Thread in Two Directions At Once

[eKretz]

interesting. so if i tilt the round shank cutter in the holder im not producing threads to specs? i would have to use a 60°+ point? and all the inserts that get shimmed? are they larger than 60°?

A round rotating cutter like used in the video, no. You can rotate that about its axis of rotation and nothing changes. That's what causes the problem that Old Mart was talking about.

If you mean a plain single point threading tool with a round shank, yes that's correct - but it will take a fair amount of rotation to cause the produced angle on the threads to change a lot. We were talking about multiple lead threads with a very high lead... For a single lead thread the tool wouldn't need to be rotated much, but yes, every bit you tilt the tool will change the included angle of the cut thread.

As regards the other inserts that use shims - Check the threads produced and see! You won't see much without a comparator probably unless they are tilted pretty far.

[old mart]

The threading insert shims only go to about 3 degrees which helps mostly with the tool clearance.

[dian]

is there atually a tolerance range for the 60° angle on threads? i dont recall seeing anything like that.

[eKretz]

Yes, there certainly is. They are not particularly common in reference materials though, nor very easy to find. Here is a screenshot from an old National Bureau of Standards handbook on threading showing a column for "Deviation of Half-Angle" for various inch threads. These fall under a sort of general tolerance on pitch diameter. Some percentage of the total in the published pitch diameter tolerancing is allocated for such errors. It seems to be sort of a gray area. Note that the total at a given "perfect" pitch diameter can not exceed that angle deviation - in other words the given tolerance for the angle is only correct if the pitch diameter is theoretically perfect. If the pitch diameter deviates from theoretically perfect, the tolerance on the angle is reduced by that same error percentage. From what I understand this also applies to lead error - so the only way the angle deviation in the table will be correct at that given value would be if thread lead and pitch diameter are both dead nuts.

[old mart]

It gets more complicated when the helix angle is not only dependent on the pitch and the number of starts, but also the diameter of the thread.

[The Metal Butcher]

What is it with these Soviet lathes and threading super coarse? 48mm lead? Nearly 1/2 TPI on 14"? 16"? lathe? Must have been some big deal about multistart threads in the USSR or something.

Kinda jelly though. Seems like a nice machine.

[dian]

Yes, there certainly is. They are not particularly common in reference materials though, nor very easy to find. Here is a screenshot from an old National Bureau of Standards handbook on threading showing a column for "Deviation of Half-Angle" for various inch threads. These fall under a sort of general tolerance on pitch diameter. Some percentage of the total in the published pitch diameter tolerancing is allocated for such errors. It seems to be sort of a gray area. Note that the total at a given "perfect" pitch diameter can not exceed that angle deviation - in other words the given tolerance for the angle is only correct if the pitch diameter is theoretically perfect. If the pitch diameter deviates from theoretically perfect, the tolerance on the angle is reduced by that same error percentage. From what I understand this also applies to lead error - so the only way the angle deviation in the table will be correct at that given value would be if thread lead and pitch diameter are both dead nuts.

great find. what is confusing is "equivalent". isnt that just a translation of a diameter tolerance into an angle? it would be different if it said "half of angle tolerance".

yes, the russians: https://youtu.be/NZc-gqSIL4w?t=272

[eKretz]

The equivalent of ½ the pitch diameter tolerance as it would equate to ½-angle of the thread. Like I said, not very clear and sort of a gray area. That is one section that could certainly use a re-definition, for sure. They sort of lumped several things into one there, not very good.

[The Metal Butcher]

Nut threaded both ways:

[old mart]

Its very clever, it only works because the fit is extremely loose.

[AntonLargiader]

I don't see why a loose fit is required for this to work. They may be loose, per the previous discussion, but if they were perfectly produced the concept should still work perfectly.

[old mart]

I believe that the fit has to be loose as the nut has to be moved slightly axially to enable it to turn in one direction and in the opposite direction to turn the other way. Watch the video closely and you can see him doing just that.

[eKretz]

If everything was just perfect it might work with small clearance, but it is no easy feat to machine 10 leads in each direction on a nut and a bolt with enough precision so that they can be fitted together without a lot of clearance. I would bet quite a lot that he's way under standard pitch diameter on the bolt and/or way over on the nut. In fact you can see it in the way that the head of the bolt has quite a lot of up and down wiggle (and resultant clanking noises) while he's threading it in and out. And we already discussed the included angle mismatch - though that could still be tight on clearance at it's tightest point.

Any way you look at it, this is no more than a conversation piece. It would be next to useless in any real world application. The threads would be extremely weak and lack any kind of resistance to axial force.