Lathe Not Cutting Straight

First thing I would do is take your rpm's up by at least 1000 , then move your tailstock away from you .003 . Recut till you achieve what is acceptable to you on this piece . After this piece , grab a lathe mandrel or something hardened and ground , put it between a turned center in your chuck and a dead center in your tailstock and indicate the bar true . Turning at 175 rpm on a 3/8 shaft is slow motion , crank it up ( especially if you're using carbide) . I'd start at your machines max rpm and work down if necessary .

I tried carbide at higher speeds,lower speeds,then went to HSS,and tried different speeds. 175 was my last cut. If it turns out to be wear in the ways,do I shim the tail stock?
 
I tried carbide at higher speeds,lower speeds,then went to HSS,and tried different speeds. 175 was my last cut. If it turns out to be wear in the ways,do I shim the tail stock?

750 rpm would be the minimum rpm I would use with hss cutter . With carbide , go for the limit . Did you pull your tailstock into adjustment ? If so , there are many ways around the taper issue if your ways are worn .
 
I am cheating and uncaring so I program the taper out of the part.

What do you do?

Reminds me of a place I used to work at. We had a worn out Cincinnati Cinturn, remember them?, that would turn nearly .030" taper that varied on a cut of about 18" in length. They knew exactly where the taper was and would program for it so the finish part was perfectly straight on the OD. It's amazing what we will do to obtain perfection in the work we do. Or close to it as we can get it! Ken
 
A lot of good advise and all have merit.
There are many ways to check for twist in your bed (ie. leveling your lathe).
I came across a new to me way to do it, requires very little investment in materials and you don't need to use a precision level.

I would chuck up a chunk of aluminum, set your compound to 60 degrees and machine a point on it. Do not remove it from your chuck. Now put a dead center in your tail stock and slide it up to touch the part you just made. Make note of what you find but don't make any adjustments to your tailstock yet.
"level" your lathe, here is the link to Tips and Tricks in Metal
after checking the level and/or leveling the lathe, run the tailstock back up to see if the points match


Good luck
Mike
 
I am cheating and uncaring so I program the taper out of the part.

What do you do?

I take the 5th!!! (Thought I did the other day! just getting back to the thread).

Lots of stuff they didn't teach in school, but that works when you just want to make a part.
 
Maybe misaligned, even probably. A cheating, uncaring machinist would set up the work piece, then run an indicator along the 3.5"', then adjust the tailstock until the two ends give the same reading. THen try a cut and see what happens. Of course that's not really aligning then tailstock for general use, but might work pretty well for a one-off.

Afraid you lost me on that one! If it makes the part straight, it lines up the tail stock, assuming that is the problem. If it is the head stock that is out, then no, it is a per part remedy. But the OP said he used the tail stock, so....... (Hey Wreck, Programming it out sure is handy, too!)
 
What indicates that this is more likely a tailstock rather than headstock (or both) alignment issue?
 
What indicates that this is more likely a tailstock rather than headstock (or both) alignment issue?

Mostly the fact tailstocks are designed to move and be offset, headstocks aren't...

It's possible that it's the headstock, but far more likely that it's either tailstock offset (very few lathes have no adjustment there) or bed twist (I once had a call out to a customer ranting and raving because his brand new lathe (his first full-size machine) cut a 6 thou" taper in 2", he hadn't levelled it so I applied precision level^ to the carriage and some shims under the cabinet - result no measurable taper* in a light 10" cut) - both are a lot easier to sort than a misaligned headstock.

For a headstock check when the tailstock adjustment and precision levelling have failed the OP will need a proper test bar (taper to fit his spindle, clean spindle taper and test bar, ground test bar extending 8 - 12", DTI checks along its length to check that it runs concentric with the spindle and some easy maths to average out eccentricity if it doesn't) and good quality DTI - and even that is subject to errors from bed wear which will move the DTI relative to the theoretical axis of the ways.

Hefty "traditional" lathes often have the headstock fitted (machined, ground and usually scraped-in) to the ways on the bed, which doesn't leave much leeway for misalignment (as long as you keep everything scrupulously clean, a piece of grit or swarf's enough to ruin the alignment) - my 50's Holbrook is built that way, there's no measurable shift in headstock alignment when lifting the headstock and replacing it (necessary to get to some of the feed gearing, dammit, especially if you want to see the gears in mesh instead of on the bench!). More recent (and particularly far-eastern) machines may have adjustments either through lockable (I hope!) screws or shims between headstock and bed ways/locating points, usually around the back of the headstock.

Hope that helps,
Dave H. (the other one)

^ my granddad the WW2 ballistician's 10 second-of-angle clinometer, used to set up howitzer sights, that's about 2" at a thousand yards...
* with a 0.01mm ~1/2 thou" micrometer
 
Back
Top