- Joined
- Apr 21, 2015
- Messages
- 653
I must owe you money.
-
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!
Scraping in all bearing surfaces on my Wards/Logan 10"
- Thread starter Rex Walters
- Start date
- Joined
- Feb 1, 2018
- Messages
- 1,867
Jeff your a little behind the times...lol Allischicks post was from 2021. But your right about Rex. A very talented guyRex is very talented, patient, methodical, good looking and rich
- Joined
- Mar 26, 2018
- Messages
- 8,645
Oops.Jeff your a little behind the times...lol Allischicks post was from 2021. But your right about Rex. A very talented guy
- Joined
- Dec 16, 2023
- Messages
- 3
nice ...
- Joined
- Apr 21, 2015
- Messages
- 653
Apologies for posting to a very old (but apparently popular) thread of my own, but I just re-read reply #44 after receiving another like, and I really must correct a slight mistake I made toward the end. The post is too old to edit.
That post describes what some call Rollie's Dad's Method (RDM) to find the true axis of rotation. As far as I'm concerned, it's the only method that works accurately and reliably. The test bar only needs to be truly round, about ten or twelve inches long, and with a consistent diameter. Test the bar with a micrometer at multiple places along the length and repeating the test after revolving the bar a few times. It doesn't matter whether the bar is held in the taper, in a collet or in a chuck. A decent piece of TGP shafting won't break the bank and will work fine (the diameter isn't critical, but something reasonably stout like 3/4" is ideal).
I think I wrote it up reasonably well, but my mistake was this sentence toward the end:
The procedure I described in that post establishes a line on top of the bar that lies on a horizontal plane that is parallel to the axis of rotation. But the end of the bar will now be pointing slightly toward or away from the operator.
If you were to sweep an indicator mounted to the cross-slide over the top of the bar by moving the carriage left and right, the indicator tip would also move toward or away from top dead center, even on a perfectly unworn lathe. The indicator reading will change slightly as you traverse, but the angled or curved line formed by connecting all the TDC points along the bar will be on a horizontal plane parallel to the axis of rotation.
To establish a straight line parallel to the axis of rotation, you need to connect two points at top dead center (one close to the chuck, one toward the end of the bar).
To summarize the absolutely accurate process:
At this point, the imaginary line connecting the two marked points will be exactly parallel to the axis of rotation.
If the TDC points are on the same side of the bar, and your dial/DRO readings are close to the same, you can safely sweep the top of the bar by traversing the carriage left and right to evaluate the condition of your ways directly (but to be completely accurate, you must find TDC at each location without rotating the bar). Any change in indicator reading indicates that your ways are worn or twisted.
When reasoning about this process, it helps to imagine a test bar with a compound bend kinda like a crankshaft, with two U bends in the vertical and horizontal planes (hopefully your real test bar is more accurately made!). Imagine it's mounted in the spindle at an angle, too.
Imagine how you would need to rotate the spindle and move the cross-slide to find TDC and min/max/avg deflection at different points along such a bar.
The easiest way to measure and remove twist, of course, is with a precision level, NOT using an indicator (ideally, before starting the process above). But this is a really great method to evaluate the condition of your ways after leveling.
Hope this helps someone. I'll sleep better tonight now that I've corrected the mistake!
That post describes what some call Rollie's Dad's Method (RDM) to find the true axis of rotation. As far as I'm concerned, it's the only method that works accurately and reliably. The test bar only needs to be truly round, about ten or twelve inches long, and with a consistent diameter. Test the bar with a micrometer at multiple places along the length and repeating the test after revolving the bar a few times. It doesn't matter whether the bar is held in the taper, in a collet or in a chuck. A decent piece of TGP shafting won't break the bank and will work fine (the diameter isn't critical, but something reasonably stout like 3/4" is ideal).
I think I wrote it up reasonably well, but my mistake was this sentence toward the end:
The procedure I described in that post establishes a line on top of the bar that lies on a horizontal plane that is parallel to the axis of rotation. But the end of the bar will now be pointing slightly toward or away from the operator.
If you were to sweep an indicator mounted to the cross-slide over the top of the bar by moving the carriage left and right, the indicator tip would also move toward or away from top dead center, even on a perfectly unworn lathe. The indicator reading will change slightly as you traverse, but the angled or curved line formed by connecting all the TDC points along the bar will be on a horizontal plane parallel to the axis of rotation.
To establish a straight line parallel to the axis of rotation, you need to connect two points at top dead center (one close to the chuck, one toward the end of the bar).
To summarize the absolutely accurate process:
- Mount the test bar as securely as possible (held in the taper, in a collet, or in a chuck).
- Mount the indicator base to the cross-slide or compound.
- Place the tip of the indicator at the end of the bar, visually near top dead center.
- Rotate the bar to find the maximum indicator reading, then feed the cross slide in and out until you find TDC (max indicator reading)
- Continue rotating and moving the cross-slide to find the maximum reading at TDC and zero the cross-slide dial or DRO.
- Rotate the bar to find the minimum indicator reading (180 degrees away if an accurately ground test bar) and zero the indicator.
- Rotate the bar to find the indicator reading halfway between the minimum and the maximum reading.
- Mark that point on the bar.
- Repeat steps 4-8 with the indicator tip closer to the chuck and without zeroing the cross-slide dial or DRO.
At this point, the imaginary line connecting the two marked points will be exactly parallel to the axis of rotation.
If the TDC points are on the same side of the bar, and your dial/DRO readings are close to the same, you can safely sweep the top of the bar by traversing the carriage left and right to evaluate the condition of your ways directly (but to be completely accurate, you must find TDC at each location without rotating the bar). Any change in indicator reading indicates that your ways are worn or twisted.
When reasoning about this process, it helps to imagine a test bar with a compound bend kinda like a crankshaft, with two U bends in the vertical and horizontal planes (hopefully your real test bar is more accurately made!). Imagine it's mounted in the spindle at an angle, too.
Imagine how you would need to rotate the spindle and move the cross-slide to find TDC and min/max/avg deflection at different points along such a bar.
The easiest way to measure and remove twist, of course, is with a precision level, NOT using an indicator (ideally, before starting the process above). But this is a really great method to evaluate the condition of your ways after leveling.
Hope this helps someone. I'll sleep better tonight now that I've corrected the mistake!
Last edited: