Replies

I don't get why they would need to be replaced in order to do metric.

Ah, this might be the sticking point. Sure, I can compute inch measurements for metric parts. But when I turn the knobs to the desired place, I will have to settle with the closest .001" and a wild guess for any .0001" of the computed result. It is common to guess the location of the .0005 position; and, the human eye can split things in half with remarkable accuracy. All bets are off if the final digits are something like .000146". So, the issue of making metric parts on English machines is NOT the arithmetic, it is turning the dials to unmarked positions.

If you had ever done any machining on a manual lathe or mill, you would know this. It is often essential for machine to be done to within .0001". That doesn't show up on machine dials, but surely does on micrometers. Those are expensive devils. In my humble shop, coming that close generally means using emory paper rather than the machine's tool point.

By the way, CNC machines are easily accurate to .0001", so they can handle the equivalences of metric measurements well enough. Still, many of those machines have position transducers made to English specifications; and, those babies are quite expensive.

Now here comes the most important problem: Machine power feeds. An Imperial lathe will cut threads based upon threads per inch. The only recourse for cutting metric threads using those lead screws is to use change gears at some multiple of 1:254, and the keep the carriage engaged for the return trip before cutting the next pass. That requires the machine to be stopped at the end of every pass, reversing the motor until the carriage is returned to the start point, and then once again reversing for the cutting pass. That is a lot slower than disengaging the power feed, hand cranking the carriage back to the start point, waiting for the threading dial to indicate the proper lead screw position, and then re-engaging the feed.