How Do You Reduce Backlash (end play / slop) for the "Y" Axis of a Sherline Metric Lathe?

Rob Martinez

NAWCC Member
May 3, 2013
167
17
18
Country
The Sherline manual and videos show how to reduce backlash in the lathe slide (X axis) but not how to reduce it for the Y axis (for the 24" metric leadscrew). Following their video, I adjusted the handwheel control for the main leadscrew shaft (basically re-tightening the control on the post) as you would for an older frequently used machine, but this machine is new so it made no difference (did not work loose in the first place). Technically, the end play comes out to be .008 mm (8 increments) which is at the end range of acceptable for Sherline so I didnt want to get with the company yet assuming they will say to just live with it. Any info will be greatly appreciated...
 

glenhead

NAWCC Member
Nov 15, 2009
1,211
231
63
63
Williamson County, Texas
Country
Region
The short answer to your question is "you don't without adding a gizmo". For any lathe or mill most people learn what backlash is and how to work around it. Even with a gizmo you'll have to work around some backlash.

Sherline specs are for 0.08-0.13mm of backlash. (You slipped a decimal place there.) That means yours is at the lower ("better") end of their specifications. As a point of reference, the y-axis backlash on my two-year-old 4410 is 0.09mm. There is a gizmo you can order (Sherline P/N 4417Z/4417ZM) to adjust it down into the 0.03-0.05mm range, but in general you have to learn how to anticipate backlash and work around it. In other words, you might be able to cut it in half, but you'll still have backlash. It's inherent to the system. Even with a ball screw you have a bit of backlash, supposedly down in the 0.01mm range.

The way you deal with backlash is to always measure the same direction for a given operation. If you're cutting right to left to a shoulder, run the carriage past the backlash point and come back in right to left. It's really easy to feel when you've taken up the backlash. Once you've fully engaged the lead screw you're into the precision of the machine. In other words, you can be pretty danged sure that moving 0.03mm is really moving 0.03mm (or close enough to make no nevermind). You can accurately measure right movement to right movement or left movement to left movement, but you can't accurately measure a right movement relative to a left movement because of backlash.

Let's say you're cutting a 1.2mm-wide slot in a piece using a 1mm-wide cutoff tool, and you want to position it relative to a feature you've already turned. You know the left edge of the feature is at X on the handwheel scale, you want the right edge of the slot to be 2mm away, and you have the left edge of the cutoff tool exactly at the edge of the feature. You can get the left edge of the slot pretty much exactly where you want it by cranking the handwheel X + 3.2mm: 2mm for spacing plus another 1mm for the width of the cutoff blade, plus the additional 0.2mm to cut the 1.2mm-wide edge. Cut the slot, back out, move the carriage to the right "a ways" (my habit is to back it about a turn), and then move it back to the left to the X + 3mm mark. You've eliminated the backlash by going way past the extent of the backlash and you've re-engaged "precise", so you can cut the right edge of the slot with a high degree of certainty it really-truly is 2mm from the left edge of the feature.

I hope this makes sense, and hope it didn't go into TMI.

Glen
 

Rob Martinez

NAWCC Member
May 3, 2013
167
17
18
Country
Did I understand you correctly -- Sherline specs are for 0.08-0.13mm of backlash -- is that 13 of those little tick marks? Wow, thats a lot of slop... I thought it was much less than that. But really? 13 of those hash marks?
 

measuretwice

Registered User
Jul 28, 2019
99
36
18
Toronto
Country
.008mm? click your heels! Are you sure its not .08? .08mm I would have thought is a low amount of backlash

Sorry to be stickler, but the leadscrew on a lathe is the Z axis - Z on machine is the axis parallel to the spindle. (following the convention makes it easier to all be on the same page)

I don't know your lathe, but here's commentary on backlash and why its not much of an issue. You've got two sources of it, how the screw is mounted to the frame, and clearance between the threads on the screw and nut. Without some cleranbce, it wouldn't turn - so backlash is omnipresent (for this type of screw). The first source can sometimes been adjusted or it can eliminated with conversion to preloaded thrust arrangement (e.g. angular contact bearings) but its unlikely you want to go down that path The second source, clearance, can be eliminated with an antibacklash arrangement or ball screws...again a big job with little advantage.

You're going to find backlash in any machine without the above noted fancy arrangements - i.e. almost all machines will have backlash. The only time its really an issue is if climb milling, so we (mostly) don't do that on manual machines (without antibacklash nuts). For everything else, whenever you need to rely on a number on the dial, make sure you always approach it from the same direction as Glen describes.

As there has to be some backlash, its doesn't much matter - i.e. as you have to negate it by coming at things from the same direction, how much there isn't mission critical. The only time to be concerned about how much backlash there is, is when buying a used machine simply because its often a good indicator of overall wear. imo thats where the backlash is bad thing comes from.
 

karlmansson

Registered User
Apr 20, 2013
2,764
179
63
Linköping, Sweden
Country
On some larger mills people use what is called a "split nut mod". Basically this means sawing a leadscrew nut in half and then adding another, adjustable mounting for it so that you can separate the two halves of the nut in order to take up the backlash. This can of course cause problems because as with overall machine wear that you will notice with gibs being tight to varying degrees depending on where on the travel you are, the same will apply to a leadscrew. It will be more worn where it's most used.

There are very few operations that I know of that actually require you to work with zero backlash. Working with form tools might be one of them. On a mill it can become troublesome when reversing the feed direction. The general workaround is a DRO as I understand it. If you know where your carriage is at all times, backlash doesn't matter.

Regards
Karl
 

Jerry Kieffer

NAWCC Fellow
NAWCC Member
May 31, 2005
2,930
553
113
wisconsin
Country
Did I understand you correctly -- Sherline specs are for 0.08-0.13mm of backlash -- is that 13 of those little tick marks? Wow, thats a lot of slop... I thought it was much less than that. But really? 13 of those hash marks?

Rob
With the metric lathe the handwheel is calibrated by 100 divisions per rotation therefore 13 divisions seems like a lot.

However in reality it's only 13/100 mm or about .005".

Jerry kieffer
 

UncleDoc

NAWCC Member
Apr 4, 2020
191
29
28
56
Colonie, NY
Country
Region
Rob
With the metric lathe the handwheel is calibrated by 100 divisions per rotation therefore 13 divisions seems like a lot.

However in reality it's only 13/100 mm or about .005".

Jerry kieffer
I have about half that backlash with my 4530 lathe. I was able to minimize it by following the excellent set-up instructions from Sherline. At first I was concerned feeling how much backlash there was (originally about 0.13mm as the OP), but once you figure out how small it is and how to manage it, it's just a "thing."

Duane
 
  • Like
Reactions: etmb61 and glenhead

Forum statistics

Threads
165,506
Messages
1,440,758
Members
86,274
Latest member
cicastol
Encyclopedia Pages
1,101
Total wiki contributions
2,873
Last edit
Weekly News 7/7/19 by Tom McIntyre