I was pondering milling and routing tonight and wanted to collect my thoughts.
When selecting a tool one must ask, "What am I trying to cut, and what is my max headspeed?" Those are the two first questions on tool geometry, although just going by manufacturers recommendations normally guides a tooling purchase. Once headspeed is set, the feed rate will determine chip loading. Chip loading must be enough that the tool is not rubbing, but not too much that the machine or tool cannot handle the load.
It has to be assumed that workholding and toolholding rigidity is taken into account, and it turns into more art than science when learning. Ive found my turning center can't mill as aggressively in aluminum, it isn't as stiff as a regular mill since it is designed primarily for lathing. By listening to the cut, inspecting the finish, and inspecting accuracy, the machine happy zone was found. Because of the maximum headspeed limit of the current live tools, higher flute counts have to be used to prevent tool breakage while getting decent production rates.
For CF and other composites on the mini mill or a gantry, a diamond pattern (aka roughers or double cut) solid carbide routers seem to give the best results. With a strong vacuum on the cut, the burr is a good workhorse- my current favorite with plunge cutting
IMCO | SB-11 Cylindrical Shape End Cut Bur - 1/8" diameter - 1/4" shank - 1/2" loc - 2" oal - double cut . 1/8" size is my standard, and if no burrs are available I will grab an uncoated 4 flute and spin it as fast as possible. As long as the tool is kept cool it will stay sharp for a very long time. Freeze the panels in the deep freeze and cut cold for 2 to 3 times tool life increase. I was tipped onto this by another machinist and it works, the brittleness of cold makes the structure chip better. If we cut composites often I would build a gantry inside a deep freeze, supposedly if everything is cold the tool life and job speed can go through the roof. The limit is airflow to cool bit and pull swarf , and rpm to keep chip loading per rev proper as feed rate increases.
Conventional VS climb milling is also something to consider. The small taig is so flexible that I don't get advantages from climb milling unless it is merely a finish pass, and it is difficult to tell the difference. Machine flex changing the part size is more noticeable with climb milling. On the Mazak, finishes are easily noticeable as better with climb milling on the rough cut or finish cut. There is more material being removed on the Mazak, so the clean ejection of chips is more important.