Sherline tailstock chuck upgrade?

[EricB]

Two caveats:

Second Caveat: I measured the vertical misalignment of the headstock spindle to the bed. I got 0.002,27 inch rise from left to right, along 2.75 inches of the z-axis. This amount of misalignment renders futile my attempt to measure and cancel the vertical misalignment of the tailstock. For comparison, I measured 0.000,09 inch horizontal misalignment (toward the front from left to right) along the same 2.75 inches.

Karl,
This is a shortcomming of the Sherline lathe design. 1) The bed is attached to the base by one cap screw at the extreme left end. 2) The bed is painted with a wrinkle finish paint that will compress over time. 3) When you tighten the screw the paint will compress at the screw end but not under the entire headstock, and the bed will flex relative to the base.


To see the problem remove the screw under the handwheel end of the lathe bed. The bed will straighten out float above the base at the handwheel end.

I solved mine by removing the paint.


The bed only contacts the base along the angled portion of the bed. I figured that out after removing all the paint.


Eric

 

[Karl A]

Eric,

The deflection resulting from compression of the paint near the screw is consistent with the vertical direction of the headstock misalignment on my lathe.

I found your thread about the paint. In it, I posted that the base of my lathe has paint in the contact areas.

Karl

 

[Karl A]

Following EricB's lead, I removed the paint from the areas of the lathe base that contact the lathe bed. The result is that vertical misalignment decreased: from 0.002,3 inch rise to 0.001,1 inch rise from left to right, along 2.75 inches of the z-axis.

Here are before and after photos of the base contact areas:

left side


right side

I used paper towel and acetone to rub off the paint. The paint softened significantly after being in contact with acetone-soaked paper towel for ten seconds.

Karl

 

[EricB]

Following EricB's lead, I removed the paint from the areas of the lathe base that contact the lathe bed. The result is that vertical misalignment decreased: from 0.002,3 inch rise to 0.001,1 inch rise from left to right, along 2.75 inches of the z-axis.

Karl

You might also consider mounting your lathe so that the entire length of the base is supported on a solid piece of material.

Eric

 

[Karl A]

You might also consider mounting your lathe so that the entire length of the base is supported on a solid piece of material.

Eric

Eric,

My lathe is mounted on an 80/20 frame. I assembled the 80/20 frame with the aid of a sheet of glass, for flatness. The frame has three adjustable feet -- two along the left side and one on the right side -- for leveling the frame. To make the lathe bed level at both sides when it is mounted to the frame, I needed to include an 0.003" shim under the right, front mounting location of the lathe base. Without the shim, the bed is twisted. I don't know the cause of the twist. I want to check the flatness of the base and the bed, before I make any changes to how the lathe is mounted.

Karl

 

[EricB]

Hi Karl,

I had seen how you had mounted your lathe. It looks like you have the headstock end cantilvered off of your supports. So the headstock, motor, and speed control assembly (almost 9 pounds), along with the end of the lathe, are floating unsupported.


I figure if you are going through the trouble of chasing alignment it would be a good idea to start with a more rigid setup. Here's a simple test. Chuck a length of stock and place a test indicator at the end. Then just press on the headstock and observe the indicator.

Eric

 

[Karl A]

Hi Eric,

Yes, the portion of the base-and-bed to the left of the support mounting screws cantilevers out.

I remeasured the vertical misalignment along 2.75 inches from left to right using a test bar. I got 0.001,1 inch rise, again.

I then removed the motor from the headstock, and measured again. The rise was 0.000,8 inch, or 0.000,3 inch less.

Thank you for explaining the rigidity issue.

Karl

 

[Karl A]

Eric,

The bed is the core of the lathe. The headstock mounts to the bed, and the motor mounts to the headstock. It seems to me that if the weight of the motor causes more deflection to the bed than is acceptable to me, then I ought to relocate the motor.

From the contruction of the lathe, it appears that the base is not meant to add rigidity to the bed. The base is not directly fastened to the bed. Rather, the base is fastened to the lead screw support on the left end, and to the lead screw thrust on the right end. Based on the types of screws used, the lead screw support and the lead screw thrust are more firmly fastened to the bed than to the base.

I think that the base ought to be mounted in a manner that does not apply uneven forces where the bed contacts the base.

Yesterday, after comparing headstock spindle misalignment with and without the motor installed, I had ideas of lengthing the 80/20 frame to match the 24-inch length of the lathe base, and adding a 3-inch wide by 1-inch thick by 24-inch long aluminum bar between the 80/20 frame and the lathe base. Plans for adding the bar became difficult, though: milling the bar flat, and milling the bottom of the lathe base flat.

Karl

 

[Karl A]

I stopped developing a tailstock spindle with offset taper (for correcting tailstock misalignment) after I measured significant vertical misalignment on the headstock spindle. Eric B suggested removing the wrinkle paint from the lathe base where it contacts the lathe bed. I removed the wrinkle paint; the misalignment was cut in half. Eric B then suggested mounting the lathe such that the entire length is supported on a solid piece of material.

On the two days after Thanksgiving, I experimented with my lathe mounted on the table of a Bridgeport-clone mill at work.

When I first placed the lathe on the mill table, I found that the lathe base rocked. Placing a 0.010-inch shim under the front, right mounting hole stopped the rocking.

The base did not rock when the base was placed on two 1-inch wide parallels, with the parallels located under the base mounting holes.

I decided to mill the bottom of the base flat, using the Bridgeport-clone mill and an indexable face mill. I attached the lathe bed directly to the lathe base, using long 10-32 flat-head socket screws, washers, ,and nuts. Here is a photo, taken shortly before milling:

After milling the base, I checked the flatness using a 1-meter-long straight edge (Kinex, 0.0014-inchj claimed straightness along length). The bottom of the base was higher at the middle of the length, due to deflection during milling I think. I used the straight edge, a file, and a permanent marker to flatten the bottom. I think that I got it flat to within 0.001 inch.

I reassembled the lathe, and measured the vertical misalignment of the headstock spindle using an alignment bar, tenths dial indicator, and Novaflex holder and base. Here is a photo:
The nuts on the strap clamps were finger-tight.

Here are the measured vertical misalignment values, across 2.75 inches from left to right:
- lathe mounted directly on mill table, motor NOT installed on headstock: +0.000,7 inch
- lathe mounted directly on mill table, with motor installed on headstock: +0.000,8 inch
- 1-inch wide 0.010" brass shims between lathe and mill table, under base mounting holes, no motor on headstock: +0.001,0 inch
- 1-inch wide 0.010" brass shims between lathe and mill table, under base mounting holes, with motor on headstock: +0.001,2 inch
- 6-inch wide brass shim under left side, 4-inch wide brass shim under right side, no motor on headstock: +0.000,9 inch
- 6-inch wide brass shim under left side, 4-inch wide brass shim under right side, with motor on headstock: +0.001,0 inch

I conclude that the headstock spindle does have vertical misalignment.

The misalignment was the least when the lathe was mounted directly to the mill table. This supports Eric B's recommendation to support the entire length of the lathe base with a solid material.

I was surprised by the small effect of the headstock motor. I expected greater increases in misalignment from its significant weight. The small effect is good news.

Karl

 

[Karl A]

I have further reduced the vertical misalignment of the headstock spindle, on my lathe, by removing the wrinkle paint on the cast base where the lead screw support (inch, 40370) and lead screw thrust (40300) contact the base. See the first photo in post #21 by Erik. The photo shows the cylindrical recess for the lead screw support.

Here is a snip from Sherline's exploded view of the 4400 manual lathe with cast base:

The wrinkle paint was especially thick near the edges of the recess for the engagement lever (15420) of the thread-cutting attachment. (The exploded view shows the plug button (40400) instead of the engagement lever.) I was able to rock (like a seesaw) a 5/8-inch diameter rod placed in the lead-screw-support recess. The rod didn't rock after I removed the paint. As before: I used paper towel and acetone to rub off the paint. The paint softened significantly after being in contact with acetone-soaked paper towel for ten seconds.

New vertical-misalignment measurement, after removing the paint: With the lathe mounted on my 80/20-frame base, excluding the motor from the headstock, I measured +0.000,4 inch rise across 2.75 inches from left to right.

Also, removing the paint significantly reduced the twisting of the bed that I have observed when the lathe is assembled.

I loosened the 10-32 x 3/8" SCHS that attaches the lead screw support to the cast base. The measured vertical misalignment changed to 0.000,1 inch fall across 2.75 inches from left to right.

Karl