows you have to bear them in mind. When I was calibrating my own flowbench 20 years ago I corrected everything for the size of the downpipe or bore adaptor that I test on. Not doing so could easily mean that otherwise accurate flowbenches could give results a couple of percent different to each other simply because they've been built a different size.

The downpipe on my own bench (which the head and bore adaptor sit on) is 4" i/d which is ample for the sizes of engine I work on. However a bench suitable for Chevy sized engines could have a 6" downpipe. That's a 1.6% difference in the flow figures of a 40mm orifice plate right there before we even get into things like manometer accuracy, location of pressure tappings or leakages.

In fact it's a miracle if you can get any two benches to agree within about 2% unless you take every possible precaution to eliminate variables in the flow setup and the bench itself or to correct for them in your target figures. I finally got my own bench to within 0.1% of the theoretical target flows but then I did take extraordinary pains to allow for every variable.

I'd also just like to note briefly that as regards your Helgesen plate, which is basically Venturi Nozzles in a thick plate rather than orifices in a thin one, there will be some element of this effect happening there too. The small holes won't behave quite the same as the large ones on a given bore adaptor and when you have several, or all of them, open at once it won't behave the same as when you are flowing one hole at a time. The magnitude of these effects is an unknown though in such a complex arrangement. Maybe the inventor looked into this in more detail.

Dave

[FlowSpecialist]

I should just add one thing I forgot to say about the specific way my bench is set up to deliver accurate results. The downpipe that the test piece sits on is 4" i/d as I mentioned previously and the manometer pressure tapping is at the very top of this. However the downpipe is not just open at the top like most benches. Over it sits a 1/2" thick wooden board with a 65mm diameter hole machined through it and a nicely radiused top edge to the hole. The top of the downpipe is then sealed to the underside of this board and the bore adaptor and head sit on the board over the hole.

The hole, effectively a streamlined venturi nozzle like in your Helgesen plate, is sized to flow about 700 CFM so massively bigger than anything I flow test. It won't therefore impose a restriction but what is does is collect the turbulent flow coming from the head down into the bore adaptor and present it to the pressure tapping just below in a more uniform way designed to minimise the effects of pressure recovery down the bore adaptor.

What I find with this is instead of getting bigger flow numbers due to pressure recovery if I test an orifice plate on a length of bore adaptor instead of directly on top of the downpipe I actually get the same or even slightly smaller ones.

This tells me that pressure recovery is being negated and hopefully the same thing is happening when I test an actual head. Whatever, I guess I could be kidding myself but set up in this way and with no further artifical corrections in the computer program the bench delivers exactly the predicted flow on every size of test plate I use on it.

It doesn't take long to make up such a board so maybe others would like to try this approach and see if it alters anything for them.

Dave

[Gary Collander]

If every man did that which was right in his own eyes.

You would get test results that where unrepeatable.

SuperFlow says to check calibration with the orifice bolted directly to the flowbench; I’m not sure how other bench manufactures instruct their owners.
Dave has gone a lot further in setting up his bench.

So this tells me that not only does there need to be an industry standard of flow @ 28” test pressure but an industry standard on bench calibration.

Also noted on your flow sheet with the head in question should include, at least, bore size along with temperature and barometric pressure for some sort of test repeatability.

I find few head porters ever complete the info on the form so you never know what changes have been made to the head before you receive it.

I also have yet to find a head that flows what the manufacture claims.

What other info would you find helpful in determining repeatable results?

Great Articles!

[Rick360]

Originally Posted by Gary Collander

So this tells me that not only does there need to be an industry standard of flow @ 28” test pressure but an industry standard on bench calibration.

There can't be an industry standard for calibrating flowbenches unless everybody's benches were built the same.

SF has you use the test orifice to "check" calibration (ie. it should be within a margin of error), but it is NOT to calibrate the bench. It was calibrated at SF and the range values placed on the cal/range sticker. I think it would be incorrect to "calibrate" a bench with only an orifice plate placed onto the bench opening.

Rick

[rookie]

Would the Flow Quick eliminate all these variables?

[brucepts]

Originally Posted by Rick360

There can't be an industry standard for calibrating flowbenches unless everybody's benches were built the same.

SF has you use the test orifice to "check" calibration (ie. it should be within a margin of error), but it is NOT to calibrate the bench. It was calibrated at SF and the range values placed on the cal/range sticker. I think it would be incorrect to "calibrate" a bench with only an orifice plate placed onto the bench opening.

Rick

Rick, If you recall we did a really non-scientific study about 2 years ago over on the flowbench forum on the "plate on top of the bench theory" and how close numbers from various commercial and homebuilt benches actually came. This included the "Mega $$$ Flowbench" at Ford Automotive.

I do not recall anyone that tested the plates having a Flowquick setup? I do know a number of people who have had that setup and now have built a cabinet flowbench and done away with it though.

There are a lot of flowbenches now being "calibrated" to cousins of those original plates around the world. This allows those users to compare their bench readings to the original set of plates, so in effect there is somewhat an industry standard being used already. Some well know head porters are also using those plates to compare their numbers. I know there is even a commercial flowbench product using those same plates for comparision now.

[rookie]

brucepts,
Which plates are you refering to?

[sir yun]

Originally Posted by FlowSpecialist

I should just add one thing I forgot to say about the specific way my bench is set up to deliver accurate results. The downpipe that the test piece sits on is 4" i/d as I mentioned previously and the manometer pressure tapping is at the very top of this. However the downpipe is not just open at the top like most benches. Over it sits a 1/2" thick wooden board with a 65mm diameter hole machined through it and a nicely radiused top edge to the hole. The top of the downpipe is then sealed to the underside of this board and the bore adaptor and head sit on the board over the hole.

Dave

Hi Dave,

the radius entry is only used when using the orifice plates right? ( it there with the bore adapter in place but not ''seen'' by the moving air then)

the setup sounds a lot like a shear plate/recovery combo used sometimes with 4 barrel carbs. but mounted backwards


cheers

[Rick360]

Originally Posted by brucepts

Rick, If you recall we did a really non-scientific study about 2 years ago over on the flowbench forum on the "plate on top of the bench theory" and how close numbers from various commercial and homebuilt benches actually came. This included the "Mega $$$ Flowbench" at Ford Automotive. .

Some flowbenches will show close to the "expected/calculated" flow when an orifice is flowed on top of it, others will not. Most of the benches in the test were "calibrated" already by your plates tested in the same manner which would give very similar results. The way a flowbench is built internally has a big influence on the air that will flow thru an orifice on top. How big is the plenum, where is the depression/pressure tap, where are the d/p taps, thickness of the top of plenum, diameter and shape of entry hole into plenum ... etc etc. Too many variables that CAN influence not only the reading, but the actual flow thru the orifice. A bench with a head on a 4" adapter distributes and slows the air significantly before entering the plenum, yet with a orifice mounted right on top creates a much higher velocity right into the plenum. What if someone has a pitot bench and the pipe IS the plenum, would it work then?

Why won't the same orifice inside different size pipes give the same flow/DP? Because the surrounding area influences the flow thru the orifice.

Originally Posted by brucepts

There are a lot of flowbenches now being "calibrated" to cousins of those original plates around the world. This allows those users to compare their bench readings to the original set of plates, so in effect there is somewhat an industry standard being used already. Some well know head porters are also using those plates to compare their numbers. I know there is even a commercial flowbench product using those same plates for comparision now.

A lot of people do a lot of things, that doesn't make it correct. SF provides an orfice plate to "test calibration" and they know how their design works with their plate to show there are no problems with a bench. You should never change your numbers on a SF bench based on the orifice "cal test". This cal method may work in a lot of benches, but not all. How would you know? Shouldn't something used to calibrate ALWAYS be correct?

Originally Posted by rookie

Would the Flow Quick eliminate all these variables?

No, it mostly eliminates measurement errors and speeds testing by eliminating manometer settling time.

Rick

[brucepts]

Originally Posted by Rick360

Most of the benches in the test were "calibrated" already by your plates tested in the same manner which would give very similar results.

Not true, even my bench was not calibrated with the plates at that point in time.

The Ford flowbench was not calibrated using my plates that's for sure. I guess this will be just something we will agree or not agree on. I do know what is working for a lot of flowbench users as it has been discussed at great lengths on my forum.

I'll go back to my part of the world now, I've hijacked this thread to much already . . .