Sheldon Lathe Theading Problems

I have a 1946, 11 inch Sheldon lathe, and I have a problem with it chasing threads. First, let me say that I am a member of the Yahoo Sheldon Group. It's been a long time since I've logged on there; now when I try, my malware protection program warns me of a possible malware on the site, so I'm not going to jeopardize my computer.

My Sheldon will not repeat a thread path. Example: If I set it up for 14 TPI, make a pass, and then go for repeated passes, it will not follow the original thread path. It's still cutting 14 TPI, but not following the previous cut; it can be off several thousands, or spilt the thread right in half. I even engage the half nut on the same number on the dial to eliminate any possible influence of using different numbers.

What I've done: I have dropped the apron from the carriage to inspect the half nut and everything there is tight and in its place. The dial face is tight and not slipping. I've checked all of the keyed gears in the threading drive train and each one I've checked so for is still keyed and tight. There is no slack, slippage, or excessive wear in the led screw or its thrust bearings. I have not been in the quick change box yet to see if anything there is out of whack. Before I go thru the trouble of dropping the quick change box, I wanted to put this problem before this forum.

I have not crashed or abused the lathe, nor do I do any hard, aggressive machining on it. Just light weight hobby style work; but one day it cut threads just fine, and the next day it will not.

Here's my question: Is there any other areas I should be looking at to determine why this thing is getting out of time/sync?

Thanks

I'd say something is loose. This may sound unlikely, but check your compound gibs for proper adjustment. If they're way loose you will be able to push the handle back and forth sideways and see movement. Do this with the compound fully retracted as this will amplify the movement at the handle end. While threading, are you only advancing the compound (doing the between pass retraction solely with the cross-slide)?

One other thing I've encountered is endplay in the leadscrew. There is an adjustment where the leadscrew enters the QC box (a threaded collar). Normally this artificial backlash is accounted for by the natural drag of the carriage, but if excessive it could affect where the half-nut actually catches the screw even though it looks like the same spot as far as the dial is concerned.

The day-to-day variation in performance you mention could be due to variables in your performance or due to things like cold temps (increasing lube viscosity, etc) or the actual position of the compound along its ways and so forth.