Morning JW:
Level, no. But as one fellow on Practical Machinist pointed out, many machines on Navy ships are never "level" but still produce good results as long as the cutter relationship with the table is correct... so I am going to agree with that and say I don't think that is the issue.
Have done single pointing (fly cutter) and multi-point cutting with a 5 insert carbide face mill. Similar results.
The issue is a lack of parallelism, so yes, the machine basically cuts a taper / wedge shape from one end to the other. Thinking recently it may be stress relief in the material, but that doesn't precisely answer the issue.
I took a milling course once upon a time in BC, and I wish I could recall how well a "known good" machine would perform. I know, for example, a 3 jaw chuck is usually accurate within a couple of thou., so moving the work will always give you a bit of error. But 0.005" seems excessive. Half a thou I would be very pleased and think that exceptional.
Don't think that the parts are heating up enough to affect the depth of cut.
Typically use cutting fluid for finish and prevent chip welding, but again, the depth of cut and feed are not sufficient to generate enough heat. I generate way more heat in parts turned on the lathe and get closer tolerances.
Anyway, that is why I wondered if anyone else had experience / reasonable expectations for what a good mill should cut.
My mill is a Chinese Bridgeport clone, which had a hard life from all evidence. It was likely beat like a rented mule and probably poorly maintained. There is a fair bit of slop in the lead screw. I set the gibs to give a fairly consistent feel through the entire travel, but if there were excessive wear on the ways (to give a 5 thou difference of cut) I would expect to see that when dialing against the floor of the vise. Ditto if dents in the table were causing the vise to be jacked up when mounted.
I am starting to think it is operator error in workholding... going to play some more.
