I have the PM 1130. I borrowed a precision level that reads in .0005" to check the bed for twist. I have the lathe bolted to the stand that I bought with the lathe. I put 1-2-3 blocks on the flat part of the ways. It takes .027" of shim to get the level close to level. I clamped squares across the ways to keep the blocks in the same position when reading the level. I'm off .0005" on one end and zero on the other end. I know I'm checking for twist and true level. My lathe does cut about .004" taper in about 8" and I'm trying to eliminate causes . Is the way I'm checking correct or not? I did some machine work in my old job but never had to set a lathe up before.
-
Welcome back Guest! Did you know you can mentor other members here at H-M? If not, please check out our Relaunch of Hobby Machinist Mentoring Program!
Checking for Twist
- Thread starter sportfan
- Start date
- Joined
- Dec 15, 2022
- Messages
- 65
I'm just a beginner, but I have the same lathe. I found both to be super helpful.
- Joined
- Feb 1, 2015
- Messages
- 9,948
There are multiple causes for cutting a taper and bed twist is just one. Usually, the bed is purposely twisted to correct for other issues. In some cases, two wrongs can make a right. If you are within .0005" over the ways, you are off to a good start. It is possible that 5the headstock is out of alignmnet but before messing with tha, I would try shimming to correct the taper. Check your ability to cut a flat face. If you are cutting a dome or a dish, that would most likely be due to a headstock misalignment. If you are good there, then proceed to shimming. You should be able to remove the taper.
I chucked up a piece of aluminum that was 3.5" in diameter. I zero it in the chuck and took a light clean up cut across the face. Checked it with an indicator and it was zero. The cam lock chuck has .001" run out across the face and on the o.d.. I took the chuck off and checked the spindle. The big o.d. of the spindle ran out .0015", the tapered area o.d. ran out .001", the face of tapered area ran out .002" (I don't think the chuck touches this area when its on the spindle), and the bore of the spindle ran out .001" opposite of where the bore ran out .001. I'm going to check it some more before I try shimming it. One friend said the jaws could be the problem.
If you're at five tenths across the ways, you're probably a lot more level than most lathes are. That's not enough to explain 4 thousandths in eight inches.
Actual level doesn't matter here. You want that, but using a level that's five tenths per division, no one, even you would know the difference if you shimmed up the low leg, took a shim out of the high leg, or shimmed the opposite side of the lathe so that both ends were out equally. That's really dang close.
Once you've got a part in the chuck (any chuck), or a collet, or hanging off of a face plate, or welded to the spindle if you wanted to, (don't do that), the part may or may not run true. After you take a pass or two along the part, (or take a facing cut), that's all irrelevent. You have reduced the headstock errors to ONLY the spindle bearings. Even if the spindle is bent or twisted, you've still eliminated that. The part, whether it was straight or not, only rotates about the axis formed by the headstock bearings. So.... Don't worry about any of those items yet.
With eight inches of stickout, did you have the part supported on a center? What was the diameter of that test piece? Is it a solid piece, or did you do a two collar type?
When you faced the piece of aluminum, when you had the indicator on it, did you indicate ALL the way across it, or only to the center? And indicator will read zero on the entire first half, from the near side to the center, it's the back half that will show the error.
Actual level doesn't matter here. You want that, but using a level that's five tenths per division, no one, even you would know the difference if you shimmed up the low leg, took a shim out of the high leg, or shimmed the opposite side of the lathe so that both ends were out equally. That's really dang close.
Once you've got a part in the chuck (any chuck), or a collet, or hanging off of a face plate, or welded to the spindle if you wanted to, (don't do that), the part may or may not run true. After you take a pass or two along the part, (or take a facing cut), that's all irrelevent. You have reduced the headstock errors to ONLY the spindle bearings. Even if the spindle is bent or twisted, you've still eliminated that. The part, whether it was straight or not, only rotates about the axis formed by the headstock bearings. So.... Don't worry about any of those items yet.
With eight inches of stickout, did you have the part supported on a center? What was the diameter of that test piece? Is it a solid piece, or did you do a two collar type?
When you faced the piece of aluminum, when you had the indicator on it, did you indicate ALL the way across it, or only to the center? And indicator will read zero on the entire first half, from the near side to the center, it's the back half that will show the error.
I had drilled a center in a one-inch piece of aluminum and supported it with a dead center and took two light cuts to true it up. Then I did the barbell cuts like in the Blondiehacks clip to keep it cool. After my first cuts I tried it unsupported also. When I cut the face of the larger piece I ran the indicator completely across the face of the part. I'm using a scrolling 3 jaw chuck while doing these checks. Should I try using a 4-jaw chuck that I have? So far the runouts on the spindle concern me the most. I'm just trying to eliminate possible causes as I go. This lathe is supposed to have the more accurate ISO 9000 bearings on the spindle.
Here is something I ran into. When leveling the lathe, I noticed that if I put my hand on the bed, it changed the level reading. That was resting my hand when moving around, setting up. If the reading changed with a pound of pressure, what is it doing with 20 lbs of steel moving around? I have the PM 1228 on original PM base. I am going to remove the lathe and place a double thickness of Advantech under the chip tray and add some supports on both side of the cabinets. I think that the base cabinets have some give in them. There is just something wrong here.
It may not be worth messing with. When you have a precision level on even a larger lathe you can push against it and change the reading. The friend that I borrowed the level from took a gunsmithing class years ago and the instructor demonstrated this to the class. I was careful while checking mine to not lean on it anywhere. Lathe flex more than you would think they would.
Interesting. I may try it later on to see if I can reduce chatter. There is so much sheet metal opposed to cast iron that lots of time the lathe screams. If I gain 100 RPM more before vibrations, it would be worth it.