PDA

View Full Version : Gear and rack discovery



dp
04-16-2009, 04:44 PM
I made a most interesting discovery today while tidying up the machines. Last year I bought a longer rack for my little Grizzly mill which gives more head travel. The old rack has been used as a spacer and any number of odd jobs since. Today I happened to pick it up and in my other hand I had a change gear from the Grizzly lathe. The mill and lathe are part of a 3-in-1 combo (G0516). The gear and rack are a perfect fit. Amazed, I grabbed another gear of much larger diameter. It's not a fluke - it's a perfect fit.

I'm now working on a gear cutter attachment for my Atlas shaper that will use the rack to allow me to duplicate my change gears as they wear out. These gears have very tiny teeth and are made of very brittle iron so there's been a lot of attrition of the smaller gears. I'll use my flat bed scanner as a poor man's comparator to get the cutter profile correct. Should be an interesting project. I may even remember to take pictures!

Optics Curmudgeon
04-16-2009, 06:24 PM
No, not a fluke at all, you've proven that the rack and the change gears have the same pitch and pressure angle. However, if you grind a tool for the shaper to match the rack you'll have a cutter to make more racks (or at least more gears with at least 135 teeth). The shape of a gear tooth is a complex curve, and the curve is more pronounced in gears with less teeth. Think of the rack as a gear with infinite radius. Its teeth have straight sides, "leaning" in at the pressure angle. In a spur gear with a finite number of teeth the sides of the teeth will have an involute curve, different for each number of teeth. Involute milling cutters are made to approximate the curve for a range of teeth, and to duplicate that cutting action with a shaper cutter you would have to grind the tool appropriately. There are ways to cut proper involute gears on a shaper with a rack form tool, by coordinating the rotation of the gear blank with the cross feed, the tool then behaves like one tooth of a rack. There was an article in ME years ago (Sept. 1950) by "Base Circle" in which the process was described. C. Bamford expanded on it in Jan. 1989. These are worth looking up.

Joe

John Stevenson
04-16-2009, 06:30 PM
Correct Joe but reading between the lines I think DP is working along these lines.

If not and I apologise here are the two articles.

http://www.stevenson-engineers.co.uk/files/first%20article.pdf

and the second article.

http://www.stevenson-engineers.co.uk/files/second%20article.pdf

.

Evan
04-16-2009, 07:16 PM
Without reading what John posted, a rack cutter can be used on a shaper to generate mathematically correct involute gears.

John Stevenson
04-16-2009, 07:37 PM
Without reading what John posted, a rack cutter can be used on a shaper to generate mathematically correct involute gears.

That's what I posted :D

Think about this, I'd love to do it and it would work but I just don't have the time.

Image an XY table under CNC control with a rotary table mounted flat on the XY table.
That gives you X, Y and A.

Now take one of those cheap import arbor presses and power it with a crank and con rod so you get a nodding donkey.

The ram holds a 50c piece of tool steel ground up as one tooth of a rack, can be done on any bench grinder it's that easy.

Now you centre the cutter on the edge of the blank mounted on the RT, move away and start the machine. It feeds in to depth in the Y and the X moves together with the A every time the cutter is out of the work.
Simple microswitch on the ram. As the bed moves in the X it generates an true involute as the RT rotates.

A lot like the shaper cut gears but simpler in that the code is one line followed by a sub routing to do 360/ n teeth
That simple 50c cutter can then do any gear in the DP range it covers instead of the 8 cutters per set.

Specials, corrected teeth ? no problem just alter one line of code.

.

dp
04-16-2009, 08:37 PM
I won't use the rack for the cutter itself (but could), but to pull an existing gear which turns a shaft which turns a blank that the shaper cutter cuts.

John's link shows the use of a wire drawn tangent to the work to rotate the work and that can be replaced with a rack. I could use a wire or metal tape to do the same, but like the idea of the rack as it is positive, bi-directional, and cool.

Here's a youtube video showing the used of a metal tape anchored to and wound around a blank. It turns the shaft the work is on as the table moves. Without a lot of friction to keep the shaft from freewheeling over part of the arc, I think this design is not robust. Substitute the tape with a rack and gear and you have my concept that provides a solid mechanical connection over the entire changing angle of cut:

http://www.youtube.com/watch?v=adRlGffXM5k

The cutter itself is a simple point like a single-point threading cutter patterned exactly after a single tooth from the rack. The angle of the cutter is the PA of the gear. Very easy to make.

J Tiers
04-16-2009, 10:32 PM
To cut gears on the shaper, it is much simpler to use a matching gear (tooth count wise) and a shop-made indexer based on using a matching gear.

If you do not HAVE a matching gear, but you DO have part of the original gear including the hub, you can use that, and carefully re-index it to continue when you run out, assuming you have a credible amount of gear circumference available on the fragment.

Grinding the form cutter is fairly easy, you just keep holding it up to the original, and grinding where it touches, until it touches everywhere. Small pinions... pretty much no can do, although you theoretically could do it with two cutters, with care........ NOT.

You will find it is a real nuisance to cut gears, unless your setup is solid, and that just makes it a bit scary. Cutting the full form towards the end of the tooth, you want to advance about a half thou, or you chatter and/or threaten to break something.

With a large shaper, a small cutter, and a setup like the rock of Gibraltar, you'd figure the worst that can happen is breaking the cutter. With a tiny shaper like the Atlas, it's a bit scarier..... clunks like anything when it hits the gear.....

dp
04-16-2009, 11:07 PM
You will find it is a real nuisance to cut gears, unless your setup is solid, and that just makes it a bit scary. Cutting the full form towards the end of the tooth, you want to advance about a half thou, or you chatter and/or threaten to break something.

With a large shaper, a small cutter, and a setup like the rock of Gibraltar, you'd figure the worst that can happen is breaking the cutter. With a tiny shaper like the Atlas, it's a bit scarier..... clunks like anything when it hits the gear.....

If you allow the shaper to rotate the index gear and the blank (using my spare section of rack) there is no problem with the big clunk regardless of size as the work is approached gradually and the depth of each cut is small. This rather than a plunge cut with a fully formed cutter and static work. And that brings up some additional benefits. You can cut gears with a small pinion. And you can gang the blanks to increase productivity and rigidity. Finally, grinding the cutter becomes a simple matter of setting the angle on your grinder guide to produce the correct PA and voila, you're done. Need to sharpen it? Not a problem as it's not a complex form - just straight sides.

On the subject of ganging, and this works for a number of shaper projects, you can gang the cutters in your tool holder so they work like a broach. I have enough room in my tool holder for three 1/2" cutters. For this project, because of the rotating shaft, that won't be needed. (Edit) If I were going to cut a new rack then ganged cutters would help to minimize the problems of interrupted cuts typical of shapers. I might even consider using a section of broach ground to shape on a spline cutting bar. People misunderestimate the versatility of the shaper.

John Stevenson
04-17-2009, 03:40 AM
Here's a youtube video showing the used of a metal tape anchored to and wound around a blank. It turns the shaft the work is on as the table moves. Without a lot of friction to keep the shaft from freewheeling over part of the arc, I think this design is not robust. Substitute the tape with a rack and gear and you have my concept that provides a solid mechanical connection over the entire changing angle of cut:

http://www.youtube.com/watch?v=adRlGffXM5k

The cutter itself is a simple point like a single-point threading cutter patterned exactly after a single tooth from the rack. The angle of the cutter is the PA of the gear. Very easy to make.

The Maag gear tooth grinders use exactly this same method with the tape.
The gear has been roughed out by other means then a wheel dressed in a vee to the PA of the gear is passed thru as it rotates in steps, then backs off, indexes round and repeats, very accurate.

.

Timleech
04-17-2009, 04:10 AM
That's what I posted :D

Think about this, I'd love to do it and it would work but I just don't have the time.

Image an XY table under CNC control with a rotary table mounted flat on the XY table.
That gives you X, Y and A.

Now take one of those cheap import arbor presses and power it with a crank and con rod so you get a nodding donkey.

The ram holds a 50c piece of tool steel ground up as one tooth of a rack, can be done on any bench grinder it's that easy.

Now you centre the cutter on the edge of the blank mounted on the RT, move away and start the machine. It feeds in to depth in the Y and the X moves together with the A every time the cutter is out of the work.
Simple microswitch on the ram. As the bed moves in the X it generates an true involute as the RT rotates.

A lot like the shaper cut gears but simpler in that the code is one line followed by a sub routing to do 360/ n teeth
That simple 50c cutter can then do any gear in the DP range it covers instead of the 8 cutters per set.

Specials, corrected teeth ? no problem just alter one line of code.

.

Can't you just adapt a slotting head to bolt onto your Beaver? You must have a stepper-driven RT around the place?

Tim

John Stevenson
04-17-2009, 04:18 AM
Not easily, the Beaver is a fixed ram, no ram swivel like a TW , Elliott or Bridgy, just bolted on full square and pinned.
I would need to remove the whole of the head, wiring etc to get a mounting face.

It would be easier to CNC the Ravensburgh slotter, in fact the Russians used to make an attachment that fitted their slotters to just this but it was mechanical.

I don't have any pictures but there is a sketch in one of the Russian design books I have, damn cunning these Ruskies and very much underrated for machine tools.

.

Timleech
04-17-2009, 04:47 AM
Not easily, the Beaver is a fixed ram, no ram swivel like a TW , Elliott or Bridgy, just bolted on full square and pinned.
I would need to remove the whole of the head, wiring etc to get a mounting face.

It would be easier to CNC the Ravensburgh slotter, in fact the Russians used to make an attachment that fitted their slotters to just this but it was mechanical.

I don't have any pictures but there is a sketch in one of the Russian design books I have, damn cunning these Ruskies and very much underrated for machine tools.

.

John
I was thinking more in terms of some sort of mounting on the front face of the column, below the head. Depends how much knee travel there is, I suppose.

Tim

Edited to add that I'll weigh up how 'do-able' something like that would be on my S1 Bridgeport when my 'new' TW slotting head arrives. Might give me an incentive to get the controller finished :rolleyes:

jackary
04-17-2009, 06:17 AM
Thank you All for this interesting discussion/information. I gives a good insight to a tricky challenge and food for thought. I've spent the whole morning reading this and have not yet started hoovering the house, I will be in trouble now when big Mama gets home.
Regards
Alan

J Tiers
04-17-2009, 08:55 AM
If you allow the shaper to rotate the index gear and the blank (using my spare section of rack) there is no problem with the big clunk regardless of size as the work is approached gradually and the depth of each cut is small. This rather than a plunge cut with a fully formed cutter and static work.

If you use a form cutter, at some point it cuts on its full outline, and you have maximum force.

If you use a gear shaper, with a rack cutter, you don't. But you need a relatively huge amount of extra "stuff" to create that on your general purpose shaper. To cut a few gears from time to time, assuming you don't have a mill, it very likely isn't worth it.

If you have a mill, the problem solves itself, you buy the cutter you need, for less than a gear will cost you, and can cut that range of gears forever after.

John Stevenson
04-17-2009, 09:10 AM
If you have a mill, the problem solves itself, you buy the cutter you need, for less than a gear will cost you, and can cut that range of gears forever after.

Yabut only that limited range.

.

dp
04-17-2009, 11:44 AM
If you use a form cutter, at some point it cuts on its full outline, and you have maximum force.

If you use a gear shaper, with a rack cutter, you don't. But you need a relatively huge amount of extra "stuff" to create that on your general purpose shaper. To cut a few gears from time to time, assuming you don't have a mill, it very likely isn't worth it.

If you have a mill, the problem solves itself, you buy the cutter you need, for less than a gear will cost you, and can cut that range of gears forever after.

If having a gear were my goal I'd just pick up the phone and call Grizzly and order them as I need them. But my goal is to develop and use techniques for machining things for the pure pleasure of knowing I did.

chrsbrbnk
04-17-2009, 05:51 PM
I kinda like the cnc shaper concept wonder how big an arbor press you'd need might have to tune up the ram play a little

J Tiers
04-17-2009, 08:20 PM
If having a gear were my goal I'd just pick up the phone and call Grizzly and order them as I need them. But my goal is to develop and use techniques for machining things for the pure pleasure of knowing I did.

That is absolutely no problem.

I am the reverse...... when I need a gear, and don't have the cutter, I buy the cutter, and then I can make as many gears as I want..... I also buy gear cutters cheap when the opportunity arises...

On the other hand, I HAVE set up on both the shaper and the mill to cut bevel gears...... so I DO understand the idea....