How Much Backlash On An Atlas Mfc Mill X Axis?

JPMacG

Active User
H-M Supporter - Silver Member
Joined
Mar 22, 2014
Messages
841
I have about .030" backlash on the X axis of my Atlas MFC. In comparison, my Y axis is much tighter - maybe .010 backlash.

How much is typical for the X axis? I'm wondering if I should replace the nut.
 
Hi Jon,
Backlash is pretty normal on any machine unless you have a machine with backlash eliminators fitted. Personally I have never had such a machine.
How much is one full turn of the screw? If it is .100" then 30% is a fair amount.
A golden rule for milling is "Never Climb Mill" so backlash is usually not a problem as the cutter is always pushing against the screw and nut.
If you accidentally go past your mark when setting a cut just go back 1/2 a turn and come forward again.
I'd normally expect the Y axis to be wear out before the X axis but X may have been previously replaced. I'd check it out visually and make sure everything is done up properly and if the nut looks severely worn. It probably come down to how much you are willing to spend and if you want to feel good about a nice tight machine :) howerver I think it is not esential to replace.

David
 
Jon,

I never can keep straight which axis is X and which is Y, except that I think I remember that they are not the same on a lathe as on a mill. But if you are calling left-right "X" and IN-OUT "Y", then on a used mill, more backlash would be expected on the X than on the Y. Say you are cutting a longitudinal slot in 100 shafts. In most cases, once you get the IN-OUT set and locked, you won't touch it again until you are finished. So the cross-feed nut will typically have more wear than the IN-OUT one.

There are three possible sources of backlash in either feed system - running clearance between nut and screw, end float in the screw, wear in the nut and wear on the screw. To determine what part or parts need to be replaced, you first need to adjust the screw end float. This is best done with the feed nut disengaged from the feed screw. But you can't do that on the main cross feed because the screw has bearings at both ends and the nut runs between the bearings. You can do this on an Atlas lathe because the cross feed has only one bearing, and when you crank the nut off of the screw, you still have an adequate engagement between the cross slide and the carriage dovetail. I don't happen to know whether you can crank the IN-OUT nut off of its screw or not.

Anyway, first adjust the feed screw end float to .001"-.002". Or to as low as you can get it without there being any significant drag on the crank when you rock it back and forth. Then measure the backlash out in the middle of the slide's travel and right near the end of the travel. If there is a significant difference between the two measurements, the feed screw is worn. Any backlash in excess of 0.005" to 0.010" near the stops is feed nut wear. Anything over about 15 thou near the stops would indicate that you should replace the feed nut.
 
My Understanding of the axis names on a milling machine table movement are (when standing in front of the machine)
X is Away from you and towards you, Y is left and right, Z is up and down.

From a CNC point of view they have + and - values but that I can't remember. :)
I think Digital Read Outs also label them like this as a standard.
 
Sorry, I should have been more clear. By X axis, I mean the left-right axis when I face the table. What I call the X axis is the axis that has the auto-feed. Maybe I should be calling it the Y axis.

I took the nut and lead screw assembly off. The nut is a sloppy fit - I can wiggle it by hand on the lead screw a distance of about 1/2 thread. But I also notice that the lead screw threads are rounded - the sharp edges that I would expect to see on a Acme thread are not there. The rounding is worst in the middle of the screw and better near the ends. I moved the nut to the ends of the screw and the amount of slop seemed about the same as in the center of the screw, so maybe the screw wear is not contributing much to the overall slop.

I will reassemble and do the tests that Robert described. If the problem is the nut I will probably decide to replace it. If the problem is both the lead screw and nut then I may just decide to live with it for awhile.

I wonder if Clausing still has these nuts.
 
Sorry, I should have been more clear. By X axis, I mean the left-right axis when I face the table. What I call the X axis is the axis that has the auto-feed. Maybe I should be calling it the Y axis.

I wonder if Clausing still has these nuts.

JP.

The cross slide nut is 10F-19, so I would assume that Clausing still has this. The table nut is M1-11, so they may or may not.

And I think that Pollardd is correct for mills - the power feed is on the Atlas mill Y-axis.

Looking at the mill parts manuals, it appears that the feed screw end float is adjusted by differential tightening of the S7-166 Nut (which is a threaded bushing) and either the 10D-262 Nut or the 3/8"-24 Hex Jam Nut (same method as on the Atlas lathes cross feed and compound feed screws).
 
Over the weekend I put the table back together. I snugged up the end float on the Y axis just to the point where the wheel started to drag. The backlash is about 7 mils - really nice.

The X-axis was terrible. The backlash was over 1/2 turn, 50 mils, and that was after tightening the end float as much as possible. I found a replacement nut and lead screw from a local used equipment dealer for a good price. It feels much better than the original, but it is not plug-and-play. The lead screw is about one inch longer than the original part. So I will need to make spacers to go under the wheels. Apparently Atlas changed the design over the years.

I tried to take apart the original lead screw, thinking that I might swap end pieces to fix the length problem. I drove the pin out but the two parts seem to be permanently pressed together. I'm thinking of using a torch to try to get them apart, but afraid I might ruin the part. Has anyone taken this assembly apart? How did you do it?

Thanks!
 
Jon,

The parts manuals do not show whether the table lead screw and lead screw extension got longer or shorter but as you have an MFC, they must have gotten shorter. But the M1, MF, M1A and MFA use Screw M1-35 and Extension M1-34 whereas the M1B, MFB, M1C and MFC use M1-35A and M1-34A. So something changed.

Also, The Feed Screw was only sold with the extension (presumably installed). I don't know why there was an extension instead of just making the feed screw and extension all one piece. But it may be that Atlas didn't intend that they should ever come apart. I would check carefully to see whether or not the screw would still work if you shortened the left end.
 
Thanks Robert. I think the extension was made removable so that the nut could be installed and replaced. The threads end in both directions, which prevents the nut from being unscrewed all the way off. I guess I would need to come up with a puller that would somehow grab the extension and the lead screw.

I did get a chance to compare the two lead screw assemblies more carefully. The new one is about 1/2 inch longer, not an inch as I said before.

I don't know the history of my mill. Although it has a tag that says MFC and has the right number of speeds to be an MFC, I know nothing of its history. It may be a Frankenstein's Monster with parts from various scrapped machines, MFC or other.
 
Back
Top