Help please? Threading on a 618

[machinistnoob]

Look at the half nuts and check for wear. Also see if they fully engage. I don’t know on the 618 , but the 10 inch atlas uses the half nut for regular turning and will show wear sooner than other lathes

Thank you for your input. That was one of the first places I looked. I can't gauge them but I didn't see anything abnormal on the half nut or lead screw. I tried Wobbly's last bit of advice, and threading went a little better but stil unsuccessful. At this point, I think I'm going to change everything and try a different TPI and see if it behaves the same way.

 

[RJSakowski]

The 618 has five positions where the half nuts will engage between marks. The thread dial and gear are pressed on the thread dial shaft so there is no adjustment but if the PO had disassembled the thread dial assembly and then pressed the gear back on, he could have reassembled it a half tooth off. Then the half nuts could be engaged a half tooth before or a half tooth after which would result in different threading paths. If there is some slop in the lead screw or the thread dial gear is not fully engaged with the lead screw, that would create even more slop. Check to make sure that the mark on the dial aligns perfectly with the index mark on the housing. If it doesn't you can make a temporary mark elsewhere on the housing that lines up perfectly. It doesn't matter where as long as the two marks are perfectly aligned.

If your thread dial is out of sync, it can be corrected by pressing the shaft out of the dial and repositioning it. To reposition it, place the shaft with gear into the housing and set the housing up for proper engagement of the gear with the lead screw. with the gear pressed tight to the housing, set the dial on the shaft and rotate to align with the mark on the housing. Press the dial on enough to prevent it from changing position. Then remove the whole assembly and fully press the dial on the shaft. Check the alignment afterwards and repeat if necessary.

Also, best practice would dictate that you run the carriage past the start of the thread and then engage the half nuts after all backlash has been taken out. Another point, trivial but worth mentioning, is that you do not want to reverse the feed using the forward/reverse tumbler. Once threading has begun, it must stay engaged until the thread is complete.

 

[RJSakowski]

Years ago, I replaced both the half nuts and the lead screw on my Atlas/Craftsman 6x18, thinking they were worn. Later, I inspected my old half nuts and found them full of grunge, a mix of machining dust and oil, which prevented them from closing properly. After cleaning them, the old half nuts were as good as new.

 

[markba633csi]

Must be a backlash issue- the gearset looks ok according to the chart
Something is shifting; either the leadscrew or the halfnuts
Do a scratch pass as before then on the second pass pull the tool back and watch as it hovers over the first pass
You'll probably see something jump or shift

 

[WobblyHand]

When figuring things out like this, I have installed a very fine tip sharpy instead of a cutter. No, it's not as good as the real thing, but it helps one understand things without going through as much metal. The sharpy will wipe off with an alcohol wipe. When I was fooling around with my electronic lead screw, I found using a marker helped me figure out what was going on without going through stock.

Another alternative is to chuck up some pvc pipe. I used 1/2" pvc pipe when I was first learning how to single point. A 10' stick is still pretty cheap, or maybe you could even get some cut off pieces for far less. This helps get your timing down without worries or fears. In any case, I found it to be quite useful when starting out.

 

[mattthemuppet2]

there are two collars on the 618 lead screw - one to the right of the headstock end leadscrew support and one to the left, behind the large end gear. There may also be a spacer between the left most collar and the gear, depending on what tpi is being cut. The rightmost collar at least is secured with a set screw, the left one may be the same or pinned, it varies. I was chasing my tail all over the place until I saw the gap between the rightmost collar and the leadscrew support. Once I snugged up both I had no double tracking issues.

 

[markba633csi]

I'm not a big fan of Atlas gearing for coarser threads, they necessarily "gear up" the ratio and it puts a lot of strain
on the geartrain and banjo, especially the smallest gears
Make sure the leadscrew and carriage move easily without a lot of drag
Also make sure the gears themselves are clean with no chips stuck in the teeth

 

[machinistnoob]

there are two collars on the 618 lead screw - one to the right of the headstock end leadscrew support and one to the left, behind the large end gear. There may also be a spacer between the left most collar and the gear, depending on what tpi is being cut. The rightmost collar at least is secured with a set screw, the left one may be the same or pinned, it varies. I was chasing my tail all over the place until I saw the gap between the rightmost collar and the leadscrew support. Once I snugged up both I had no double tracking issues.

I will check on that. Thank you!

 

[bill stupak]

Have you checked to see if your tool post is moving?

 

[RJSakowski]

There are two screws behind the half nut lever. These mount the half nut cam plate. I noticed that with the half nuts engaged and moving the carriage with the hand crank, I could move those screws from side to side slightly.

Normally, this would be taken up with lash but it is possible if the screws are sort of tight and the carriage moved to the left against the engaged half nuts, you could move the half nuts slightly to the right and they might stay in that position while making a threading pass. One the following pass, if the carriage was moved to the right against the engaged half nuts, the half nuts would move slightly to the left and a threading pass wouldn't track with the previous pass. Ironically, either loose or tight screws wouldn't see this issue. If the screws were loose enough to allow free movement,the movement would be accounted for when taking up the backlash before starting the thread

Tightening the screws eliminated the movement on my 6x18.