Play in cross feed and compound

[ARC-170]

On my Atlas Craftsman 101.07403 I'm not able to accurately turn a diameter. When I get close to the final measurement, if I rely solely on the DRO I'm usually off by a few thousandths. I've been able to make a deeper or shallower cut by moving the cross feed in or out. I'm thinking this is not good! Well, I noticed the cross feed has about 0.018" of movement and the compound has about 0.012" of movement. I've tightened everything down (I think), so I'm wondering if this is indicative of screw wear.

Is there a way to lock the cross feed and compound in place? Maybe a mod I could do?

As a workaround, when I get close, I take measurements, then move the cutter so it just starts to cut, then look at the DRO and move in the required amount. I was able to turn a steel rod to within 0.001" of spec. However, that was more luck than skill; I stopped short to check and it turned out to be what I needed.

Any advice or solutions?

 

[mickri]

All machines have some backlash. No way to eliminate that. There will also be spring which I believe doesn't show on a DRO. But then I don't have a DRO so I don't know. My lathe has 0.007 to 0.008 backlash on the compound and the cross slide.

I often have the same problem as you are having when turning to a diameter. When I am turning I never use the compound and bring it as far back as I can to eliminate flex in the compound. I line it up with the cross slide Then snug the gibs down so I can't move the compound. What I do after taking a cut is to back off the cross slide at least 40 to 50 thousands. This ensures that when I move the cross slide back in for the next cut I have taken all of the backlash out of the cross slide. I also do all of my measurements with a micrometer. Never use a caliper when I need a really accurate measurement. Even with this I rarely hit a diameter spot on. If I can get within 0.001 I am happy.

 

[mike44]

Not familiar with an Atlas lathe. Assuming the compound and cross feed are adjusted properly , at least as far as you can then. The fix you did was proper. I have digital scales on the X axis and Z axis ( tail stock) but not on the Y axis. I start by turning the cross feed til it moves . Then continue in the same direction , clockwise til the cutter touches the work. I then set the micrometer to 0. This does not actually eliminate back lash but works around it. You did the same thing but with your DRO. I do not think the lead screw has any thing to do with the back lash on the X axis.
Possibly a worn lead screw on the cross feed would cause back lash but I doubt it. Not positive but I think the cross feed lead screw would not get enough wear , unlike the X axis lead screw. If there is an Atlas forum, you should be able to answers there.
mike

 

[AGCB97]

Is there a way to lock the cross feed and compound in place? Maybe a mod I could do?

If the gib screws are easy enough to get at you could replace one of them with a thumbscrew, even a DIY one. I've done that on my 100 year old Springfield.
Aaron

 

[markba633csi]

Sounds like you need to re-adjust all the gib screws on your carriage, including the compound. It's a fiddly job and takes time, but the results are worth it.
You want to find the point where the parts are just tight enough but not too tight to prevent smooth sliding. One or more of the gib strips may be worn and need to be replaced.
All lathes have backlash so you just take that into account when using the machine.
-Mark

 

[JPMacG]

I lock the gib screws on the cross feed and compound prior to my final cut. Also, I make practice cuts to see the actual cut versus the amount I move the cross feed (or compound). This is particularly important for very light cuts. For example, a change of .001 on the dial might make an actual cut of .002. I think this is because of flex. I can achieve .001 accuracy on diameters with some care, but it is not easy on these light machines. If I need better than .001 I use a file.

Note that just the act of locking the gib screws can cause movement. The whole process requires practice and patience.

 

[craniac]

I have the same problem on my 1937 Atlas and have been able to get reasonably accurate diameters by following pretty much all the advise given above: tight gibs, replace a gib screw with something easy to tighten, use the cross slide not the compound for the most part. I would add that you can sometime get some additional backlash if the nuts on the cross-slide screw are loose at the dial. There is a thread on that somewhere that I read. Might try searching that.

This only thing I differ with in the above postings is the statement that youdon't see wear in the cross-slide lead screw. My lathe is quite old though, but you can see wear in the middle of the screw, and the nut, being softer, has considerable wear- even more that what you're seeing. This is a well documented problem with solutions being getting a replacement nut. They are available but not cheap around $40-$50.

I don't expect to work to tenths- though I sometimes get there by accident. It requires constant awareness of the backlash and being careful to back out each time. I also use a dial indicator and don't rely on the cross-slide dial, which I can barely see anyway.

You can do it.

Tim

 

[Bi11Hudson]

Having learned to use a lathe BC (before computers), I don't have any use for an electronic gizmo DRO. Larger dials would serve the same purpose. Using trig on the compound helps as well. Mine stays at 29* most of the time. As much to keep the hand wheel out of the way as threading. That gives .866 feed, an odd size. Adjusting to 60*(M/L) gives 1/2 the feed. As the angle increases, the feed decreases. But it gets hard to work in your head.

The Atlas machines, including the Craftsman (mine is a 101.27440) have a rather aggressive longitudinal cut. It doesn't relate directly to your question, just something else to keep in mind. However, with a finishing cut, you might consider a smoothing pass before moving the cutter. This in addition to the comments above about tightening the gibs. My machine has a lot of wear but I compensate for it by tightening the gibs for smooth work. And never depending on backward adjustments. I always allow for backlash on smooth jobs.

Not the proper video, it just happened to be in memory. But can lead to the proper one. . .

.

 

[ARC-170]

The gibs are new and adjusted a bit on the tighter side. Everything else is snug, like it's supposed to be. I just find it hard to believe there is no way to lock these down.
I thought of a solution as I was drifting off to sleep about adding a lock screw to the bank of gib screws. There's room on the lathe and the parts would fit in my mill vise so I could drill them properly. I found a YouTube video where they guy did the same thing. Anyone done that with these lathes, or see an issue if I do it? I'd like to keep the gib screws out of this, so that means a new hole. Thoughts?

 

[jaek]

Stefan Gotteswinter had a very good video about how to cut accurate diameters in the presence of backlash: