Shop Made Tools

[wbc]

I too am easily corn fused picturing rotary broaching at work. Wikipedia may help:

https://en.wikipedia.org/wiki/Broaching_(metalworking)

The necessary condition is that the center of rotation of what I am calling the "shank" and the center of rotation of the "wobbling" broach intersect
at the face of the broach. This varies with angle and length of broach. Many designs allow the "shank" to be adjusted relative to the "barrel"
to accommodate differing length broaches. I settled on fixed 1.5" long broaches and then "dial gauge found" the proper shank
offset to get the zero intersection. No math or geometry dizziness. Then screwed the two pieces together and welded for good measure.
Then turned down the frankenstein assemblage into a marginally less crappy looking rounder assembly. Beauty, it ain't.

As noted before, a mill, lathe, or drill press, etc. may be employed. Rotate the tool, or rotate the toolee, no matter.

[Dave C]

Ought to be a youtube video showing a rotary broach in action.

[Bob Engelhardt]

Ought to be a youtube video showing a rotary broach in action.

This Old Tony made a rotary broach and video of it working. Two videos actually - the 1st was on the lathe and with the work and broach turning it was not at all clear as to what was happening. So the 2nd one had the broach's shank in the drill press, with the work & cutter stationary. Very clear.

[BCRider]

I believe it was AVE that also did a dead simple version too.

It occurs to me in a sudden flash as I read your last reply that maybe it doesn't need to be all that complicated. If we can live with a slight chamfer at the opening of the broached hole the job could be done with this "loose broach" as seen in the sketch below.

There is an obvious issue with the sketch since clearly the broach cannot be removed from the retraction cap. I saw that right away but wasn't about to alter the whole "quick" sketch. I think what I'd do is make the ball end "knuckle" a separate and larger piece. And due to the rotational factors the broaches themselves could simply thread into the half ball "knuckle". I would simply assemble the broach and knuckle with the retraction cap fitted to the pair then install the whole works in the offset.

Did I miss anything on this conceptual sketch?

What I like about this option is that the broach is centered by the chamfer of the part. So no need for any actual set length. Well, other than the length must be sufficient that the side clearance angles do not go past being parallel with the center axis. So depending on the amount of offset in the bearing body the broaches would have some minimum length. But longer would be OK. And in fact.... .longer would need less of a taper angle with this scheme. Yes? No? With this way of doing it the smaller "waist" at the end which will screw into the half ball can be a set size. And the broach length can be set to any value. Well... within reason. If the broach is too long there won't be enough wobble and actually forming the hole would take a lot longer or may not occur after some value due to spring in the parts. So there is a limit to all this.

Anyway if this option works it would take away the need for an oddball angle being drilled.

[MikeWI]

I like this one as well:
https://www.hobby-machinist.com/thre...9/#post-272748

[BCRider]

That's a nice option Mike. I just need three screws to come in from the side instead of that cap like threaded on retractor. That makes the job even easier.

[Bob Engelhardt]

I like this one as well:
https://www.hobby-machinist.com/thre...9/#post-272748

That is nice - it is so simple, cuts to the chase, if you will. As a dedicated KISS-er, I am impressed.

[BCRider]

In that simple design with just the single ball as the pivot and rotating bearing he mentioned not using too much relief angle to avoid the broached hole forming a spiral. I wonder if that is related to the drag in the socket due to the spinning? I'm thinking that there may still be some advantage to using a tapered roller as a lower rotational drag socket. I do like the idea of a hard ball and the mild steel with socket formed by a ball end mill though.

[Bruno Mueller]

Some time ago I ordered a set of Hemingway Kits for a Rotary Broach.
After I built this device and was successful in using it, I realized that it would be more sensible to build this device a little bigger. I increased the size of the tool by a factor of 1.5.
The result I would not like to withhold from you.

First the original tool from Hemingway Kits.

5mm square hole in a key for square screws from my Mulifix Aa steel holder.
The wrench is made of unhardened C45 tool steel.

[Bruno Mueller]

and here comes the enlarged version.

[Bruno Mueller]

I have pre-milled and pre-ground the punches from tool steel. Afterwards they were hardened and tempered.
Finally the punches were ground to size.

The smaller punches were partly made from broken off screw taps or center drills. They were glued into a steel holder with Loctite and then machined.


grinding the hexagon Punches, ...........................................Square hole 2,5mm and 2,0mm


10mm hexagonal hole in aluminium

[Bruno Mueller]

And now a rough drawing for the enlarged version.
I cannot guarantee the accuracy of the drawing.

[TGTool]

Very nicely done. I haven't seen drawings for the Hemmingway design, but where is the tilt or offset produced to get the rotary broach action? It's a small amount and not apparent from the photos. In theory it could be on the face of the shank or the back face of the body where the slot is for compensating for cutter length. I suppose it could also be a slanted bore in the body, but that gets more complicated for setup.

[Bruno Mueller]

Very nicely done. I haven't seen drawings for the Hemmingway design, but where is the tilt or offset produced to get the rotary broach action? It's a small amount and not apparent from the photos. In theory it could be on the face of the shank or the back face of the body where the slot is for compensating for cutter length. I suppose it could also be a slanted bore in the body, but that gets more complicated for setup.

The 1° angle is attached to the end of the body and after assembly, the parts were turned over together with a fictitious clearing pin.
The bevel is no longer visible at the end. Only when a broaching punch is clamped in place can this be seen.

This is a detail of an original Hemingway drawing.

[TGTool]

Thanks, Bruno. It looks like a very robust design with the support and thrust bearings and you've done a nice job expanding on it.