Another obligatory New PM-833TV and PM-1340GT Ordered!

mksj said:
Try it with a 4J independent, sounds like it is an issue with the 3J chuck, they can have a TIR issue, but also a skew or axial deflection issue.
Click to expand...

Thanks Mark. I just re- measured with test bar. Along the Z I’m getting in the tenths over the length of the bar, but testing Y (top of bar) I see a 1.5 thou rise approaching the tail. I know this won’t impact much, but wouldn’t the impact increase parabolically the smaller the stock I’m cutting?
 
Should not make that much of a difference over short distances, also depends on the cutter height. That being said it is hard to dial in the tool height to that degree, sounds like the jaws are not holding the work true. If you have another chuck, I would try that.
 
You may have a chip or other gunk on the jaw scroll groove. Try removing all the jaws and making sure they are clean. While they are out, look on the scroll as well to make sure there is not anything in there that could be causing the problem.
 
I got $5 says it’s your chuck.
 
Tried my ER40 set tru chuck today. Indicated the inside taper and got no run out there. Loaded up a bar, faced and center drilled and pulled it out about 10" and supported with live center. Took a cut and I was getting 3 thou under at the chuck. I must be doing something wrong at some point, but I have been unable to measure it. I can only measure my work, and it is coming out with a taper.
 
If your tailstock is out slightly, i.e. not forward enough you will cut a taper that is smaller at the chuck. Also to indicate the TIR at the chuck when using a set-tru type of chuck, I usually chuck up a precision ground rod (1/2" and 1") and zero out the TIR at the chuck.
 
So I think I have ID'ed the issue, or at least a contributing factor. I am getting a dead nuts reading on a test bar with a DTI and a small ball tip, on a cylinder. Well this could be an issue, so I decided to measure the spindle using a DTI on a flat surface, and check my work with my DRO.

The problem in pictures.

I advanced Z to zero on my DTI, and then zeroed the DRO.
IMG_3702.jpeg

IMG_3703.jpeg


Then I backed off Z, and advanced X to the other side of the spindle. Advanced Z to zero on my DRO. The result is obvious.
IMG_3705.jpeg

IMG_3704.jpeg


The theory is that me using a ball tipped DTI on a cylinder, would mask the results in a 10-12 inch sweep if the head had nod. I was able to measure some nod. BUT in order to measure and fix one, I need to know the other is right. I think if I had started this process with 2 dial indicators with FLAT bottoms 90 degrees opposed to each other, (like in Davids picture) this mystery would have been apparent.

TL/DR - My head is out of alignment.
 
Starting completely over. Removed the taper but the test bar is wildly off. Going to re-lvevl the lathe and carefully measure for twist, then deal with the headstock again I need to stop chasing my tail and start from the beginning. My precision level kinda sucks though...and its a real PITA to dial in in my built in 1960 house with no good reference level anywhere.
 
Your headstock is way out of alignment. Level is relative, the only thing that matters is that it reads the same front to back moving down the ways. A test bar should work, but when doing the headstock alignment on my lathes I use Rollie's Dad method, just chuck up a round rod and make sure the +/- swing is the same in both directions moving down the rod and rotating it at each test point. I use a dial indicator with a mushroom/flat head. Then one can do some light test cuts. Just about everyone I know with the 1236T/1340GT, the headstock needed alignment after the lathe was delivered and leveled.
 
I have had it. Just threw a *******t cheap eBay "master" precision level in the trash..The glass on it was flawed. badly.The adjustment on it was awful. The glass must have been banana shaped as it would creep towards the middle and then fly off to the other side making tiny adjustments.Back and forth I went, but it dawned on me that it wasn't even holding its calibration when I thought I was getting close, and looked at my Starrett 98 and it was WAY off. Thats what I get for buying a cheap POS I guess.

I hear what you are saying Mark, but I am very perplexed.

If I use a test bar (confirmed cylindrical over its length to the .00015" accuracy). and a DTI to align my lathe, I will cut a taper of AT LEAST .004" over 3.5 inches,

If I align my lathe against the spindle, the taper is greatly reduced. .0015" over 4 inches, but then when I run my test bait is WAY off. Spinning the test bar at any given location shows no runout.

Something is incredibly wrong here.
 
You must log in or register to reply here.
Back
Top