A head scratcher here …

You use bed twist to compensate to some degree for bed wear, on a new machine you should not have bed wear and it should cut straight if the head is aligned properly. Beds do settle, but I would not use bed twist on a new machine to compensate for poor head alignment. The head alignment is fairly straight forward, I use a 1" ground rod, or one can get a test bar. Using Rolle's method it also compensates for the chuck not holding the stock straight.
 
Just aligned my lathe again and will do some test cuts shortly. The level should be accurate to .0002"/10". I am guessing that Mark is correct and my headstock will need to be adjusted. I am kind of worried about doing that though.
 
Have you swept the indicator on the top or bottom of the work to check for nod? Which side of the taper is narrow?
 
Okay re leveled the lathe - so the ways should be aligned to within .0005. Inserted a 16" ground test bar into my spindle and ran a .0001 indicator on the side of the bar and also another .0001 indicator on top of the bar.

indicator on the bar's side: looks like the head is off by around .0007 over 12".
indicator on the bar's top: looks like the spindle is nodding - that was off .003
 
I assume the narrower side was at the Chuck side? Edit, I guess tool height would have more to do with that. Disregard this post I think

poorboy said:
Okay re leveled the lathe - so the ways should be aligned to within .0005. Inserted a 16" ground test bar into my spindle and ran a .0001 indicator on the side of the bar and also another .0001 indicator on top of the bar.

indicator on the bar's side: looks like the head is off by around .0007 over 12".
indicator on the bar's top: looks like the spindle is nodding - that was off .003
Click to expand...
 
I just ran the two collar test - here are the final numbers:

1.60005" collar closest to to the chuck
1.59725" collar closest to the tail stock

It seems that i still have a .0028 taper (which is much better than before). i am taking .003 cuts - so there might be some deflection in the rod as well.

I also ran a .0001 indicator from my tool post - that sees an approximate difference of .0007 between the two collars.
 
Headstock is out of alignment. It is not a big deal to align it, and as I mentioned it is frequently needed with the 1340GT. Please note that it can also be that the chuck is not holding the stock true, thus checking with Rolle's method, all you are looking at is that the swing in the +/- direction is the same.
 
poorboy said:
I have an interesting issue with my PM-1340GT that I am trying to figure out.

I put a 2”x12" bar in the chuck and performed a collar alignment test (with no tailstock). Took light cuts (.002") -- I then put a .0001" indicator on the carriage and ran the indicator over both collars -- I am off by .0003" over 10 inches. Everything looks great - but ...

If I use a micrometer to measure the collar closest to the chuck and then measure the collar farthest from the chuck and the difference is .010". There seems to be a taper in the bar. How can I have a taper and not have the indicator on the carriage show that there is a taper ?

Any ideas ?
Click to expand...
When you cut the collars, the cut is referenced to the carriage. Replacing the tool with a dial indicator, you are again referencing to the carriage. The dial indicator should read the same in both positions. A small difference could be caused by deflection due to tool pressure.

This doesn't mean the two diameters will be the same. Head misalignment, bed twist, or wear in the ways will cause a difference in the diameters.

Note that what you did is essentially the opposite of what the RDM method does. It measures the difference in indicator reading for two positions on a bar of uniform diameter.

Before attempting to adjust the head, check for alignment on a facing cut. A facing cut is minimally affected by bed twist. If a facing cut produces neither a domed or dished surface, more than likely, your head is aligned. Rather than cutting a face, I use a dial indicator to measure the relative displacement of a boss on my faceplate when in a front vs. rear position. Details are here. https://www.hobby-machinist.com/threads/on-the-level-further-adventures-with-the-g0602.72338/
 
RJSakowski said:
When you cut the collars, the cut is referenced to the carriage. Replacing the tool with a dial indicator, you are again referencing to the carriage. The dial indicator should read the same in both positions. A small difference could be caused by deflection due to tool pressure.

This doesn't mean the two diameters will be the same. Head misalignment, bed twist, or wear in the ways will cause a difference in the diameters.

Note that what you did is essentially the opposite of what the RDM method does. It measures the difference in indicator reading for two positions on a bar of uniform diameter.

Before attempting to adjust the head, check for alignment on a facing cut. A facing cut is minimally affected by bed twist. If a facing cut produces neither a domed or dished surface, more than likely, your head is aligned. Rather than cutting a face, I use a dial indicator to measure the relative displacement of a boss on my faceplate when in a front vs. rear position. Details are here. https://www.hobby-machinist.com/threads/on-the-level-further-adventures-with-the-g0602.72338/
Click to expand...
Thank you - i will go ahead and perform a facing cut and see how that looks.
 
Keep in mind that the bar you're cutting the collars on shouldn't extend more than 4 times it's diameter without support from the tailstock. Any longer extension will result in some deflection of the bar. It could be due to the strength of the material or the quality (or preload) of the bearings used in the headstock.
 
You must log in or register to reply here.
Back
Top