Home lathe precision question

Your average home manual lathe of reasonably quality can probably repeat to +/- 0.0001 or better, but this does depend on the skill of the operator. Notice I said ''repeat'' rather than ''accurate to'', the accuracy is dependant on the operator also.

Jim, I am thinking you have an extra zero in the +/-0.0001 number. +/- 0.001 would be closer, I believe. Most of us cannot even measure to repeatably a tenth, let alone machine “better” than that. Please clarify?
 
Jim, I am thinking you have an extra zero in the +/-0.0001 number. +/- 0.001 would be closer, I believe. Most of us cannot even measure to repeatably a tenth, let alone machine “better” than that. Please clarify?

Depends on what I'm working on, but typically my target is +/- 0.00000''. That's pretty hard to hit consistently, but most people can learn how to hit and measure +/- 0.0001. It just takes practice and some dedication. I got used to measuring to these tolerances back in my old automotive machinist days when building high performance engines. We had measuring tools that would measure to 0.00005 so I had something to compare hand measurements to, so it was just a matter of practice to get consistent results. This is normally only important with bearing fits, or other sliding or press fits so you are trying to hit those targets over short distances which is much easier than trying to hit that target over a foot or so. Although I did turn a 9.5 inch long 9/16 inch diameter aluminum shaft to 0.0001 end to end a few weeks ago, but the only critical dimensions were the 12mm bearing journals on the ends.

Most of the time on a critical dimension I turn the part to within 0.001'' and finish with a file and emery cloth.

I have mics that will measure to 1 micron (0.000039'') resolution, but the absolute accuracy is more like +/- 5 microns. Now I have to say that this is a relative measurement rather than an absolute measurement because I have no environmental controls in my shop. This means that the parts are smaller on a cold day than on a hot day. Also, after making a cut you have to let the part cool down to ambient temperature before measuring else it will measure large.

My lathe is nothing special, just a 13x40 Jet, Taiwan import. But I do have it dialed in properly and it's in good shape for a 25 year old lathe.
 
So, what kind of precision can it do? Can it do piston pin level fitting? I don't know what the gap is actually. But things like that, or injector plunger. All very smooth and tight fitting, would a decent home machinist lathe with a person who has good skill, but not god like-can do anything type, be able to do it?
Short answer, no.
But yes, a talented, experienced machinist can make parts to tolerance on machines not ideally suited for the task, but he'll wish he'd paid the relatively modest cost to have it done correctly.

To do many parts often, automotive machine work on any scale requires specialized machines designed for the task. Everyday production work routinely holds tolerances to .0005" over a set of eight rod bores, pin bores, cylinder bores, block decks, valve guides and if the customer wants to pay a bit extra for it, .0001" can be had. But as mentioned, it's done with hones and surfacers, not home shop lathes and mills.

Random stories:

Before we got a Tobin-Arp rod boring machine, we spent hours making a jig to hold connecting rods on a mill table to bore the small end oversize. Even with all the setup, the results were sloppy compared to what the machine designed for the task can accomplish in a fraction of the time.

With sufficient setup and a large flycutter, it's possible to surface a cylinder head on a vertical mill. On a dedicated surfacer, the head will be on, off and out the door before the guy with the mill can even jig it up enough to begin leveling. The finish will also be correct for the material and the head gasket being used.

Same with cylinder heads with two different valve angles. A head shop will have the seats and guides cut before the guy with the mill can even find the fixtures and clamps to try to get one angle correct.

Don't even think about working with piston pins on a lathe. They're centerless ground and harder than the hubs of hell. It takes carbide tooling and a rigid lathe toolpost to even scratch them.

We work on obsolete Studebaker and Packard engines for which parts are not always available. Another enthusiast designed an adapter to enable an Oldmobile V8 oil pump to be used on the Packard V8. However, he couldn't find anyone to do the manual machining consistently at an affordable cost. The tolerances were just too difficult to hold doing each setup by hand. He asked me to take over the project and I happened to know a talented CNC guy who had a one-man shop. Once this guy did the program, he can run off this complicated part at an affordable price and each piece is dead nuts, when the manual machinist just couldn't do it.

jack vines
 
Last edited:
The finish will also be correct for the material and the head gasket being used.
Good stuff. Thanks. About correct finish, would you share which finish would be for which type of gasket?
Before, I was very surprised to find out the surface doesn't have to be mirror finish, but even need some kind of teeth for the gasket to bite.
 
Many times a simple flycut finish on aluminum parts is satisfactory for gasket sealing
 
What everyone already said. I just want to add that no lathe, professional or personal can achieve accuracy with consistency smaller than 5 ten thousandths. It takes different operations and equipment to achieve greater accuracy. For instance Jim Dawson regularly achieves much greater accuracy. But not with a lathe/lathe tooling alone: there's reaming, grinding, and lapping. They help ypu move toward superstardom accuracy.

My normal tolerance on the lathe is to the nearest thou (that is +- a half thou) On the mill 3 times that. With work I can get far better than that, but for almost everything I do, there is no need.
 
The cnc lathes I ran would hold a tenth (.0001) all day long after I ran a few cycles to get them up to operating temps . Everything was in a controlled room and it was for DOD which meant 100% inspection . My Mahr snap gages with Millimess indicators can make .00005 look pretty damn big on the dial . You're not likely to achieve these results on any lathe , not Hardinges , Monarchs , nor LeBlondes . That's why we make cylindrical and centerless grinders , laps , hones etc . I will say we had some nice Deckel and P&W jig bores/grinders that would hold less than .00005 true position daily , NASA used our machines to inspect there parts vs . the CMMs .
 
A lathe, any lathe, is not suitable for piston pin, lifter internals or that type of operation. For the average "home type" lathe to repeat and hold .0005 especially over any length I would say is remarkable............Bob

I beg to differ, A properly trained and experienced operator on a quality toolroom lathe can regularly obtain these dimensions, when necessary. However I do concede that these jobs are better performed with the aid of precision grinders. But just because you and/or your machine cant achieve these results doesn't mean it can't be done.

Remember while some of us hobby machinists have been properly trained, very few of us have or can afford a toolroom quality lathe.
 
OP said "home lathe" not "tool room quality" lathe. The OP also referenced "smooth" tight fitting piston pins and lifter plungers. While I agree a talented operator on a quality lathe can often hit the needed tolerances a lathe is not suitable for machining hardened steel to the tolerances and finishes needed. Can it be done on a onesy twosy, hit or miss basis? It likely can.
As a practical matter the lathe is not, other than roughing out, the correct tool for the job. That answers the OP's original question asked...................Bob.
 
Last edited:
Wow, I just saw this video. Not a normal lathe, but appears doable and affordable

40 millionth of an inch precision.

 
Back
Top