Strange threading problem

Jeff_G · Apr 25, 2020

I have a 9A from 1950. I set up a piece of scrap to test-cut 12 tpi. First cut was fine, second cut didn't register with the first cut. Checked the threading dial - 32 teeth and no play. Checked the headstock gear train 20-80-56. Got out the thread pitch gauge and measured 13 tpi. Rechecked the thread chart and verified that the levers were in the right slots.

Does anyone have any idea what could be going on? What do I check next? I'm baffled.

Thanks,
Jeff

johnoder · Apr 25, 2020

20 and 56 have their keys?

SBLatheman · Apr 25, 2020

Are you using the half-nuts ?

Jeff_G · Apr 25, 2020

Yes, keys and yes half nuts

Am I right to assume that the gears in the QC box are right, else the tumbler gears (correct terminology?) would not engage properly?

SLK001 · Apr 25, 2020

Jeff_G said:
First cut was fine, second cut didn't register with the first cut.

Checked the headstock gear train 20-80-56. Got out the thread pitch gauge and measured 13 tpi. Rechecked the thread chart and verified that the levers were in the right slots.

Doesn't make sense. How do you know that the first cut was correct? Was it checked with a thread gauge? The first cut should be a scratch cut, barely enough to make a mark, then CHECKED with the proper thread gauge.

Regardless of your thread pitch, your second cut should have lined up with your first. Since you were cutting 13 TPI, you should have started at the 1, 2, 3 or 4 on the thread dial. This is most likely why you didn't get the same start.

Please explain, in detail, your entire operation.

johnoder · Apr 25, 2020

I guess a good question would be when is the last time it worked properly?

A sure fire test anytime - engage half nuts and turn the chuck enough to take all lost motion out all the way from spindle to half nuts

Now stop and/or move nothing

Set a mag base DTI on the bed with its business end against the carriage

Mark chuck and turn it exactly one turn

The carriage MUST move the reciprocal of the thread being cut - like .083333" for 12 TPI

Jeff_G · Apr 25, 2020

John, Tried what you suggested. Gearbox set for 12tpi. Carriage moved .077. Almost exactly 13tpi. Also tried setting the QC box for 8tpi. Carriage moved .116 - about 8.6tpi.

Were there different threading charts depending on when the lathe was built? Maybe a previous owner got the wrong one?

Or could there be something metric going on? What is the pitch of a SB9 metric leadscrew?

The headstock gears were stamped on the rim with the number of teeth - I did not count teeth. Could there be an error ? Seems unlikely.

iwananew10K · Apr 25, 2020

its *acting* like it has a 3mm pitch leadscrew.
suggest counting teeth and measuring leadscrew pitch.

if it is an 8tpi leadscrew its turning roughly 1.08 times too fast.

iwananew10K · Apr 25, 2020

1.08 times too SLOW......for some reason it wont let me edit.

SBLatheman · Apr 25, 2020

Jeff_G said:
John, Tried what you suggested. Gearbox set for 12tpi. Carriage moved .077. Almost exactly 13tpi. Also tried setting the QC box for 8tpi. Carriage moved .116 - about 8.6tpi.

Were there different threading charts depending on when the lathe was built? Maybe a previous owner got the wrong one?

Or could there be something metric going on? What is the pitch of a SB9 metric leadscrew?

The headstock gears were stamped on the rim with the number of teeth - I did not count teeth. Could there be an error ? Seems unlikely.

The 9” metric machine has a 3mm leadscrew, and metric half-nuts, and a metric gearbox, and different end gears. I doubt anybody mixed them all up :nutter:

Show us a picture of your machine

Jeff_G · Apr 25, 2020

Picture here: South Bend 9A Lathe - Jeff's Old Iron Workshop

Lead screw is 8tpi.

I counted gear teeth. The only one I don't know the correct tooth count for is the one behind the stud gear. Anyone know how many teeth its supposed to have? Mine has 24 teeth. Its listed in the parts book as PT28NK1, but no tooth count shown.

johnoder · Apr 25, 2020

.077" is also very close to 2 mm

Carriage moved .116

And that is quite close to 3 mm

SLK001 · Apr 25, 2020

Jeff_G said:
Its listed in the parts book as PT28NK1, but no tooth count shown.

28 teeth .

Jeff_G · Apr 25, 2020

Problem appears to be the gearbox. One rotation of the spindle should rotate the screw gear by 20 teeth, if my math is correct. It does. Also, when set to 8tpi, one rotation of the spindle turns the leadscrew less than 1 full turn. So, tomorrow, I guess I need to pull the gearbox.

Jeff

SBLatheman · Apr 25, 2020

Jeff_G said:
Problem appears to be the gearbox. One rotation of the spindle should rotate the screw gear by 20 teeth, if my math is correct. It does. Also, when set to 8tpi, one rotation of the spindle turns the leadscrew less than 1 full turn. So, tomorrow, I guess I need to pull the gearbox.

Jeff

My guess is the leadscrew gear is on backwards (I’ve seen it several times)
BTW; on your web site, the year that South Bend Lathe was founded should be 1906
Ted

Jeff_G · Apr 25, 2020

Ted, Thanks for correcting me on the date.

When you say "leadscrew gear" are you talking about the 56 tooth gear on the left end of the gearbox? Or something inside the gearbox? Turning the 56T gear around would make no difference.

malcolm rypauf · Apr 26, 2020

Several years ago, I found the "cluster" gear in the Q.C. gearbox had some gears out of position laterally on the shaft.

michiganbuck · Apr 26, 2020

I’m not going to get into this because there are plenty of lathe guys here smarter than me.

*When your thread gauge a tickle cut shows wrong then you should stop on that part and find an existing finished thread sample to gauge your lathe problem.

Chucking up or centering a thread sample and eyeball or loupe setting a bit to the thread major diameter, then watching a number of travels beats cutting a bunch of wastes material.

https://www.amazon.com/Stainless-Cu...87901199&sprefix=thread+gauge,aps,185&sr=8-11

With your thread pitch gauge a tickle cut should only make a light scriber wide line so it does not scrap the part it it is wrong. With a thread pitch gauge it is easy to see an error with only that light cut.
For 10 bucks every one should have a thread pitch gauge set.

SBLatheman · Apr 26, 2020

Jeff_G said:
Ted, Thanks for correcting me on the date.

When you say "leadscrew gear" are you talking about the 56 tooth gear on the left end of the gearbox? Or something inside the gearbox? Turning the 56T gear around would make no difference.

The leadscrew gear is in the gearbox on the end of the leadscrew.

Paolo_MD · Apr 26, 2020

michiganbuck said:
I’m not going to get into this because there are plenty of lathe guys here smarter than me.

*When your thread gauge a tickle cut shows wrong then you should stop on that part and find an existing finished thread sample to gauge your lathe problem.

Chucking up or centering a thread sample and eyeball or loupe setting a bit to the thread major diameter, then watching a number of travels beats cutting a bunch of wastes material.

https://www.amazon.com/Stainless-Cu...87901199&sprefix=thread+gauge,aps,185&sr=8-11

With your thread pitch gauge a tickle cut should only make a light scriber wide line so it does not scrap the part it it is wrong. With a thread pitch gauge it is easy to see an error with only that light cut.
For 10 bucks every one should have a thread pitch gauge set.

This is pretty much drifting from the true topic. Probably triggered by a few things implicit in Jeff's first description. First, when you're cutting (or, in this case, expect to cut) an even TPI thread, you can re-engage the half nuts at any of the marks on the threading dial (if the desired TPI is a multiple of the pitch of the leadscrew, you can engage at any position, disregarding completely the threading dial). "Luckily", on the piece of scrap Jeff used to test the threading, he engaged once on a numbered mark and once on a mark between numbers. This produced a second lead on the (actually 13 TPI) thread.

It looks like he has never cut a thread from that specific lathe before and assumed that gear train and QCGB were assembled correctly. But rest assured that he has the proper tools to measure pitch, etc.

Paolo

Strange threading problem

Hot Rolled

Diamond

Titanium

Hot Rolled

Stainless

Diamond

Hot Rolled

Diamond

Diamond

Titanium

Hot Rolled

Diamond

Stainless

Hot Rolled

Titanium

Hot Rolled

Cast Iron

Diamond

Titanium

Stainless

Similar threads