Bolting down a lathe

[cybor462]

I finally got a REAL machinists level and will be checking my lathe right this time. If all is ok or if I need to level should I bolt it down or does it really matter.

I never bolted it from the begining because I do not feel a 13x36 could ever be worked hard enough (within reason) to make bolting it down needed. The other way to think would be since it is light it should be bolted.

Now it will be a bugger to bolt it so I wanted to have some discussion on bolting them down.

I am serious about this so lets have a discussion.

[tattoomike68]

bolting down is not needed unless you are running the hell out of it and it starts walking around when running balls to the wall.

put it on a flat place with some studs to hold it still and run the hell out of it. lathes dont need bolted down tight at all.

[Bob Ford]

I live in EARTHQUAKE country! Bolting down is very important.
Bob

[darryl]

There's a few ways to look at this. Here's three- bolted solidly, not bolted at all, or set over pins that are just locators.

First we'll assume your bench is solid. If you place rubber discs under the lathe's feet, then put some rubber hose pieces over some lag bolts, then bolted those through the feet and into the bench- not torquing them down over the lathe's mounting lugs- that's what I'm calling set over pins. It keeps the lathe in place, but doesn't add to the rigidity of the bed at all, plus some of the vibration is kept from transferring to the bench by the rubber discs. If the bench isn't flat, the lathe will be wanting to rock to some degree, unless it has only three mounting lugs. I'm assuming for now that it has four.

Next is solidly bolted down to the bench. In this case, either the bench is warped to suit the lathe, or the lathe bed warps to suit the bench, or you shim carefully to keep the bench solidly on all four legs while preventing any twisting of the bed. (or actually applying some twist to the bed as required for alignment) If this is done right, the bench will help the lathe bed to stay aligned. This assumes of course that the bench isn't warping for any reasons, like humidity changes, floor level changes, etc.

Not bolted at all- well, the lathe would be free to move around, and could get into a vibration mode that would be dangerous. Some off-center chucking and certain speeds are most definitely going to cause some problems.

It's possible that it boils down to this- if the bench is solid and doesn't warp for any reason, then bolting the lathe solidly to it allows the bench to control the bed and can help with the alignment. If the bench is queasy you would not want the lathe to be solidly bolted to it since it would alter the alignment- but you lose the capability of aligning the bed because there's nothing to torque against to do that.

Does that lathe have a three point mounting? If so then the bench can't be used to help align the bed, unless you bolt a bar under the tailstock end, and have both ends of that bar as mounting lugs. My own preference, because my lathe is short and a three point, is to solidly bolt the headstock end and leave the other end snug but not tight.

It would seem to me that a longer bed would benefit a lot from a solid bench, as long as that benchtop stays flat. In that case, I'd say bolt it up solidly, but carefully shim and align the bed as you do so. It doesn't take much movement of a bench's feet to throw it out of whack. Some means of keeping them in place on the floor can make a lot of difference.

[oldtiffie]

I have an open mind on this topic but I've yet to see definitive explanation of why "leveling" or "correcting twist" is either really necessary or practical in the general case.

Well, that's as maybe.

Yeah, I know, that is just about the biggest heresy of all to some.

Perhaps some of this "Holy Writ" stuff as regards machines is "Bull**it" and I can't see why I can't question it.

If it's all "cast in stone" - then show me.

I'm not interested in the "everybody says/knows/does/etc." or "So and SO - an acknowledged Guru says so" stuff.

In the immortal words of Prof. Julius Somner-Miller: "Why is it so?"

Generally, the "box" or "ribbed" section of a lathe bed frame is a rigid as ................ and will not be moved much without significant torque and a base that is capable of resisting it.

How are you going to limit the "correction" to any part or parts of the lathe bed? How are you going to stop it twisting at the head-stock or tail-stock.

Does it really matter in the grand scheme of things?

Why is the assumption always that the distortion is a "twist". What if it is a "hog" or "sag" (vertical and/or horizontal) and if so how are you going to correct THAT? What if it is all three? How would you know?

I am reasonably sure that if my lathe was 5, 10 or 15 degrees - or more - out of level - in both the longitudinal or transverse planes that it would not make much if any difference to the operation or efficiency of the lathe.

I have a "super-good" Machinist Level as well, and to be candid, it is a PITA to use and so back in its box it went and there it is staying.

I get by very well with a good accurately machined square box-section Carpenter's level which has an accuracy of better than 1mm per meter - 1:1,000 which is 0.001 which is the ArcTan or ArcSin of 0.057 arc degrees = 3.438 arc minutes.

Just to put that into perspective: your average "Chippie" has a tool with an accuracy order of magnitude better than the calibration of most rotary tables and vernier protractors which are calibrated/read to 5 arc minutes!!!.

Similarly, how many worry themselves stiff getting a lathe head and tail-stock aligned only in the lateral plane? How many actually run the DTI over the TOP of the test piece and see if the head or tail stock centre lines are parallel to the lathe bed in the vertical plane?

What about my lathe? Is it bolted to the bench (steel)? Yep? Why? to keep it on the bench. What's going to "give first? The bench (I hope!!).

If all this is so critical, why isn't it mandatory on precision grinders and Tool-Room lathes etc? Most are on 3-point suspensions and usually on resilient mounts.

The only thing I make sure I get level is the grinding wheel balancing stand for setting the balance weights on my surface grinder balancing weights/flanges. That's all. And I set it up each time I use it where-ever it's handy and I use my Carpenters level!! Any good? Too right. How good? When I've set the balance on the wheel and set it on the knife edges, the wheel either runs end-to-end freely without stop, or if stopped - stays stopped.

If I were concerned about anything on my lathe bed it would be to put an accurate straight-edge over the top and sides of the bed. If there were "bending" in the horizontal or vertical plane/s I'd really have a problem and all the "corrective twisting" in the world applied at the lathe-to-bench/floor supports won't change it.

And while I'm "on it" - what if your mill table or vertical slide-ways were bent or twisted? How would you know? And if you did, how would you correct it? Or could you? Would you?

Anything else?

Yep.

I will get a very accurate "Machine Square" aka "Precision Frame Level" for checking my squares and machines for "square/ness".
http://www.hareandforbes.com.au/sample_2/home.php (top left) as it had the same order of accuracy as the "Machine Precision Level" (0.02mm/1,000mm), will do the same job, is marginally cheaper and is ground "square" to a very high order of accuracy.

All of these "high end" "levels" need to be checked and re-set (if necessary) each tim ethey are used. Know how to do it? No? Why? The theory is OK but this too can be a real PITA in the application of it as its as "fiddly" as bug**ry.

[JCD]

None of the commercial shops I’ve been in have lathes bolted down.
50-60.

[cybor462]

So far it looks like it is 50/50 the good thing is that we have a discussion and I am seeing reasons why I would or would not bolt it down. With that I can make a decision.
That being said I should have said I have a Grizzly G4003 which weighs a little over 1000 lbs. It is bolted to the stand which is sheet metal cabinets with heavy tops and bottoms. There are two connected with a sheet metal strip. By themselves they are somewhat weak but together I feel they are barely adequate.
The lathe does have 2 solid pedestal mounts like a square column. They have 4 bolts each.
I did level it with a high quality carpenter level actually a Starret which Enco played up to be a machinist level but it is a carpenter level but a very high quality one.

Always wanting a machinist level I finally got one accurate to .005 in 12" which is not the NASA version but should do fine in my case. If it is off it should be consistently off so .005 on the way should not be a problem.

Good point so far on all sides. I am not ready yet to decide. Anymore real life stories about this that can be shared?

[sch]

From a practical point of view, if the mounting holes for the
lathe bench are not oversized, unless you use a diamond
core drill to make the hole, putting 4 holes at precise spots
in concrete is a bugger. Aggregate in the concrete or steel
if present can shift holes all over the place.

On another point, once I had a machinist level, I was amazed
at how easy it was to move the bubble when the level was
across the bed of the lathe, lathe bolted to the stand and
sitting on 1/2" bolts and pads just by pushing down on the
bed with my hand. Gave a new insight into how rigid 1200
lbs of CI/steel was, or perhaps I was just flexing the concrete
pad.

[Ed Tipton]

At the risk of seeming to be overly diplomatic, I think it depends upon the type of work you are doing and how much accuracy do you really need. I have a 14 inch Clausing and a 16 inch Monarch lathe and neither are bolted down. I also have a concrete floor that is reasonably level, and generally speaking, I don't work either of these lathes very hard, but then there is always that "special" job that falls outside of my "usual" type of work where I'm either hogging heavy cuts, or where I need extreme (for me) accuracy. I can honestly say that for each of my lathes, I have leveled them each once, and then I check them periodically. Neither has ever moved that I can detect. I would work with it, and if repeatability or accuracy becomes an issue, then I would take the necessary time to level it out and then bolt it down. As with anything else, if you have a problem, you either live with it, or keep going until the problem is resolved to your satisfaction. I do know from experience that a twisted bed will not give repeatable accuracy over the entire length of the bed. This is not a myth, it is a fact. But having said that, most lathes are pretty heavy, and made from subsatantil castings, and due to that, a lathe bed that is actually twisted is somewhat rare assuming that you set it up properly to begin with.

[cybor462]

I have a question Ed, what does either of your US made lathes weigh? I am curious if Chicom made is as heavy as US. In my eyes this plays a very important role in whether to bolt or not to bolt and yes how hard you will work it and how accurate you want to be is also a factor.