Subject: [alfa] Re: Jim K's porting

[RonHorowitz <RonHorowitz@xxxxxxxxxxxxxx>]

Hi, Erik!

>
>That is kind of a blanket statement you have there,

Of course. Not to be flippant, but I would say the kinematic equations and
such hold in the general case....

>and may not actually apply
>to reality.

No hard feelings, but obviously, we disagree!

>The generality works if you use the typical very long duration, high
overlap
>cams that Alfa is well known for, where you need the momentum to keep the
air
>flowing into the chamber.

For a given CFM, you want to maximize the momentum regardless of cam
timing. If you understand what I'm saying, and you still disagree with me
here, I don't know what else to say!

>But if one uses the cams that Jim K tells you to use in the book, his
ports
>probably work quite well-

Well, Kartamalakis gives a lot of advice, some of it very good, but I would
say the ports work only acceptably well. They're certainly nowhere near as
good as they could have been

>his are higher lift, with relatively shorter
>duration.

The cam itself can't make the rest of the components any better than they
already are. The cams can only make a more nearly optimum set of
compromises for what you've already got. The better the ports flow, the
more benefit you get from cam redesigns and associated hop-ups. You want
the best ports you can get, no matter what.

>It all works as a system- need to match the cams to the ports to work like
you
>want.

You don't want the ports to work sub-optimally in any case. The
Kartamalakis port design, regardless of the relative merits of the balance
of his engine design ideas, is sub-optimal, but I'm getting a bit
repetitive.

>And good CFD software that will properly predict the flow in an
>individual port is very hard to come by-

What I was suggesting was that without a flow bench or at _least_ some
_approximately_ appropriate 3D FEA model, you're flying blind.

>the non steady condiditions are a PITA
>to work with.

True enough, except that even flowbench data is only quasi-steady state.
One often finds flow patterns toggling between two semi-stable flow
patterns at a given lift and depression. Never mind that the flow patterns
often stabilize differently (and give significantly different flow numbers)
depending on whether you flow the ports on the openning or closing ramps.
Nevertheless, there is a very high correlation between raising the
flowbench velocity for a given CFM, (i.e., making the port cross-section
smaller for a given target flow-bench CFM,) and increasing the power
integral of the engine on the dyno, improving the throttle response, the
driveability, etc. Note that this is _not_ the same thing as merely raising
the CFM without regard to velocity, which has been shown to have a negative
correlation (f'rinstance, some of the new Chevy heads, which make _more_
power the _less_ CFM they flow.)

>Especially all the back flow from the combustion gasses into
>the intake when the valve first opens, during overlap.

Here you've just contradicted yourself.

If energy is stored in the (mv**2)/2 term (which it is) you should be a
_proponent_ of high velocity ports. Ask yourself, from the standpoint of
conservation of energy, where does the energy go when high velocity gasses
are suddenly stopped up behind a closed valve? Does the pressure stay
constant as the velocity falls? Impossible.

>And I know a guy who DID use the Jim K book, as a system, and that car was
a
>ROCKET.

I wouldn't doubt that an engine built with features such as higher
compression, advanced ignition, oversized valves, aggressive cams, custom
headers, manifolds, air filters, exhaust, etc., would be faster than a
stock engine. Thatdoesn't mean it's as good as would have been had it been
done with optimal porting.

Anyway, if you like the Kartamalakis ports, suit yourself, but I wouldn't
use them and I wouldn't I recommend them. I hope my advice is usefull to
the group.

RON
--
to be removed from alfa, see /bin/digest-subs.cgi
or email "unsubscribe alfa" to majordomo@domain.elided