KnightEngines wrote: ↑Tue Mar 23, 2021 6:03 pm
Max effort porting on OEM castings is something that cannot be done by cnc.
Cnc cannot see the core shift & adjust each port to stay out of the water jacket, a machine can't hear/feel a random thin section & decide that one can be a little smaller.
Not a 100% sure on that KE. This comes from two perspectives, one, porting OEM castings to a level that is economic for the casting [not eking out the last possible potential at a cost that makes it uneconomic against a performance casting or complete product] then scanning and digitizing the port, then running that as the NC program to port other OEM heads of the same casting can be economic. It can also get to a point where the cost of the CNC work exceeds the cost of doing it by hand, this is a decision we can make when presented with this. I have been there.
The other is [with the product i work on] the manufacturer produces a performance cylinder head with cast ports, these work quite well out of the box, but with the demand for more power from the same engine, the next level is to optimize the already supplied cylinder head,, so,, the manufacturer optimizes the port, digitizes it and puts this into a NC program and cuts the port.
What happened in earlier versions of the NC ported head was that all areas of the port did not have tool paths over them and had areas that were left cast. As we know the parts not cut were already out of cutters way that is, big enough already, running your finger over these areas showed they were not out by much at all. Next iteration from the manufacturer came with tool paths over the entire port, a measurement check showed they did not increase the CSA's but cast the port smaller in production so that there would be complete tool path coverage. My point is that the customers bitched their so called CNC head was not done properly in the early head as the tool paths had gaps, the later head was very popular as it looked sexy for no tangible gain, but with complete tool paths,woohoo!
The amount required smaller in the casting for this to occur? something like .5mm-1mm, the accuracy of casting design datum points to actual finished casting point location before machining these days is pretty good. It is a while since i have seen this manufacturer have head core shift variance of 1mm.
As you had correctly pointed out, breaking through is a consideration as well, but if we have been having a crack at the same component over years we identify where material can be removed and where it has to be left behind to insure integrity.
A while back a person took a finished exhaust port i did to task as not being optimized, he correctly identified areas that i could have easily picked up flow, what he didn't know is that this head can suffer loosening valve guides when excessive heat soak is experienced [ high ambient temp, high load, long distance touring air cooled] , so keeping enough supportive material around the guide boss is imperative, not the flow gain by removing it, and as you pointed out, you can't keep cutting, you will find air.
Cheers.