PDA

View Full Version : Lathe Bed Straightness



kNucKlbustr
09-21-2015, 08:37 PM
While trying to solve a different prob, I ran into something Im not familiar with.

Running an indicator from tailstock end to headstock end, along the outside of the bed way from the carriage, there is quite abit of unevenness or warp.

Heres a drawing of the measurements.
http://www.rocketsaway.com/LatheBed01.jpg

Is there something wrong w/ the bed ? or, because the carriage does not "ride" along this part of bed, maybe its not machined correctly from factory ? or the carriage is being "forced" out at those areas ?

Richard King
09-21-2015, 09:00 PM
What sort of lathe is it? What size?? How about a picture what you wrote and drew may make sense to you, but to me it doesn't make sense at all. What the heck are you saying..? "" because the carriage does not "ride" along this part of bed, ??????"""

Are you indicating a test bar or are you cutting a bar and miking it? What size diameter is the bar?? Is the error gradual and tapered ?

JoeLee
09-21-2015, 09:05 PM
Being an insignificant surface it may have just been machined that way.

JL...........

Steve Middleton
09-21-2015, 09:09 PM
Until it's leveled can't comment?

Doc Nickel
09-21-2015, 09:10 PM
You're going to at least have to let us know how you "ran" the indicator across the bed. If you just stuck a mag-base on the carriage and put the tip somewhere on the bed, you could easily have more error than you show or less error.

In either case, there's several possible causes- the main one being simple wear. Unless it's a relatively new lathe, or has been relatively recently refurbished, chances are it's going to have some wear, typically in the form of a hollow (or low area) close to the headstock.

Another common cause is the bed isn't properly levelled. If the floor is out of level, or one hasn't properly adjustted the levelling feet, or any number of other seasons, it could very well have a "high" corner or a low one- which would show a progressive error along the bed. Compound that with the bed wear, and yes, you could easily see the results you're showing (assuming I'm reading your chart right.)

Doc.

kNucKlbustr
09-21-2015, 09:21 PM
What sort of lathe is it? What size?? How about a picture what you wrote and drew may make sense to you, but to me it doesn't make sense at all. What the heck are you saying..? "" because the carriage does not "ride" along this part of bed, ??????"""

Are you indicating a test bar or are you cutting a bar and miking it? What size diameter is the bar?? Is the error gradual and tapered ?

Google pic, but exactly like my 40" lathe.

1 is where indicator base set on carriage , 2 is outside bed way where indicated from TS end to HS end.
Zeroed at tailstock end.
http://www.rocketsaway.com/LatheBedIndic01.jpg

JoeLee
09-21-2015, 09:22 PM
You're going to at least have to let us know how you "ran" the indicator across the bed. If you just stuck a mag-base on the carriage and put the tip somewhere on the bed, you could easily have more error than you show or less error.

In either case, there's several possible causes- the main one being simple wear. Unless it's a relatively new lathe, or has been relatively recently refurbished, chances are it's going to have some wear, typically in the form of a hollow (or low area) close to the headstock.

Another common cause is the bed isn't properly levelled. If the floor is out of level, or one hasn't properly adjustted the levelling feet, or any number of other seasons, it could very well have a "high" corner or a low one- which would show a progressive error along the bed. Compound that with the bed wear, and yes, you could easily see the results you're showing (assuming I'm reading your chart right.)

Doc.Read post #2........nothing rides on that surface, unless at one time some one mounted a Trav-A-Dial and that wouldn't cause it.

JL................

kNucKlbustr
09-21-2015, 10:33 PM
Ok, Nothing rides there, so Ill assume it isnt a prob.

As far as leveling goes, I didnt use a machinist level. Used a a good carpentry level, 4yrs ago.
Its a gap bed and gap has never been removed.

Trying to dial everything in today. Trued the 6jaw chuck jaws, inner and outer.
Best I can machine a piece is within 0.001".
No taper on 24", dead on 0.000". The tailstock is adjusted for each diff. length.

lakeside53
09-21-2015, 10:56 PM
lol.. no issues. Wish my lathe had 00000 taper in 24 inches!

kNucKlbustr
09-21-2015, 11:26 PM
lol.. no issues. Wish my lathe had 00000 taper in 24 inches!

I jam 6061 tube on a expanding collet up to 42" on this machine and "try" for .002-.003". No follower.

wierdscience
09-21-2015, 11:44 PM
On an Asian lathe that surface can be simply knocked down with an abrasive belt and not even machined,it just has to look shiney.Try taking the same measurement on the V-way itself and see what you get.I would bet the v-ways are pretty bang on.

Paul Alciatore
09-22-2015, 02:17 AM
Yea, but measuring the ways with a DI that is mounted on the carriage, WHICH IS FOLLOWING THOSE SAME WAYS, is not a very good test. I would expect a low error. Heck, they could go in a complete circle and test OK within millionths of an inch.

Generally ways are checked against a straight edge, not a DI on the carriage. Once you have the ways that the carriage rides on straight and parallel, then you can use the carriage to check additional way surfaces, like the tail stock ways which are separate on some lathes.




On an Asian lathe that surface can be simply knocked down with an abrasive belt and not even machined,it just has to look shiney.Try taking the same measurement on the V-way itself and see what you get.I would bet the v-ways are pretty bang on.

wierdscience
09-22-2015, 02:50 AM
Yea, but measuring the ways with a DI that is mounted on the carriage, WHICH IS FOLLOWING THOSE SAME WAYS, is not a very good test. I would expect a low error. Heck, they could go in a complete circle and test OK within millionths of an inch.

Generally ways are checked against a straight edge, not a DI on the carriage. Once you have the ways that the carriage rides on straight and parallel, then you can use the carriage to check additional way surfaces, like the tail stock ways which are separate on some lathes.

Only looking for relative movement Paul,if he slapped the DTI on the v-way and still got the same run out something would be amiss right?It is very possible for a carriage to lift for reasons other than bed issues.Crud built up under the rear ledge is one possible cause for example.

Rosco-P
09-22-2015, 07:26 AM
Only looking for relative movement Paul,if he slapped the DTI on the v-way and still got the same run out something would be amiss right?It is very possible for a carriage to lift for reasons other than bed issues.Crud built up under the rear ledge is one possible cause for example.

I agree with Paul A. "Leveled" with a carpenters level, data collection process flawed, result: data useless.

Richard P Wilson
09-22-2015, 08:26 AM
If the OP's getting the results he quotes in post 8, how he levelled the bed, how he measured straightness doesn't really matter if the lathe turns that well.

JoeLee
09-22-2015, 08:54 AM
On an Asian lathe that surface can be simply knocked down with an abrasive belt and not even machined,it just has to look shiney.Try taking the same measurement on the V-way itself and see what you get.I would bet the v-ways are pretty bang on. I have to agree........ It could have been a second op set up when they cut the front of the bed way, poor setup, something moved or maybe the bed had to be moved and repositioned on the machine that cut it due to size restrictions etc. One way to tell would be to look for any interruption or pattern change in the cutter marks at the point in question. That would be a tell tale sign that the bed was repositioned to finish the cut rather than cut in one complete pass. Either way, as said before nothing rides on that surface. I find it doubtful that the bed warped but not impossible. We don't know if these are seasoned castings or not, I'm guessing not.

JL................

J Tiers
09-22-2015, 09:27 AM
Yea, but measuring the ways with a DI that is mounted on the carriage, WHICH IS FOLLOWING THOSE SAME WAYS, is not a very good test. I would expect a low error. Heck, they could go in a complete circle and test OK within millionths of an inch.

Generally ways are checked against a straight edge, not a DI on the carriage. Once you have the ways that the carriage rides on straight and parallel, then you can use the carriage to check additional way surfaces, like the tail stock ways which are separate on some lathes.

Separate on most lathes....

There is no way to test for bed twist whatsoever, without an external reference. A level, etc, with you looking for a change as the level is placed in various locations, manually or with the carriage.

You can test for what you care about (turning parallel) with the "two collars" test, which is done without ugse of the tailstock.