I have a Sheldon 13x40 lathe and just got it leveled with the Starrett
199 level. I now want to use the 2 collar method to check the
alignment. I read in an old post that some folks like to use aluminum
so I picked some up some 1 1/8" 6061 and put it in a 5C collet.

I cut off a 10" piece and trued it up and the hogged out the middle to
form the collars. I had about 8" hanging out of the collet. I sharpened
the cutter again and honed it with a small radius and took a cut across
the collars at 1000rpm and .00085 feed. The bar chattered terribly on
the end but cleaned up beautifully at the headstock end.

I know that is alot of overhang but was not sure how long the bar had
to be to be accurate. All SB specifies is 3-4 inches between collars
and the picture in the book looks like the bar is only 5-6 inches long.


So how long should the rod be to be accurate and how wide should the
collars be?

Also, when the Starrett is "level" the bubble overhangs each side of
the "level" mark by about 1 graduation. Is this normal? Also, what if
you leveled each end of the lathe and only brought the bubble to the
end of the inner line on each end making sure you had the front of the
level facing the front of the machine at each end. I assume the lathe
would be level in one plane but equally twisted a small amount along
its length by 1/2 a graduation?

Thanks, Steve

You need to support the stock with the tailstock, center drill and use a
center. This "collar" test is done between centers most of the time. If you
use a collet you still need to center drill and use the tailstock.
wrote in message
ps.com...
I have a Sheldon 13x40 lathe and just got it leveled with the Starrett
199 level. I now want to use the 2 collar method to check the
alignment. I read in an old post that some folks like to use aluminum
so I picked some up some 1 1/8" 6061 and put it in a 5C collet.

I cut off a 10" piece and trued it up and the hogged out the middle to
form the collars. I had about 8" hanging out of the collet. I sharpened
the cutter again and honed it with a small radius and took a cut across
the collars at 1000rpm and .00085 feed. The bar chattered terribly on
the end but cleaned up beautifully at the headstock end.

I know that is alot of overhang but was not sure how long the bar had
to be to be accurate. All SB specifies is 3-4 inches between collars
and the picture in the book looks like the bar is only 5-6 inches long.


So how long should the rod be to be accurate and how wide should the
collars be?

Also, when the Starrett is "level" the bubble overhangs each side of
the "level" mark by about 1 graduation. Is this normal? Also, what if
you leveled each end of the lathe and only brought the bubble to the
end of the inner line on each end making sure you had the front of the
level facing the front of the machine at each end. I assume the lathe
would be level in one plane but equally twisted a small amount along
its length by 1/2 a graduation?

Thanks, Steve

In article , Waynemak says...

You need to support the stock with the tailstock, center drill and use a
center. This "collar" test is done between centers most of the time.

No. The two-collar method as outlined in the South Bend book is
done *without* a center in the outboard end of the stock.

Jim


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================

In article om,
says...

I cut off a 10" piece and trued it up and the hogged out the middle to
form the collars. I had about 8" hanging out of the collet. I sharpened
the cutter again and honed it with a small radius and took a cut across
the collars at 1000rpm and .00085 feed. The bar chattered terribly on
the end but cleaned up beautifully at the headstock end.

You could go up to a larger diameter piece of steel stock,
do NOT use a radius on the end of the tool. Keep the feed rate
and spindle speed down and you will probably get out of the
chatter zone. 1K rpm is probably too high.

The radius on the tool is probably the issue, how are the
spindle bearings on this machine?

You can also reduce the overhang to six inches and try again.


--
==================================================
please reply to:
JRR(zero) at pkmfgvm4 (dot) vnet (dot) ibm (dot) com
==================================================