cutom gearing profile

Dear Giorgis:

...

I assume each 120 degree section can be represented by sections of sec( theta ). What needs to be established is shaft-to-shaft distance, and triangle geometry.

A three-lobed triangular tooth will slip quite easily in three places, and will create very high axial separation forces both before an after this slip point.

David A. Smith

What do you mean three-lobed triangular tooth ? Do you mean the wheel would look like a three toothed pinion ?

Giorgis

Dea Giorgis:

Three lobed, perhaps. Maybe you need to provide a picture to go with this description:

I have a weird application and I think I will need to go to first principals, but I will run it past you guys as an interesting challenge.

Think 2D I have a wheel that runs on a flat. This flat has a triangle on it. As this wheel rolls over the triangle, think of the triangle as a cutter. I need the shape of the cutaway in the wheel.

I *assumed* the triangle rotated on an axis normal to the plane of the triangle. I *assumed* the each point on the rotation axis was equidistant from each vertex. I further *assumed* this rotation axis was parallel to the wheel's rotation axis.

If the triangle in rotating (?) in the rolling plane, so that the triangle's axis is perpendicular to the wheel's... What is the correlation between triangle speed (if any) and the wheel speed?

David A. Smith

With a wheel 4 units in diameter, a triangle 1 unit thick, rotating 1:1 with the wheel, the triangle equilateral, and 2 units from rotational center to tip, and rotating as in the paragraph above, the mating gear would look something like: URL://members.aol.com/dlzc/three_lobed_gear.pdf The "outies" (like navels... aka belly buttons) in the three pockets are actually deeply carved "innies".

I created it in AutoCAD, rotating the wheel 1 deg and the triangle 1 deg then subtracting the triangle (like a cutter). To simplify machining, probably best to not go for full engagement... the outer 30% of the "contact face" is pretty uninteresting.

David A. Smith

I don't get how this differs from a standard involute form.

So what's the question again, and didn't you just answer it whatever it was?

I don't get how this differs from a standard cycloidal gear form.

What do I win for all this hard work?

Fred Klingener

I guess you're right, I was after some pointers on how to go about it.

You must be new here. Usenet is a place were people often ask questions, and others come to there aid.

Dear Giorgis:

You must be new here. You get what you pay for. Advice is free, and worth everything you paid for it. Sometimes you get people in a good mood.

David A. Smith

:-) ... Good or bad mood, a sence of humor is always priceless (... free ?).

Giorgis

Just for info I have an Octagon about 240mm accross flats. I have a roller that runs on the flat of the octagon as it turns. I want the octagon to drive the roller by friction. I want to place "teeth" on the corners of the octagon to bring the roller back in synch if it goes out.

I have come up with a possible solution that I am about to make. It is somewhat of a hack, as I realise I would have issues with the pitch circle if Iuse meshing teeth of any design. (I am restucted by existing hardware)

Thanks for the pointers so far. It did trigger a need to expolore the shapes of unorthodox meshing teeth but time and other distractions do not allow me to persue it. I hope to re-visit it one day. CNC machining allows for amazing shapes to be cut.

Giorgis

If it goes out of sync., you're screwed anyway since the tooth won't match up with the groove. The diameter of your roller would have to be strategically chosen as well, so that you get an integral number of grooves spaced at the proper arcs to synchronize with the octagon's teeth. You will not be able to have a fixed center distance either. One member will have to be spring-loaded since the center distance will change as you roll. The whole thing sounds pretty Mickey-Mouse to me.

As far as tooth profile is concerned, they don't have to be of involute design. Any two conjugate profiles will roll on each other. Since you're not going to have smooth motion anyway, I wouldn't worry too much about involutes.

For more elaborate profiles, wire EDM is also an option.

Don Kansas City

Dear Don A. Gilmore:

...

Even laser-cut plastic gears might suffice to evaluate the design... Or possibly a stereolithography model, for "deep" designs.

David A. Smith

Mickey mouse in Australia means flawless :-)

Thanks for the pointers, I would love to explore the maths behind conjugate profiles. That comes with free time ...

Actually conjugate profiles are more of a geometric science and are rather simple.

If you draw a line between the centers of both gears, this is the "line of centers". For rolling contact, the pressure point where the two gear teeth contact each other must be along this line. It's that simple.

The two convex faces of the gear teeth will theoretically contact at a point (in 2D). At this point a line (plane) drawn perpendicular to both faces will be the same line. This is the "common normal" and it must intersect the line of centers for any position of the rotating gears. It doesn't even have to remain in the same place along the line of centers as long as it intersects it somewhere and the motion of the point along the line is continuous. You can literally plot points along a curve and design your own gear teeth.

Don Kansas City

Naw. That would take a pencil, some tracing paper, and a couple of push pins. Too hard.

Fred Klingener

Actually Fred, you may find it hard, but it is one of the first things I did. After playing around and figuring out what I want approximately, I wanted the elegance of a precise curve to make a CNC machiened compnent. (or a laser cut curve as is the final solution)

Thanks for your input Giorgis