How accurately can you machine?

I think many of us work to the precision of the measuring tools we have.

A contemporary example:

Since the advent of the ubiquitous 'wixey-type' angle gauge many folks now seem to work to 0.1 degree, whereas before you could estimate down to about 0.5 degrees with a protractor or work to much closer tolerances with a sine bar or dividing head. these little beasties have provided a quick and convenient way of working to a degree of accuracy somewhere in the middle with lots less bother and yet good enough for many purposes.

Neil

Well I once produced a 240mm parabolic surface to better than 50nm by hand (not machining I know).

regards Martin

I wouldn't dispute that the high tech industry is working towards non contact measuring with good reason to, it's just in reality, the actual operators of these super high tech optical/laser measuring devices have found that it isn't as convenient as the PR image suggests, there can be alot of variation between readings depending on your lighting rig and setting up a program to accurately measure with effective repeatability takes quite a long time. Having said that, once it's set up, it's done and you can just speed through all the measuring that would normally take a very long time.

It's a given fact that pretty much all day to day measuring tasks are still done with manual tools because, for whats needed, it's still effective. A computer needs instruction to carry out it's magic whereas i can just measure what i like without having a need to define parameters. I know we'd be up the creek without them and i'm pretty much beholden to hand tools and many industries are still finding it hard to let go of them.

Michael W

Edited By Michael Walters on 09/06/2016 15:33:31

Richard, you ask for thoughts on the two methodologies.

Some while back ME (or was it MEW) had a lengthy and very detailed article by somebody who had investigated tipped tools. His conclusion, arising from talks with tool manufacturers, was that these are designed for production and thus need to be worked hard - very fast with deep cuts and fast feeds, all of which are carefully calculated. Of course, it requires absolute certainty of where you will end up. Model engineers don't work like this, first Our machines cannot perform to the limits of production CNC machines, second, if you mic'd up an item and you knew that is was 2.6mm oversize on diameter, who would, with absolute confidence, wind in to take a final cut of 1.3mm in the knowledge that you would end up well within a thou of the required size? Not many.

We tend to approach the final measurement and thus the final cut is likely to be very small, much too small for a tipped tool. If you look at a tipped tool it is quite blunt compared with HSS tooling. Thus there will always be room for HSS tooling which can make very fine cuts.

Well, that is one view. On the other hand I went to a lecture at a model engineering exhibition where the speaker was extolling the virtues of being able to take very fine cuts with tipped tooling. He showed his results and the finish was very good.

Also, JB Tooling sell very shiny (they appear to be chromium they are so shiny) tips which are very pointed and very sharp. Jenny says that they are designed for aluminium, but in her experience they are excellent on steel, with the proviso that they are ONLY brought into contact with moving metal, otherwise the tip is likely to chip-off. I have used these and Jenny is right.

If I lived my life at my lathe, then I would get somewhere near, take a final measurement, and move in for the kill with a decent final cut (not less than 0.3mm). As it is, I will continue to get within 0.1mm and make this my final cut and I would use either a tipped tool (preferably one of the shiny ones, but a gold coloured tip would do), or a sharp HSS tool.

Charles

And I bet you tested it by a non-contact method as well

Neil

Good points, Charles. I have discovered for myself that the big issue with carbide is being willing to 'take the hit' of wrecking a number of inserts, because if you do you will discover how to get the best out of them with your machine. Unfortunately, unlike HSS a mistake it tends to be terminal, rather than meaning a quick trip to the grindstone. Also trouble usually means you need to speed up or increase the cut - which is not what you expect.

Neil

Just because carbide tips are liable to chipping, doesn't mean you can't sharpen the edges and reduce the radius to allow lighter cuts to be taken. The tips are made for production use, but for home use you can play around with them as you are not on the clock in any sense.

So buying cheap tips on eBay and experimenting with a grinder can produce useful results. My grinding wheels c from RDG were as cheap as chips, so again, no great cost involved.

Steve

I don't find carbide inserts prone to chipping. This steel conrod was machined from rectangular using a round insert at 800rpm, so a pretty severe interrupted cut, and the insert was fine:

The only time I've chipped inserts is by running the insert into something stationary when moving the saddle, like the chuck.

Andrew

Machining hardened steel ballnuts complete with mounting holes is a pretty good "interrupted cut" test. I used std general purpose (but "proper" Korly inserts and after 30 mins or so I'd removed a sizable flange at the cost of a few edges (ie 2 inserts, IIRC). Of course, there are grades of insert available that are specifically designed for prolonged machining of hardened steel with interrupted cuts but it's hardly worth the hassle of getting a few in specially for one job, given the cost of more general purpose inserts which work well enough to get the job done.

I now know enough about the myriad grades, coatings, chipbreakers and shapes of inserts available to recognise that the ones we see offered through the usual hobby channels are a very limited subset of what is available. On the other hand, I'm some way short of being able to offer an expert article to navigate through the subject myself. It would be great to see such an article but before offering to do so, I hope any would-be authors take the time to do some meaningful research into the matter, ideally speaking from professional experience rather than enthusiasm or (limited) opinion.

Murray