Notes on Workshop Techniques



3. Lathe Tools

3.6 Obtaining the best finish on the Workpiece

Obtaining a good surface finish is of more interest to the model engineer than to industry, mainly because the model engineer doesn't usually have access to cylindrical grinding equipment. I've already pointed out the requirements in terms of tool shapes and sharpness thereof, but perhaps I should expand on what exactly is meant by 'good finish'. A good finish may be merely decorative, it just needs to look smooth and shiney, or (more likely) the good finish has to be at an exact pre-determined size and is an attempt to emulate a ground finish.

In the first example it's possible to get an even turned surface using a pointed tool. I've read comments where it was stated a good finish can be obtained using a knife tool with a sharp point. With a slow, even feed the surface will indeed look good, with the light glinting off a nice shiney surface. In reality the surface will consist of a very fine thread with a helical groove the depth of which is a function of the feed rate used. If you don't believe me take a piece of fine emery paper and rub longitudinally (across the cut) and look at it again. Any measurement made of the turned diameter will be across the crests and not the troughs. There are circumstances when you might *want* such a surface as it holds oil quite well, and is also useful for Loctited joints. However, if it's a rotating shaft in a bearing both bearing and crests will wear leading to a less precise fit, and if the shaft is made from silver steel and subsequently hardened you will have quite an effective fine rotary file which will do your bearing no good at all!

The real trick is to obtain a truly smooth finish within a tenth or so of the required diameter. Such work is really in the realms of cylindrical grinding, but precision machining such as this is not too difficult - though it does take a methodical approach and is reached in stages. Having turned your workpiece to within about 0.01" of finished diameter bring your newly honed finishing tool into play. If you are working right up to a shoulder you have no choice but to modify the tool point so that the leading corner is square (or very nearly so). Its better not to have to do this as there will be a danger of gouging the workpiece, and it's preferable to undercut the shoulder instead. You should allow a little time for the workpiece to cool down if rough turning has heated it appreciably, and assuming you are working with tailstock support, check the adjustment of the center after it has cooled as the workpiece will have shrunk in size. You need good cutting fluid to get a good finish, either soluble oil or a straight cutting oil will do, but you need lots of it to flood the work. Figure on taking 0.002" off as your last cut so take a 0.003" cut now to check your math (!) but mainly to zero your cross-slide dial. Watch for the 'drag-cut' when you wind the carriage back to the start, this shouldn't happen with well-adjusted slides, a sharp tool and tailstock support, but if it does happen take your measurements after both passes. In fact, this is more of a problem when trying to fine-finish bores rather than shafts, a maladjusted carriage will twist inwards under drive from the leadscrew - and twist outward as the tool is traversed back towards the tailstock. In boring work this will cause a 'drag-cut' to be taken when the tool is withdrawn. A remedy is to mount the tool upside-down to cut on the opposite side of the bore. If your cross-slide feed is a bit flakey because of old age (the lathe's...) you might consider setting the top-slide over and using that to feed in at a narrower angle. If you set it to 30 degrees instead of 90 each 0.002" indicated infeed will actually equal 0.001". Back to the job - you just took 0.003" off according to the cross-slide dial, and a measurement with a good micrometer will tell you what you *really* took off (you might find it different from what you thought). You can make any compensation necessary and take the final cut, err on the big side as you can't put metal back on once it's turned off. As mentioned earlier, finishing cuts should be taken at an even, fast rate to avoid intermittent cutting, twice the carriage speed you would use for turning is about right, the finish being a result of tool shape and a keen edge (a bit like reaming really). To remove that odd remaining 0.0001", if you can measure it, you can resort to the finest grade of emery cloth, preferably well worn and dipped in oil. For a glass-like finish polish with chrome cleaner paste.

I should point out that, should you make your shaft dead to size and subsequently harden it the dimension is likely to change a little. Silver steel gets a little larger (we are talking 10ths here), yet I've also read that if you use the old 'soft-soap' trick to prevent scaling during the hardening process that the diameter can actually end up *smaller*. I've not investigated the latter effect but have been caught by having a shaft that was once a perfect fit not fitting the bore after hardening. If the shaft is to run in gunmetal or phosphor-bronze bearings you will probably not need to harden the shaft anyway, if it's cast iron bearings and you do need to harden then you might get a problem. If I was pressing the shaft into a ball-bearing sleave or suchlike I would aim to get the diameter a couple of 10ths undersize before hardening. Again, whether any of these effects are of consequence really depends on the class of work - it's no likely to cause you problems building the average loco, but IC engines would be different.

I don't like using carbide tipped tools for finish cuts, whilst the turned finish tends to be better than obtained with a roughing tool it does not approach that of a well-honed and shaped HSS tool (in my experience anyway). Very fine finish tools can also be made from hardened silver steel which in fact will provide a harder, keener edge - if somewhat less durable. To get the best finish possible with carbide tools the lathe should be run about twice the speed used for HSS tooling.

All the forgoing relates to mild steels of various types and also low carbon steels such as silver steel. However, working to the same degree of accuracy in tougher stainless or alloy steels might be a bit more challenging (understatement). For a start, I use different cutting fluid for these materials. I find that a high pressure cutting agent (like Trefolex) works well with stainless. If the workpiece is a substantial chunk of metal the standard finishing tool might not survive to the end of the job. You can try reducing the top and side rakes to produce a more durable cutting edge, or you may be reduced to using a carbide tipped tool - a new one kept only for finishing jobs, or perhaps a new insert, will give you the best chances of obtaining a reasonable finish. Watch that the extra tool pressure required does not spring the work. I have one anonymous piece of hexagonal stainless steel of about 1" AF, and even a carbide tool will hardly scratch it. There are free-turning varieties of stainless steel available so my best advice is to stay with these and leave the tough stuff to industry.

Photo shows (left to right) the finish achieved using the same tool on brass, silver steel, free-cutting mild steel and ordinary mild steel. The silver steel is not as rough as it might appear in the photo, but is a sort of semi-matt finish typically obtained.

Return to top

(c) Chris Heapy 1996.

Back to previous page