Hendey T&G Metric threading

Some are confused by "gear feed in thousands" in that table. 100 tpi is 10 thousands feed. What am I missing?

I had to think about this. It does seem confusing.
The OUTPUT of the Norton box is
either to a Leadscrew (and then to the half nuts)
or to the Feedrod (and then to the feed clutch lever).
So if you divide it out, the Feedrod moves the carriage
4.167 times slower than the Leadscrew. So that is where
those feed numbers per rev come from.

Now the other thing the chart says is
Belt feeds equal one-half of Gear feeds.
This is because the INPUT to the Norton box is
selectable through the change gears from the spindle
OR a Vee belt coming from the spindle. So the belt
feed ratio is 1/2 the change gears feed ratio.

I once had a thought to replace the Vee belt with
a tooth timing belt and pulleys, so it would not slip.
That way you could cut threads twice as fine as
indicated on the chart. But 240 tpi might be useless.
The Hendey T&G lathe is an amazing machine.


So don't forget, the feedrod is 4.167 times slower,
so 120 tpi (in threading terms) becomes 500 tpi (in threading terms).
Change the input to the Norton box from gears to belt,
and that gets cut in half again, so the finest feed on the Hendey
becomes 1000 tpi (in threading terms) or .001" per rev.
Make some diffraction gratings anyone ?

So can one cut metric threads with the factory
Hendey Norton gearbox with the 24:48 (1:2)change gears ?



1mm would be double 13 tpi, or 26 tpi in this case.
So 6-1/2 tpi would be equivalent to 4mm pitch.





So my math is

1" / 13 tpi = .077" pitch

2mm / 25.4 = .079" pitch


--------------------------


1" / 26 tpi = .039" pitch

1mm / 25.4 = .039" pitch


----------------------------


1" / 6.5 tpi = .154" pitch

4mm / 25.4 = .157" pitch

------------------------------


1" / 17 tpi = .059" pitch (my Hendey doesn't have that one)

1.5mm / 25.4 = .059" pitch

A workaround is to set the Norton box to 4 tpi, and select the
feed rod (not the leadscrew), which gives...

1" / 4 tpi = .25" pitch
.25" / 4.167 (feedrod reduction factor) = .060" pitch

This works by leaving the bed feed clutch engaged
and reversing the spindle to rewind each threading pass,
because you are feeding with the rack, not the leadscrew.

----------------------------------


1" / 20 tpi = .050" pitch

1.25 mm / 25.4 = .049" pitch


------------------------------


1" / 14 tpi = .071" pitch

1.75mm / 25.4 = .069" pitch





-Doozer

This is an interesting idea. I had never considered that the T&G has so many feed rates that you could possibly get close enough to do metric as well.

However, I'm not really convinced that, for example, a 26 tpi nut will thread onto a 25.4 tpi bolt. Sure the pitch is close, but is it close enough?

Maybe when you replace the belt with a toothed timing belt, you should use a 100/127 tooth sprocket pair. Or gears maybe? Then just flip a lever and have real metric feeds.

1" / 26 tpi = .038" pitch
1" / 25.4 tpi = .039" pitch

So one thousandth is not close enough,
as per your example ? Just asking.

-Doozer

jwearing said:

...Maybe when you replace the belt with a toothed timing belt, you should use a 100/127 tooth sprocket pair. Or gears maybe? Then just flip a lever and have real metric feeds.

Click to expand...

Yes I was thinking about that. Seems like a great idea.

-D

Doozer said:

1" / 26 tpi = .038" pitch
1" / 25.4 tpi = .039" pitch

So one thousandth is not close enough,
as per your example ? Just asking.

-Doozer

Click to expand...

That doesn't sound like much, but a normal nut has what, 5 threads? So the offset between the first and last thread will be more like .005". That seems like it would be at least a tight fit.

This might only be a problem
if the nuts thread pitch diameter
and the bolts thread pitch diameter
are the same, which they never are.
(I love reading about thread in the
Machinery Handbook, call me crazy).
There is clearance in there.
It is actually a slight benefit for
strength to have some mismatch in the
thread pitches. Look up asymmetrical
thread profile. Theory is as the first
threads that are engaged take the tensile
load, the material deflects some, and
progressively more threads then begin to
take the tensile load. If you have
conventionally cut threads, all the
threads that are in engagement take the
load at the same time, and as the bolt
stretches, it only stretches at the point
where it enters the nut. With asymmetric
thread geometry, it spreads the bolt stretch
throughout the length of the nut.
Similar concept is used to create uniform
roller chain loading by running the chain on
nylon sprockets. The gullets of the nylon
teeth deflect, and the tight and stretching
chain applied its load to more than one
sprocket tooth.

-Doozer

Bump because someone might be interested in this.
-Doozer

Hmmm... a little late to the discussion, BUT.....
would it not be easier to just make a set of change gears? I did and it really wasn't that big of a deal. while I was at it I made the ones for the DP set as well.

Peter

Yes I think I have the blue prints.
I have not gotten to it, checking what gears I need
and which one of my many indexers with their different
plates will space them. But you are right. Probably easier
to do it right.

-Doozer