Turning threads off rod fine but now rod advances out of chuck.

[RaisedByWolves]

Got 10 made got tuckered out and quit. Lotta set up time and fussing around. Here they are, I added a bevel to take the braze silver. Thx Wobbly for the advice of using a bull nose. Never tried my new m-Carbide bit. I will say that it also cut well with a triangular insert bit just coming straight down the ways at the thread. I'd recco a crush bushing like this cause it does the job and it's super fast to reload. Thanks Wolves for the relief cut idea and to Mr. York for the added relief, it's working like a champ.

Looks good!

 

[RaisedByWolves]

Yes, that's what I say; the issue is that you want the threaded workpiece NOT to shift, so any torsion caused by the cutter should
not twist the work, which is what the distal-end clamping encourages. Proximal-end clamping means the torsion
only acts on the extended work (where the cutting forces occur) and that keeps the springiness of the workpiece out
of the problem.

Work holding is most effective near the cut location. Holding elsewhere encourages deflection.

I’m of the mind to hold the entire thing if at all possible.

 

[helmbelly]

I did not read the entire thread, and this may have been suggested..

Make a collet that is drilled and tapped, then split it. The internal threads should give it more bite to hold on w/o slipping.

Make a steel collet that is drilled and tapped (not split), use a jam jut on the cut side to lock in at desired length + .125, then put a dial indicator on the ways for your carriage to give you precise Length Of Cut

Make a collet that is drilled and tapped, with a cross drilled hole for a pin or bolt at each desired depth for LOC.

Thanks Sticks, love the cross pins idea for LOC. That's great. I tried the tapped and split coupling and found that having threads engaged that way can be a problem, with a heavy cut they can engage and start walking on you- right into your toolbit, pow. The smooth bore locknut works well, Ive done in past but some fiddly set up if your doing several. What I'm using I like is smooth bore delrin (alum be fine) split once, relieved twice. it's like a silver bullet bc it holds good and set up is fast. Gonna use your pin idea.

 

[helmbelly]

I’m of the mind to hold the entire thing if at all possible.

I am too, just instinct, as long as it's within a few thou end to end, max the clamping surface.

 

[helmbelly]

with a cross drilled hole for a pin or bolt at each desired depth for LOC.

With my crush bushing it's compressing along the whole length, so a pin won't work well, altho some 3/32 delrin rod might be good and I have some. But I was thinking what would work maybe is a round wooden toothpick- ha! and if you dont stock those in your shop (and wooden tonge depressors and pipe cleaners) your missing out! When you need to just get a dab of oil somewhere or a tricky glue up or clean out a slot real quick - I use the heck out mine. Toothpicks!!

 

[helmbelly]

tried that micro carbide on the 316 stainless. there's a noticeable difference. Like night and day.

 

[whitmore]

I’m of the mind to hold the entire thing if at all possible.

Supporting the whole thing is good, but (the lore is) three or four threads is all it takes
to develop the full strength of a threaded rod; if you don't want the rod to act as a torsion
spring, moving during the cut, that suggests you want clamping pressure on the
first three or four threads to completely immobilize the work. I'm dubious that
a three-jaw chuck is the best vise-like way to hold, if you want the best
clamp on the tip end, and was imagining a 5C emergency collet machined for the purpose.

 

[helmbelly]

I asked the smartest engineer I ever knew how many threads do I need to engage for full
strength and he just said "3". So I'm with you there. But that's for full strength of a mechanical fastening against sheer/pull. How does that relate to chucking a rod I don't see. A straight rod does not become a spring when held by a clamp parallel to its surface. or I don't see how.
Where is that coming from? The more clamp in contact with the rod the better the grab which is what chucking is all about -holding the work firmer than the tool trying to stop it.

 

[RaisedByWolves]

I asked the smartest engineer I ever knew how many threads do I need to engage for full
strength and he just said "3". So I'm with you there. But that's for full strength of a mechanical fastening against sheer/pull. How does that relate to chucking a rod I don't see. A straight rod does not become a spring when held by a clamp parallel to its surface. or I don't see how.
Where is that coming from? The more clamp in contact with the rod the better the grab which is what chucking is all about -holding the work firmer than the tool trying to stop it.

And is not true.

Rule of thumb is 1.5X the diameter of the bolt for steel and 2X for aluminum.

If you're only chucking three threads on soft crap material like off the shelf threaded rod you're likely to deform it.

3-4 threads may be doable in an extreme situation, say turning the head of a short cap screw down using a collet, but other than a situation like this......Why?

No need to play fancy games with things said around the lunch table, hold everything the best you can.

 

[whitmore]

I asked the smartest engineer I ever knew how many threads do I need to engage for full
strength and he just said "3". So I'm with you there. But that's for full strength of a mechanical fastening against sheer/pull. How does that relate to chucking a rod I don't see. A straight rod does not become a spring when held by a clamp parallel to its surface. or I don't see how.
Where is that coming from? The more clamp in contact with the rod the better the grab which is what chucking is all about -holding the work firmer than the tool trying to stop it.

The best high-quality springs are rods (quartz fibers) in torsion. To make them deflect more, you just use a longer
length. Angular deflection threatens cut quality, and keeping the workpiece nearly rigid means holding
it near the cut. The screw-thread issue is about stretching (also elastic deformation, but different geometry).

Simple (good-but-not-perfect) physics model of friction is that the static (holding force) friction is a coefficient
times the force normal to the surfaces in contact. Area doesn't come into that model; as long as the forces
don't significantly deform the surfaces, it's theoretically NOT better holding/more friction to use larger area, only
to use higher normal (clamping) force. This implies one should use a clamp caul with enough contact area to
avoid plastic deformation, but NOT to otherwise maximize area.

Much of the 'normal force' in threaded objects is from the geometric ~30 degree slope of the contact surface; that
roughly doubles the normal force and friction, compared to cylinder-in-cylinder clamping.