Debugged a function to mill a cone

It is great

You bet. It is a huge convenience.

i
Reply to
Ignoramus6705
Loading thread data ...

I'm gonna bet it's the sprocket on the transmission output shaft . I've got a '39 Harley Flathead that uses a tapered fit sprocket (and clutch hub , and motor sprocket) , mine also incorporates an oil seal into the sprocket . I'll be boring the hub and machining the seal portion on the lathe . Might fit a modified sprocket section on my homemade hub , unless I can pick up a sprocket cutter for "real cheap" .

Reply to
Snag

There's a good chance that this bike has a belt driven final drive with bicycle style sprockets and chain to start it. My curiosity is peaked anyway.

Newb

Reply to
newb

I think so.

No oil seal here IIRC.

i
Reply to
Ignoramus6705

You could use a straight endmill, with the hub mounted on a rotary table mounted on a sine plate set to the half angle.

Reply to
James Waldby

What happens -- especially in aluminum -- is that the previously cut chips are forced against the just cut surface and weld themselves there. With the climb milling (unsafe with heavy cut -- especially on a worn manual mill and doubly so with pocket milling) the cuts remove the material and leave a nice clean surface. With the ball screws and the control from the servo motors, you should not have problems with reasonable levels of climb milling on your machine.

Enjoy, DoN.

Reply to
DoN. Nichols

A year and model would be of great help. And from a previous reply, while you might be able to generate the general angle by interpolation, you will in fact, have a series of concentric rings of narrow contact formed by the cusps of the successive passes. What is likely to happen if this is ridden much, is the high points will wear or deform, affecting the fit of the sprocket to the shaft. Any movement at all will result in fretting and wear of both parts. You are dealing with a combination of cusp height between milling passes, and the milled finish itself, which might look smooth to the eye, but is really rough at the microscopic level. All tapered sprocket bores I've ever seen are ground, and generally it is the friction of two finely matched surfaces that transmits the torque. Reduce the surface area in contact due to surface roughness, and you reduce the torque that can be handled. If however, this sprocket has a key, that would go a -long- way toward helping your approach! I don't imagine a 1920's era Indian is going to be flogged, but with the scarcity and expense of parts, it would be wise to try and insure the best possible match.

A step over of .001 would improve things quite a bit. But then there's the issue of just how round a hole can you interpolate on that thing...

I'm certainly not trying to knock what you're doing, just that I foresee potential problems and that's coming from my combined experience with machining and motorcycles.

Jon

Reply to
Jon Anderson

Jon, with a ball endmill, the approximation of a surface is relatively good even at somewhat larger steps. But I definiteky take your point.

i
Reply to
Ignoramus5687

Right, but there is another effect. With conventional milling, the cutting edge skates along the just-cut material until enough pressure is produced to bite under the surface. This sliding wears the cutting edge. I do ALL my work as climb milling now, and so very rarely do I ever have a cutter wear out, it is great! I do have a CNC machine with tight screws, and I think Igor does, too.

Jon

Reply to
Jon Elson

This meand that with normal right handed end mills, I would be better off doing counterclockwise circular cuts when finishing inside circles?

i
Reply to
Ignoramus5687

Yes, you would do the interior of all pockets counterclockwise to climb mill.

Jon

Reply to
Jon Elson

Very food idea.

I am realizing that CNC is easy in concept -- "the machine follows instructions from computer" -- but it is complicated in practice.

i
Reply to
Ignoramus18915

Ignoramus18915 fired this volley in news:RvydnYSW-9EJV8_RnZ2dnUVZ snipped-for-privacy@giganews.com:

It boils down to this, Iggy:

Even a complex, path-optimizing package depended at least originally on the skills of an accomplished machinist to establish the methods and approaches to a given task.

When you're working at a 'lower level' coding your own toolpaths, your work will only turn out as good as your machining skills would have allowed without the computer. All the CNC machine is doing is automating your personal approach to the task.

LLoyd

Reply to
Lloyd E. Sponenburgh

The key difference with CNC is the lack of interactivity. You have to plan everything out before the milling begins, unlike manual milling where you can adapt your speeds and feeds as you go and also monitor the surface finish and adapt as you get to the finish pass. The spindle speed override and feed rate override knobs on a typical CNC control are to allow the machine operator to optimize the process a bit from what was coded in the G-code. The spindle load meters are also a very useful item to have and I believe the VFDs all provide for a load meter output.

Reply to
Pete C.

I think that my VFD does not do that. (current output)

i
Reply to
Ignoramus18915

Ignoramus18915 fired this volley in news:Q9adnZ65hYwETs_RnZ2dnUVZ snipped-for-privacy@giganews.com:

A guy with some electronics background, three 0.01ohm shunts, and and handful of op-amps could accomplish that bit of feedback himself.

LLoyd

Reply to
Lloyd E. Sponenburgh

Those Toshiba VFDs certainly do, with instructions for it in the manual. Other VFDs I've looked at also have analog outputs to drive load meters.

Reply to
Pete C.

This is VERY well put. Exactly. I coded this function to cut a mold. The mold represents a cone inside a cone:

\___ ___/ \____/

And yet the code to mill it out, while it works is very suboptimal. It reflects my poor understanding of the process.

i
Reply to
Ignoramus18915

That's beyond me. What it really needs is a power meter. Anyway, I will try to stick to what I can at least figure out.

i
Reply to
Ignoramus18915

I ran out of Toshibas. This is not a Toshiba, it is a TECO FM50 203 or some such.

i
Reply to
Ignoramus18915

PolyTech Forum website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.