I knew that the siamese section was not properly shaped. As this was the first head I ever ported, I was a bit scared to hit water, something you really don’t have to worry overly about I have since found out. The shape was just off and kept a large part of the original MG bowl profile. The MG heads had a simple machining operation done to them improving airflow at minimal cost and labour. I was advised to leave it alone ( not a bad advice seeing that some hamfisted grinding can make things a lot worse) . Downside is that you are left with a weird lip and a convex siamese section. Vizard states that this region needs to be concave, and on later head I just did it. Now I had the chance to see what it does. I ground it out using my big grinder so a used the lightest possible touch at the lowest speed and a quite scary aggressive flame shape double chip breaker burr. The amount of material I took off is about a grams worth , so very very minimal. However it changes the shape significantly and is now much more convex and took out the weird shape left over from the original MG shape bowl machining.
The gains are surprisingly large, gaining all over the place. A result this good has to be investigated with a degree of skepsis. Maybe calibration is off, on the other had the shape look pretty contra productive and could well have upset flow for a large part of the valve circumference . Well, for now i’m happy
Never repeat a successfull experiment 😉
Well if results seem good enough to be suspicious… they usually are. Hence a retest.. and as suspected something was indeed fishy. The fact that the whole cuve had gained, just did not concur with the other 15 test done before showing very little effect at low list before gain started after 6mm lift. This is the retested version.still gain is gain. Why? I suspect the bore adapter shifted a bit, but it is still odd.
Interestingly the point where you lift the valve beyond the chamber flow refuses to go up . This could have something to do with the Cd. As suggested by Darin MorganI will try and velocity map the periphery of the valve and try to keep flow uniform around the entire valve during the whole of the lift cycle. I would need pitot tubes for that and some way to measure pressure differences or preferably airspeed directly .