Help identify the mystery set screw

Kelly, when I searched the downloads section "Atlas 3996" I find these four documents: so none with the exact title you show.

That's because shortly after my post Robert updated the file from rev 12 to rev 13. He must have delete the previous revision when he did. Not sure of the difference.

Best,
Kelly
 
When tensioning the belts do you pull the handle up or do you move it down to tension. I googled pics and see the handle in both of the positions. I will assume that to tension you pull up and move down to release.

I pull down to tension because I know that's the direction the lower pulley moves and it's more intuitive for me. One other related topic while you're setting that up, you mght want to consider the three belts and order of tensioning.

I tensioned the dual belts to the spindle first because that moves the top outboard cone pulley downward and changes the centerline distance to the other outboard, adjustable cone pulley. When I had my spindle apart I took the opportunity to replace those belts. I bought a matched pair of 43" x 1/2". Not sure if that's exactly what is called for but I used one of the online belt length calculators and worked well for me. Then I did the same for the cone pulleys and it called for 41" x 1/2". I haven't bought the third belt yet because I don't have the two step motor drive shiv yet so I just temporarily used a single groove pulley and belt I had laying around, but adjusting the motor belt tension is the last of the three, otherwise it's a tail chasing exercise.

Best,
Kelly
 
Last edited:
Either I'm getting too tired to be writing or too old, but Kelly had already answered the tension lever direction question so just disregard the next small paragraph.

@MHW, are you referring to the 3986 or the 3996? I don't really know about the 3986 but on the 3996, no. You push the outside-mounted lever down to tension and pull it up to release. That's the way that mine was when I bought it soon to be 40 years ago. Be advised that the thing doesn't want to stay all of the way up. I thought about trying so see whether or not I could reverse that but concluded that it was more important that it have no tendency to come loose while running than while not running.

The (actually two) other parts books were for a 3986, not a 3996. The 3985 and 3986 were the 24" and 36" bench models. The 3995 and 3996 were the 24" and 36" cabinet or as Atlas or Clausing called them, pedestal models. The only differences between the 24" and 36" were the bed, lead screw and rack lengths and part numbers. Besides the cabinet or no cabinet, the only other differences between the bench and cabinet models were the rear-mounted or the underneath mounted motors and 8-speed countershafts. And the spindle pulleys. The bench models had the 4-groove cone pulley and the cabinet model had a 2-groove one. The cabinet model also had two spindle drive belts, both the same size. Other than those points, the four lathes were exactly the same.

I spent some time yesterday deleting the extra 3986 manual and the correcting a few errors in the remaining ones. Most errors might be considered only cosmetic. But the stub sticking out of the top of the cross feed nut was drawn almost as large as the exposed part whereas it is really about one-half the diameter. The carriage gib was drawn as about half as long as it really is.. And I also replaced the bed and lead screw page with the one that I cleaned up and corrected several years ago for the 3996. For one thing, there is a single socket head screw shown above the bed. It was added during production to help hold the QCGB on. And if you don't know about it and try to remove your gear box, you will be pulling your hair out trying to figure out why it won't come off. It is well hidden down inside of the headstock. Anyway, I think that the two manuals are now both correct. Anyone who had previously downloaded either one should download the revision.

Back to the belts for the 3996, the one for the cone pulleys and the motor belt are standard FHP belts. 4L410 for the cone pulleys and 4L360 for the motor. The two belts for the spindle are A42's, not 4L430's. These are for some reason 43" long but the "A" belts are called Multiple V-belts, have a tighter length tolerance or variation, and are designed to run with up to either 8 or 12 belts in parallel. These are what came with my lathe new and are the same as what I got from Clausing 3 or 4 years ago.
 
Last edited:
Back to the belts for the 3996, the one for the cone pulleys and the motor belt are standard FHP belts. 4L410 for the cone pulleys and 4L360 for the motor. The two belts for the spindle are A42's, not 4L430's. These are for some reason 43" long but the "A" belts are called Multiple V-belts, have a tighter length tolerance or variation, and are designed to run with up to either 8 or 12 belts in parallel. These are what came with my lathe new and are the same as what I got from Clausing 3 or 4 years ago

Well this explains it. Before I bought the 43" spindle belts I searched the subject here and saw the 42" part number most frequently cited, but when I measured it, that would have put the upper spindle at the extreme upper end of the slotted travel range and looked like it would be very hard to mount the bearing with the belts in place or get the belts on once the bearing/pulley was mounted. When I measured and calculated the length of 43" looked better so that's what I used.

When I calculated the length required for the outboard pulley with the 43" spindle belts in position, I got 41" so same there. A longer than required spindle belt would consume the adjustable range for wear on both of those but I still must have at least 3/4" available which per side would be 1 1/2" of belt stretch, and I don't think that's going to happen. Unless something damages them, these new ones will likely outlast me.

Curiosity question: Do the 10" lathes have shorter spindle belts or taller bed feet to account for the height difference to the spindle?

Best,
Kelly
 
On the spindle belts, if you haven't already installed and run the 4L430's, I would recommend taking them back to wherever you bought them and exchanging them for A42's. The 4L430 would be OK if there were only 1 but the FHP belts aren't speced to run belts in parallel like the Multiple V-Belts are. They may be OK or you might find one of them to be noticeably looser than the other.
 
....They may be OK

The lathe is cutting chips with them in place. Must have gotten lucky. They seem very uniform in length and tension. Having disassembled and reassembled several spindles, it doesn't bother me if need be but don't think that will be the case.

Take care,
Kelly
 
I looked through the section on V-belt drives in Machinery's Handbook and if the two belts have approximate;y the same tension, they should be OK since they are already installed.
 
Back
Top