Gas flowing cylinder heads

[zefarelly]

I know this is a 'dark art' but are there any publications and or guidlines . . . I've played with a few precrossflow Ford heads now and I'm having themn tested on a flow bench to see what they do, I'm also testing a std GT head to check Im not going backwards !

I guess there must be a fairly simple calculation to determine a maximum flow though a fixed dia bore, but athese heads for example have a far from perfect inlet/outlet tract !

any comments, opinions appreciated

[Revracing]

You could set up your own flow bench based on water flow, if you could arrange it so as the port being tested had the valve fixed in it's maximum lift position, then if you timed the flow of a known quantity of water through the port then that would serve as a very good indication of the improvements or not gained by your work. You would have to test it before any work was done of course.

In this way you would be able to balance each port as well.

It does work and although you wont get a CFM reading once you're finished you could get it tested by a third party and that would give you a benchmark to work to.

As far where to remove metal and where not to, that is a very difficult one to answer, but if you concentrate on the outside radii of the exhaust port and the inside radii of the inlet it would be a good place to start

[zefarelly]

thanks . . .Id not thought of water flow . . .although it wouldnt behave the same as gas, it would be a good indication of capacity.

my local engineering shop has a proper flow bench . . . .when I find out how much wonga he's going to relieve me of Ill decide how much I'll use it !

I have 2 heads modified by others and half a dozen to start playing with . . . .

[imull]

we were told that air cannot flow faster than the speed of sound and by using this and the minimum diameter of the inlet tract, a threoretical maximum cfm can be found.

obviously it assumes that flow is perfect... dont know how accurate the statement about speed of sound is but our lecturer said it was all he used as an engine designer for Ford once upon a yesterday...




if there is a uni near you with a half decent automotive or plain engineering faculty, they will probably have a flow bench.

[GolddustMini]

there are litterally hundreds of books on engine tuning, many of which are excelent at the principles etc behind gasflowing, have a look on amazon

[imull]

especially fords

[GolddustMini]

yup although we found one book that advertised itself as having lots of information on xflow heads, the entire information was "open ports to circular in shape and put 3 angle seats on it...

erm yeh! loads of info there

[thebear]

Originally posted by zefarelly
I know this is a 'dark art' but are there any publications and or guidlines . . . I've played with a few precrossflow Ford heads now and I'm having themn tested on a flow bench to see what they do, I'm also testing a std GT head to check Im not going backwards !

Early MG & Sprite heads could be improved by reducing the profile of the valve guides that extended into the `stream'. However, if you remove too much you will be replacing guides and possibly valves very often due to excessive oil consumption and insufficient lateral support for the valve stems.

[avsfan733]

Originally posted by imull
we were told that air cannot flow faster than the speed of sound and by using this and the minimum diameter of the inlet tract, a threoretical maximum cfm can be found.

Basically right thats one way of going about it, but a better way would be to figure out the cfm you need to flow by calculating your cylinder volume times rpm, then figure out your port area so that there is no way the flow can exceed the speed of sound at the redline with a reasonable safety margin (the same thing has to be accounted for in turbocharging, the tips of the compressor can't exceed the speed of sound or flow will stall, reverse, back up and destroy turbos). YOu also need to account for the opening into your intake manifold if you are pushing port velocities that high. It would also be very wise to consider using the velocity of that airflow to some good in mixing up your cylinder and getting the charge flowing equally around the cylinder. Air at that velocity has a decent amount of force in it and will kick around create a lot of turbulence and give you a better burn which is more efficient and will gi ve you better gas mileage and power. Remember that the number that is your minimum port area is FOR THE SMALLEST POINT ON THE RUNNNER otherwise while it may not hit sonic flow at the ends, it may hit it at other points and bends. like the guys have said though, a book is the best way to go as it provides a lot more info especially if it is specific to your application as it can provide a guide to angles and finishes and curves and radii. If your doing it yourself, be very cautious lest you destroy a cylinder head...maybe best to have someone who knows the art/craft show you before hand, and practice a lot with the air grinder.

Best of luck

[imull]

Originally posted by avsfan733


(the same thing has to be accounted for in turbocharging, the tips of the compressor can't exceed the speed of sound or flow will stall, reverse, back up and destroy turbos

I didnt know that. Learn something new every day.

[zefarelly]

thanks Avsfan733, I have 2 ported heads, one with GT valves, one with Lotus twin cam valves, and a std GT ford head, theyre all on the flow bench at my engineers for comparisson . . . .based on the results I'll decide which way to progress.

I was looking at the set up he has and some comparissons . . . a pair of Lambo heads flowed 9% better just by recutting the seat angles . . . .they wont flow anymore if you increase valve lift, and they flowed approx half as well as a good pug 16v head, fascinating stuff . . .Ill soon probably find out that my first attempts have reduced efficiency !!!