Mike Everman wrote:I think that's too much. I believe the acoustic properties of the intake will have a node .6r out from the inlet, true, but it will also have an antinode on the closed end of the cc. I believe the intake tube doesn't act as a true open end on the inside, but that the inlet to the closed end of the cc acts as a closed cylinder, where only odd modes are possible. Strangely, I think that all modes are supported by the whole tract because, while the exhaust and intake share an antinode at the end of the cc, the nodes needn't be perfectly coincedent, allowing the tail to have odd and even number of nodes.
On another note, if UFLOW isn't already correcting for the open ends and showing me real geometry, I may just open a vein!
All right, now we have Steve [the only one of us who is ready to butcher steel on this thing] waiting in suspense while we fight out where to make the cut. What a test of nerve and resolve! Hang in there, Steve ... we'll get it right yet!
This time, unless I'm misunderstanding what you've said above [and that is certainly possible], I think I'm closer to the truth. Here's why: The intake pipe in this type of engine has to work both ways
! Remember that Bill laid out two conditions for the intake: a length of L/5 and an outer node locus of L/3, both relating to the acoustic length L of the pipe as a whole. I think the premise of these two criteria is that the pipe has to behave itself in both modes -- its own characteristic open pipe [two node] oscillation has to blend with its operation as a closed pipe [single node] horn against the reflection of the front wall.
What that has to mean [I think!] is precisely this: The acoustic length
of the pipe has to satisfy the L/5 criterion, while the acoustic position
of its outer lip has to satisfy the L/3 criterion, within some fairly tight tolerance. We can surely do that.
NOTE TO BILL H: Please correct me publically if this is a misinterpretation of what you meant to state! We need to properly understand this.
I perhaps did not make it clear that my L/8 'derivative' is just a convenient approximation -- it does not serve us the same critical way that the Hinote Criteria do, since these are actually based on the way the harmonics fall into line in the pipe, while my L/8 is not. It is definitely these two that have to be hit within some [as yet undiscovered] close tolerance. But in saying that, I think we have to admit that the intake pipe has to function both as an oscillator in its own right and
as an element of the larger picture. And we can only do that by making it suffer the indignity of being chopped off at its open-pipe acoustic length.
Unless you can refute this reasoning on sound theoretical grounds, I have to let my claim stand for the radically shortened length.
NOTE TO STEVE: Looks like I owe you a drawing with corrected dimensions. Hang on ...
As to UFLOW concerns, I think that the correction is almost surely accounted for by the way it works in simulating the action of the real pipe. I can't imagine that we need to overlay a correction of our own on top of what it does. If that weren't true, it would be nearly useless in situations where the L/D ratio is low [short pipes]!