RevTheory wrote: ↑Fri Mar 19, 2021 11:37 am
I've seen these types of arguments play out probably hundreds of times. I'm pretty sure that the guy who's been behind a flowbench for 30 years and was really good at it won't listen to a guy with an expensive computer program and vise-versa.
A brand new engine platform, absolutely! Something already known that needs to be optimized for an application, give me a guy who's really good with a bench and probe.
Either way, the wheels on this particular bus will continue to go round and round.
I see where you are coming from. But it does not have to be an either/or situation.
To me it makes sense to use every available tool in order to accomplish more with less time.
Cost does not matter: time is the only non-renewable resource. Something every racer knows.
Empirical testing, no matter how it done, will not provide an understanding of the governing principles.
A low pressure bench and probe will never provide the real insight necessary to make serious
improvements, only incremental changes.
Empirical testing usually involves one segment or function at a time.
But engines are a series of inter-connected sub-systems, each of which must be optimized.
There is no such thing as changing only one thing at a time. Every small change results in
numerous downstream effects, many of which are not obvious. Especially thermal effects.
Simulations of connected components provide a better insight than testing of each separately.
Improved port flow is only a small part of the solution. Upstream manifold design, valve seat
design, cylinder and piston crown design, combustion chamber shape are equally important once
the total system is understood. Prototyping of new designs is expensive, time consuming, and
limits the number of options that can be examined. Bench or dyno testing still require prototypes
to evaluate.
If a CFD model of an existing design is constructed, it becomes possible to quickly explore the
available possibilities and limits of multiple new designs. Once we know what air (mass) flow
really looks like, and where the constrictions, turbulence, separations and re-circulation areas
actually are, we are better prepared to make effective changes.
Progress based on empirical testing is a slow, shallow sloped, process. There is very little that
we know about engines that was not known almost 100 years ago.
I am much too old to wait around while we take baby steps forward.