PDF

Lynn, have you experimented much with the boundary layer inside ports and manifolds? I have a friend who did a lot of race development with Mazda in the 70's and he got best results on the engine dyno with rough finish ports. He believes that if the port is a perfect shape then it will benefit from polishing but if it is not quite right air will bounce off the high or incorrectly shaped spots and ruin flow. But, if the port finish is a little rough the boundry layer formed seems to straighten flow and therefore effectiveness of the port. He said one of the well known gp engine builders from the 50's discovered this, it may have been coventry climax? I have a memory like a seive). Whoever it was ran 2 identical engines, one with polished ports and one with rough cast ports(due to a lack of time or something like that). The engine with rough cast ports was faster so they investigated, hence the boundary layer.



Personally I dont have time to mirror polish ports/runners and am quite happy with the results I get with a rough finish, what dictates a perfect port shape anyway?



If you read this Lynn what are your thoughts? (you may have heard of my friend, Bill Shiells, he used to build Rod Millens engines when he lived in NZ)

Cheesy

That makes perfect sense, the rougher surface will force a turbulent boundary layer which will stay attatched longer over adverse pressure gradients as is the case if the port shape is not perfect. I the port is shaped well a laminar boundary layer will stay attached, the laminar boundary layer is thinner than a turbulent layer so that is where some of the advantage of a laminar boundary layer comes from

Lynn E. Hanover

Sorry about the late start. I have been in Texas pissing away my wifes hard earned money on airplane parts,

and visiting long lost friends.







First of all..................everything affects everything.



Once you have th flow bench working, and you have some example pieces to play with, the rate of smarts induction will have you grining.



A little dab of clay here or there and there will be a change. Better in some cases or worse in others. A punch mark stood up and the fix works on this piece, but not on that piece. And so it goes on everything you test.



The problem is not with Nature. She gets the same result for the same input every time.



The problem is that we bumbling fools have so little chance of doing anything the same twice, by hand.

It is just not possible. By machine, yes, close enough to be called identical, but never by hand.



The saving grace is Nature gets the math right, so if we get it close it will act ABOUT the same each time.



So we look down some tubes in a manifold that have sand cast features inside and out. No two are alike.

It is a stock piece. It is designed to just work, and nothing more.



The rules stay the same for each piece but to see the flow in your mind, and say well right here there is a unmachined ledge sticking out, and you grind it back and flow goes down. ****, what now? weld it back in?



Glue some bits of thread along the inside of the piece you want to see and then add and remove some clay here and there to see what kind of shape change would make sense and improve flow.



If there is a problem, smooth or streamline around the problem and flow again.



Can a piece with just a bit of work flow better with very little work, while the same piece with a ton of work and all shiny not be worth a damn. Sure. It happens all of the time. Every rule applies every time.



Keeping in mind what rule applies to what situation is the key.



Sit down and play with the suction end of the shop vac. Make shapes with your hands. An increase in shop vac revs indicates poorer flow, lower revs better flow.



Better yet build a flow bench.



In some cases a rougher surface improves flow, by adding energy to the boundry layer. In another area of the same piece that same roughness may be too much and trip the boundry layer and reduce flow.



A polished surface on the inside of a turn is of little value. The "D" shaped runner is better. A bit of a VG upstream may be just the trick, along with the polished outside of the turn. Everything is different and requires a complete rework. So, clay and bits of string first. Then grinding and polishing.



The highest runner velocity attainable is just about always the goal. Bigger gets more total flow, and nearly always at a much lower velocity.



Keep in mind that the bowl shape has too much volume stock. So a nearly stock runner diameter with a high velocity will lose much of that velocity when it gets to the big bowl. So what if you hog out the runner, producing a low runner velocity, and it gets to the bowl shape?



For the same RPM (CFM?)



Godzilla way up high but a dog to drive around at lower revs.





Lynn E. Hanover

bill shurvinton

So are there known techniques for filling in and reshaping the bowl that will last in service? I assume that the right Devcon will do the trick, but obviously you don't want bits falling off.



Bill

Lynn E. Hanover

I have not gone to this area, because there is a SCCA rule that material may not be added.



It is probably going on, but not by me.



You could proof a system with Devcon plastic steel, and if there is enough of a difference, than do it with



conventional braze, or if more volume is required, a piece of steel sheet stock and silver braze. Some holes might be drilled along the edges of the sheet, into the iron to privide a more positive grip.



I doubt that there would be much stress in the intake. More from heat soak after shut down than anything while running.





Lynn E. Hanover