What's new
What's new

Another new guy with threading issues, SB 9B

Slow Dave

Plastic
Joined
Jan 25, 2018
This is my first post here but I have been reading a ton of old posts trying to figure out a few things. I picked up what looks like a 1946 9 B from an auction guy who had it sitting in boxes in his wharehouse for years. It was a bit of a gamble but I figured out how to put it together. The lathe ways look like it has had very little use.

I picked up a real SB thread dial from Ebay. The dial gear has 32 teeth. My lead screw is 8 tpi. My stud gear is 32 teeth as is the screw gear. I am trying to cut 8 tpi threads on 1" A36. When I thread using the dial each pass is slightly off to the left. After a few passes I have a fine threaded (sort of) mess.

When I thread using reverse and leave the half nuts engaged the threads turn out perfectly.

The thread dial appears to fully engage and the gear does not look worn out.

At this point I am thinking it has to be either the thread dial gear is in fact too worn or my half nuts are not engaging identically each time. Due to the lack of wear on the lathe overall I suspect the issue is the thread dial gear. Does anyone have any other ideas as to what is going wrong? Does anyone know of a source for a steel replacement thread dial gear?

Thanks

Dave
 
I picked up a real SB thread dial from Ebay

Welcome...so the thread dial you have is not steel.? is it one that some one made.
 
if you drop on the very same TD line is it still messing up?
Are you sure your tool bit is not moving in the holder or the holder moving?
Gibs close/tight?
Part moving in chuck or collet?
Any end play in lead screw?
 
Just my guess, but it would seem it would have to be related to the half nuts themselves, some sort of "play" that's only manifested when you disengage/re-engage. Are you re-engaging on the exact same number/line? If not I would try that, its going to slow down the cutting but I think that would be a way to rule the gear itself out-or in...

Some minor wear on the gear isn't going to matter, to the point where the half nut would engage on a completely different thread than it should. It's just an indicator...
 
If your lead screw is 8tpi and you are cutting 8tpi, the one thing that we can say with absolute certainty at this point is that the problem is NOT with the thread dial. You should not even be using the thread dial when cutting 8tpi, engage the half nut anywhere it will engage and you'll be fine.
Lots of other things can be causing your problem.
It's a change gear lathe, are you sure you have the gears set right to cut 8tpi?
Are you infeeding with the cross slide or compound? If the compound, do you have it at the correct angle or is it 29* off the bed instead of 29* off the cross slide?
You say you just put the machine together so I assume it's untested cutting threads. Are the half nuts in good shape and working properly? Does the lead screw have excessive end play?
Could the work or tool be moving?
Maybe post a couple of pics of your set up if none of the above helps.
 
You should be able to just look at the thread path with a loop so not cutting a whole bunch of scrap parts till you figure it out..perhaps just chuck up a good thread screw bolt and stay .015 away..and travel a number of times.

Good point Derek to be on the right compound angle..

Very high cutting forces on no-rake tool bit. I like a little side and back rake for mild steel.

Be sure tool bit is near center so clearance angle is good. Be sure tool holder is is good angle so bit clearance is good to the part..

Going straight in (not at about <30) the drag on the follow side of the tool bit is huge..often enough to make something move.

A negative tool bit will have huge tool pressure if used for mild steel..
 
Wow, thanks for all the responses. I thought it might be the gibbs so I already tightened them. I also thought it might have something to do with my tool grinding so I purchased a pre-ground threading tool. The threading dial is a South Bend factory made dial with a 32 tooth gear. Since it threads perfectly if I reverse it and don't disengage the half nuts it seems to me there are only two variables left, the dial or the half nuts. Right now I am thinking it is probably the half nuts are not engaging exactly the same each time they are engages either due to wear or being packed up with chips and old oil and grease. Most likely they are full of gunk since the rest of the lathe has such little wear. I will tear it apart in the next few days and have a look.

I believe I have the change gears right and I have tried different change gears for different TPIs but I have come back to 8 TPI for the reasons stated above and it is really easy to see if the second and third passes are right.

I grew up around machine shops as both my father and grandfather were machinists. I have only been trying to learn for the last year or so both on a milling machine and lathe. I really regret not taking advantage of the leaning opportunity that was available to me as a teenager.
 
What is your compound set at? Or are you feeding with the cross slide?


And this wouldn't be one of those blazed carbide tools would it?...you have to final shape those.
 
The compound is set at 29.5 and I am moving the tool in with the compound not the cross feed. The tool works perfectly when I thread by withdrawing the tool and reversing so I doubt the problem is related to the tool.

Thanks

Dave
 
Yes a half nut full of chips might matter.
What happens with just a touch thread , perhaps .010 deep then travel over that same path 6 times with no in feed does the tool path get wider/change?
If you hold a straight edge vertical to the tool bit lead and follow side as it is installed can you see clearance on both sides?
Is it a positive rake fingernail shaving sharp edge tool bit?
If you drop on the very same line does it still error?
You are not taking crazy deep cuts?
is your compound reasonably centered so not hanging way over board to the far out position..towards lathe center.

You said feeding at about <30 good.
With feeding straight in (cross only) the tool bit follow side can act like a nut running on a thread and can actually push the carriage ahead so taking up all the slack in the screw to make the tool bit go ahead of the intended path..and it can also rub hard enough to move the tool bit or the tool holder forward.

Mostly with knowing my lathe 30 is good I just set at 30*.. When unsure I set back a little.
 
Try 27 degrees...the scales can be off a bit...

wait- you threw me with the withdraw and reverse part....you Always have to with draw the tool no matter what.

I will add that the mounting for the thread dial might be slipping(it's jus a set screw on a round pin so not very secure) or your toolset/workpiece moving...someone could provably figure it with a pic.
 
When cutting threads at a pitch that is a multiple of the pitch of the leadscrew, no thread dial is required. It this case, since it has an 8 pitch leadscrew, the thread dial is not needed for 8, 16, 24, 32, 40, Etc.pitch threads.
Something else is the problem.
Ted
 
Thanks again for all the help. Yes I withdraw the tool at the end of each pass using the cross slide. Once everything is moved back I re-set the cross slide to the original setting and then I adjust the tool for the next pass using the compound. I have tried several cutting depths from very light to fairly deep. At one point I was wondering if I was cutting too light. When I was using the reverse with the half nuts engaged I would back off the cross slide at the end of the pass, reverse with the nuts engaged and then re-set the cross slide and adjust the depth of cut with the compound. I have not tried using the dial and doing a second pass without moving the compound yet. Maybe I will get a chance to try that tomorrow.
 
Guess you have to go through everything..
Head stock take up nut(for end play)
Cross and Compound everything tight and solid to the screws..
Cross and Compound lead screw nuts properly fixed.
Lead screw collar set for no end play.

What happens if you only out feed with the compound?
and then only with the cross? (yes still only infeed with compound)
This to see if one is not properly fixed.
Can you test the lead screw for end play? (.002-.003 is OK)
are you sure the bit has clearance when installed. (yes you know the lead side does because it cuts and the follow does not matter much because you fed at<30))

Might zero cross and compound dial to something perhaps tool bit to the chuck OD..then after running a hey-wire thread go back to see the zeros (or numbers) are still there. if one is off that may be the problem. (to do this you may zero to the part..then go see where the dials are touching the chuck and make a note of it. so the end check will be again off the part. yes that could come from tool bit moving so you need know that is tight.
 
Are you sure you're cutting 8 tpi? Count the teeth on all the gears and make sure the spindle rotates exactly once per revolution of the leadscrew. If you were off by a tooth on one of the gears, it might look right, but isn't 8.
 
Are you sure you're cutting 8 tpi? Count the teeth on all the gears and make sure the spindle rotates exactly once per revolution of the leadscrew. If you were off by a tooth on one of the gears, it might look right, but isn't 8.

Yeah, I've begun to think this is the most likely cause myself. It would cause the exact problem the OP is having and yet be difficult to detect because everything would "look" right.
 
Dave,
The thread dial index plate is not loose on the shaft, right? There's a set screw that holds it in place.
 
His gearing is right for a 9" model B and as Ted noted anywhere the halfnuts close should be fine as its 8tpi.

A good pic will clear things up I bet.
 








 
Back
Top