PDA

View Full Version : wait a sec....would this work?



doctordoctor
10-09-2014, 04:44 PM
okay so I want to make change gears for my Atlas lathe because its missing all of them.

Many people have suggested the usual ways of making custom cutters or buying hobs etc..much has been done on youtube, etc..

The teeth are your typical involute spur gear shape.

What I would like to do in a perfect world is design the cutter geometry in solidworks, then machine the cutter on my CNC mill, so there is no hand grinding or approximating of the proper shape. It would also be awesome to make arbitrary cutters this way for all kinds of things.

Problem is, machining HSS is not exactly easy. Machining tool steel and then hardening it is also a possibility, but still subject to distortion and would need some hand treatment afterwards. And a furnace I dont have. (although I would like to get into this..for discussions sake lets say its not an option)

But I just realized..if I am going to make the gears out of delrin or glass reinforced nylon, then do I really need an HSS cutting tool? Cant I just use an unhardened 1018 cutting tool? Or possibly a hard steel, but not one that is difficult to machine (relatively speaking)?

Because if thats the case, I could pump out several involute cutters, and actually make an insert based gear hob that uses those cutters.

And end up with a powerful system that lets me easily make multi tooth hobs and have precision control in CAD over all the geometry.

So..the question is, can you cut plastic effectively with steel that has been machined but not ground, and is still relatively friendly to machine on a CNC mill?

And then..can you do the same with aluminum? What happens if you try cutting aluminum with something softer than HSS?

If this all makes sense..what would be the best steel for the job? Hard as possible without being difficult to machine I would imagine. 41L40?

BTW anyone that helps me figure out a way to do this that I end up doing will almost certainly be rewarded with a gear at some point :)

boslab
10-09-2014, 05:53 PM
Why not, you could case harden them after if you had the urge.
Mark

J Tiers
10-09-2014, 06:39 PM
There's "hobs" and there's "cutters" for gears.

A "hob" is driven at a particular RPM as the gear is turned in synch, and will cut ANY gear of that pitch. It looks like a worm, except it is gashed to form teeth, and the cross-section is same as a rack tooth. It corrects for all variables geometrically to form any tooth count of it's pitch that may be desired.

A gear "cutter" is formed to cut one range of gears of one pitch reasonably well. It has a shape exactly the shape of the space between gear teeth for the "best fit" tooth count. It might cut as few as two tooth counts acceptably.

Gears may be made of anything that will work, from delrin or micarta to 4140PH or a case hardening alloy.

Aluminum is surprisingly abrasive. Not a lot softer than 1018, and in some alloys harder. Best cut with HSS

doctordoctor
10-09-2014, 06:50 PM
There's "hobs" and there's "cutters" for gears.

A "hob" is driven at a particular RPM as the gear is turned in synch, and will cut ANY gear of that pitch. It looks like a worm, except it is gashed to form teeth, and the cross-section is same as a rack tooth. It corrects for all variables geometrically to form any tooth count of it's pitch that may be desired.

A gear "cutter" is formed to cut one range of gears of one pitch reasonably well. It has a shape exactly the shape of the space between gear teeth for the "best fit" tooth count. It might cut as few as two tooth counts acceptably.

Gears may be made of anything that will work, from delrin or micarta to 4140PH or a case hardening alloy.

Aluminum is surprisingly abrasive. Not a lot softer than 1018, and in some alloys harder. Best cut with HSS

Hmm...I think making the hob would present the same problem as making the cutter for me..so I suppose I am talking about making a cutter then. And if aluminum cant be cut well with non-hss, then I guess i'll have to stick to a strong plastic for the gears.

lane
10-09-2014, 06:58 PM
Just grind up a tool bit on your pedestal grinder, use are make up a boring bar and fly cut the gear . a done deal. That is the way we have done it for years in a working machine shop for rush jobs when we did not have a cutter . A piece of 1/4 tool bit will work fine.

doctordoctor
10-09-2014, 07:08 PM
Just grind up a tool bit on your pedestal grinder, use are make up a boring bar and fly cut the gear . a done deal. That is the way we have done it for years in a working machine shop for rush jobs when we did not have a cutter . A piece of 1/4 tool bit will work fine.

The problem is hand grinding a precision involute gear shape. And I would need about 3 or 4 different shapes for the various size gears I need to make.

Plus it doesnt let me control the process and make custom gear cutter arbors and use the inserts I make in them. I could do all kinds of stuff if I can get the whole process in CAD and machined CNC.

loose nut
10-09-2014, 07:24 PM
Your over thinking it.

Change gears on lathes are not the high precision type and don't need perfectly shaped cutters. Lots of people have cut gears using simple methods to form the gear cutter or hob with success, my self included. Get a copy of Ivan Laws book on gear cutting, it will show you what you need to know.

alanganes
10-09-2014, 08:24 PM
OK I don't really know the answer to your question, but I THINK the answer is that you ought to try it! Not much to lose, I think.

It seems that there would be no reason that you could not cut many plastics with a steel tool that was not hardened or HSS. It may not last as long as something harder, but if it stays sharp long enough to make the part you need, then you could count it as a success. I have no idea how long such a thing would last on aluminum but again it sounds like it may make for a worthy experiment.

Please let us know what you find out!

Toolguy
10-09-2014, 08:31 PM
Make the cutters out of tool steel. It is annealed to start with. I would get W-1 or 0-1. Heat it to glowing orange with a propane torch and quench it. It will be hard enough to cut mild steel and softer. You can cut it on the CNC mill. With a tool bit the size to fit between gear teeth the propane torch will be hot enough. You will not get enough warping to ever know if it does. Keep the spindle RPM down when cutting the tool steel to make the end mills last. The 1018 will not work well at all.

Paul Alciatore
10-09-2014, 08:36 PM
If you are going to hob the gears, you really don't have to make the cutter. There are any number of them available on E-bay and other sources, new and used, for reasonable prices. For change gears, you do not need the very best quality hob or even extreme long life, so an inexpensive hob should do the job.

http://www.ebay.com/sch/i.html?_sop=3&_mPrRngCbx=1&_udlo=10&_udhi=49&_nkw=gear%20hob%20cutter&rt=nc&_pppn=r1

Likewise if you are going to use indexing and cut one tooth at a time. There are any number of inexpensive cutters that will be as good as or better than any that you could make.

I would worry more about the rest of the set up and just buy the cutters.

As for the material, I would use aluminum, perhaps some of the 6061-T6 stuff. Easy to machine and good for home use gears.

J Tiers
10-09-2014, 09:41 PM
The problem is hand grinding a precision involute gear shape. And I would need about 3 or 4 different shapes for the various size gears I need to make.

Plus it doesnt let me control the process and make custom gear cutter arbors and use the inserts I make in them. I could do all kinds of stuff if I can get the whole process in CAD and machined CNC.

Hand grinding the cutter isn't so bad. if you have a gear that would be cut with the same "number" of standard cutter, you can use it as a template to fit against. Just grind freehand and keep at it, comparing to the gear until it fits well. It will be good enough to work.

I've done that and cut gears with a shaper. Bevel gears, in fact.

Fasttrack
10-09-2014, 10:16 PM
Ditto what everyone has said, but if you are dead set on cutting them out with the CNC, then you can make them from something like 4140. That could be heat treated later or left as is and hold up reasonably well to delrin. Glass reinforced nylon is more abrasive.

Really, you are over thinking. Your CNC machine won't make them "perfect" or "exact". Then, if you make the cutters from ordinary steel, they will wear pretty quickly as you cut, so the tooth form will change as you go. If your gears are plastic, they will wear and distort in use. As others have said, it doesn't need to be a honed and lapped gear train to be effective in this application.

bubby-joe
10-09-2014, 11:16 PM
What model and size, I have a very disapointing atlas that I bought and first sight was I paid what for this POS about the ony good thing was a full set of change gears I'm keeping the 4 jaw for a tailstock offset and the rest can go piece meal.

rcaffin
10-10-2014, 12:38 AM
Well, for a start, you can probably use plastic for the gears - GOOD plastic that is. I would suggest nylon or acetal (or delrin) for the gears, although other plastics are also possible. Aluminium will wear very fast in that application. A plastic like PVC will likely shatter. PC could actually melt!

Tooth shape: it sounds bad, but with nylon or acetal gears it is probably a case of 'close enough', using either bought cutters or MYOG using another gear as a profile guide. Just don't force the meshing.

Gear width: err on the wide side here. Plastic will work fine, but give it as much strength as you can. Hey - you should see the tiny Gates Powerband belt driving my lathe: it amazes me what it can do.

Now, MYOG cutters. Use HSS if you can, not 'mild steel' and not carbide. The reason is that you can get a much better edge on a bit of HSS than on other stuff, and plastic strongly prefer a sharp edge. Cuting plastic is very different from cutting steel. If you want to use carbide you should hone the edge.

Cheers
Roger

Baz
10-10-2014, 08:43 AM
The cutter can be made from any high carbon steel. Drill rod in the USA, Silver steel in UK both of which are most commony round rod. For larger disc cutters you would need flat sheet 'gauge plate' in the UK, not sure what it is called in USA. These can be softened for cutting by heating to red and cooling slowly. After cutting they are hardened and tempered in the usual way. (look it up) When softened they can be cut by a hardened version of themselves. This tool will cut aluminium for hours. HSS is only needed for things that are run so hard and fast (ie in production) that they will get red hot. Make the small gears in aluminium and the large ones in plastic and they will run together nicely.
Have you actually looked up the button method and cone drill method of making cutters as recommended to you in previous threads? It is easy peasy. You do not need CNC to make gear cutters it's like saying you need a Cray computer to add up your grocery bill and it takes ten times longer. If you cannot manage this forget screw cutting for a few months until you have mastered basic metal work and lathe use.

Lew Hartswick
10-10-2014, 09:53 AM
As to working "plastics" with non-ferrous tooling:
A friend an I made quite a few high voltage connector parts out of Nylon with a brass tool .
This was for the 10 Kv connector for a corona regulator and the configuration was an
annular slot about 1/8" wide, 1" dia. and maybe 3/4" deep . Made about a dozen for the
company I worked for at the time and they were used in a military system. That brass
tool just pealed the nylon off like you wouldn't believe.
...lew...

doctordoctor
10-10-2014, 10:52 AM
Your over thinking it.

Change gears on lathes are not the high precision type and don't need perfectly shaped cutters. Lots of people have cut gears using simple methods to form the gear cutter or hob with success, my self included. Get a copy of Ivan Laws book on gear cutting, it will show you what you need to know.


This is 1/3rd making a cutter to make gears from my Atlas, 1/3rd fun/learning, 1/3rd creating a highly controlled, parametric process that will let me make arbitrary cutting tools for my mill for use on plastic. I try to avoid manual processes like the plague in all my machining, all they do is eat up labor and lower the consistency of whatever it is thats getting done.

doctordoctor
10-10-2014, 10:52 AM
If you are going to hob the gears, you really don't have to make the cutter. There are any number of them available on E-bay and other sources, new and used, for reasonable prices. For change gears, you do not need the very best quality hob or even extreme long life, so an inexpensive hob should do the job.

http://www.ebay.com/sch/i.html?_sop=3&_mPrRngCbx=1&_udlo=10&_udhi=49&_nkw=gear%20hob%20cutter&rt=nc&_pppn=r1

Likewise if you are going to use indexing and cut one tooth at a time. There are any number of inexpensive cutters that will be as good as or better than any that you could make.

I would worry more about the rest of the set up and just buy the cutters.

As for the material, I would use aluminum, perhaps some of the 6061-T6 stuff. Easy to machine and good for home use gears.

Just buying the cutters doesnt do me any good as far as learning about making cutters or creating a process I can use later to make arbitrary cutters.

doctordoctor
10-10-2014, 10:56 AM
Hand grinding the cutter isn't so bad. if you have a gear that would be cut with the same "number" of standard cutter, you can use it as a template to fit against. Just grind freehand and keep at it, comparing to the gear until it fits well. It will be good enough to work.

I've done that and cut gears with a shaper. Bevel gears, in fact.

Id actually like to try this as a project of its own, but more just for fun and to learn. Otherwise, I try to avoid manual processes as much as possible. I just need a good excuse to hand grind a bit. Im sure one will come up eventually with the Atlas.

doctordoctor
10-10-2014, 10:59 AM
Well, for a start, you can probably use plastic for the gears - GOOD plastic that is. I would suggest nylon or acetal (or delrin) for the gears, although other plastics are also possible. Aluminium will wear very fast in that application. A plastic like PVC will likely shatter. PC could actually melt!

Tooth shape: it sounds bad, but with nylon or acetal gears it is probably a case of 'close enough', using either bought cutters or MYOG using another gear as a profile guide. Just don't force the meshing.

Gear width: err on the wide side here. Plastic will work fine, but give it as much strength as you can. Hey - you should see the tiny Gates Powerband belt driving my lathe: it amazes me what it can do.

Now, MYOG cutters. Use HSS if you can, not 'mild steel' and not carbide. The reason is that you can get a much better edge on a bit of HSS than on other stuff, and plastic strongly prefer a sharp edge. Cuting plastic is very different from cutting steel. If you want to use carbide you should hone the edge.

Cheers
Roger

Id use HSS if I could cut it on the mill. Its probably possible, but if we are talking plastic gears, then Im hoping to get away with a softer steel that I can easily machine and possible harden with the propane stuff I have.

doctordoctor
10-10-2014, 11:05 AM
OK I don't really know the answer to your question, but I THINK the answer is that you ought to try it! Not much to lose, I think.

It seems that there would be no reason that you could not cut many plastics with a steel tool that was not hardened or HSS. It may not last as long as something harder, but if it stays sharp long enough to make the part you need, then you could count it as a success. I have no idea how long such a thing would last on aluminum but again it sounds like it may make for a worthy experiment.

Please let us know what you find out!


Thanks!! Half the fun of this will be seeing what happens..I will definitely be posting all info and pics about it

doctordoctor
10-10-2014, 11:06 AM
Make the cutters out of tool steel. It is annealed to start with. I would get W-1 or 0-1. Heat it to glowing orange with a propane torch and quench it. It will be hard enough to cut mild steel and softer. You can cut it on the CNC mill. With a tool bit the size to fit between gear teeth the propane torch will be hot enough. You will not get enough warping to ever know if it does. Keep the spindle RPM down when cutting the tool steel to make the end mills last. The 1018 will not work well at all.

EXCELLENT information Toolguy!! Being able to harden tool steel with my propane stuff is a CRITICAL part of this idea. Now that I know that things are going to get real.

doctordoctor
10-10-2014, 11:09 AM
Ditto what everyone has said, but if you are dead set on cutting them out with the CNC, then you can make them from something like 4140. That could be heat treated later or left as is and hold up reasonably well to delrin. Glass reinforced nylon is more abrasive.

Really, you are over thinking. Your CNC machine won't make them "perfect" or "exact". Then, if you make the cutters from ordinary steel, they will wear pretty quickly as you cut, so the tooth form will change as you go. If your gears are plastic, they will wear and distort in use. As others have said, it doesn't need to be a honed and lapped gear train to be effective in this application.

Its not so much I am trying to make precision gears for the Atlas, which is probably not warranted. Its that I want to have a controlled , parametric process from CAD to machined cutter that lets me make arbitrary cutter shapes and arbors to hold multiple custom cutter inserts. Ill be working on lots of other gear types and some of them will need many cutters and they will all need to be the same. Plus I'll need to be able to tweak specific things about the cutters and arbors and do it easily. I try to get away from manual processes as much as possible and do everything in CAD and CNC, for me, its much faster and more consistent that way, and I learn more when I have control of the entire process and all the geometry.

doctordoctor
10-10-2014, 11:11 AM
The cutter can be made from any high carbon steel. Drill rod in the USA, Silver steel in UK both of which are most commony round rod. For larger disc cutters you would need flat sheet 'gauge plate' in the UK, not sure what it is called in USA. These can be softened for cutting by heating to red and cooling slowly. After cutting they are hardened and tempered in the usual way. (look it up) When softened they can be cut by a hardened version of themselves. This tool will cut aluminium for hours. HSS is only needed for things that are run so hard and fast (ie in production) that they will get red hot. Make the small gears in aluminium and the large ones in plastic and they will run together nicely.
Have you actually looked up the button method and cone drill method of making cutters as recommended to you in previous threads? It is easy peasy. You do not need CNC to make gear cutters it's like saying you need a Cray computer to add up your grocery bill and it takes ten times longer. If you cannot manage this forget screw cutting for a few months until you have mastered basic metal work and lathe use.

Excellent info! For me, CNC and CAD are actually the easiest way for me to get anything done. All the geometry is controlled and defined and the process is very consistent. Plus if I need to make 10 cutters because Ive come up with some arbor that uses them as inserts, I wont have to hand grind all of them. And I can tweak the cutter and arbor geometries very specifically if needed. And also whatever I draw up in SW will be parametric and therefore can be extrapolated to different sizes with ease.
.

doctordoctor
10-10-2014, 11:21 AM
So it looks like this is feasible.

To test out the process, I'll do a single cutter and use it "fly cutter" style.

Ill make it out of unhardened 1018, and try it on delrin.

Then I'll harden it using the methods described by Baz and Toolguy, and try it again on Delrin, and also on Aluminum.

There are alot of ways to draw involute spur gears in SW.

I could draw the gear I want, and then cut out the tooth shape and use that for the cutter.

But the cutter needs other geometries like relief angles and what not to work well.

Anyone got any suggestions for those? Pretend we are starting with a cross section of an involute tooth, square on the ends with no reliefs at all, just a perfect extrusion of the tooth shape. What angles would need to be added?

And what is the ideal cutter thickness?

Once I get this going, I'll start another thread with detailed pics and info so that this may help others millions of years from now.

Toolguy
10-10-2014, 11:56 AM
The 1018 will not harden. Use a common tool steel. McMaster-Carr, Enco, MSC etc. has it for pretty cheap. For small gears (1 diam. or less) I would make the cutter 1/8" thick. For bigger sizes, 3/16 to 1/4" thick.

For back clearance use a tapered end mill, 5 deg. min, 10 deg. max taper per side, depending on how much back clearance you need for a given situation. That will give you back clearance in all directions while still being able to cut the part in a horizontal orientation to preserve the CAD geometry. You will need to start a little oversize, cut and measure, then adjust the cutter comp to get the exact size. I tend to go for more back clearance on soft materials (ali or softer) and less on harder materials (mainly different steels).

doctordoctor
10-10-2014, 12:19 PM
The 1018 will not harden. Use a common tool steel. McMaster-Carr, Enco, MSC etc. has it for pretty cheap. For small gears (1 diam. or less) I would make the cutter 1/8" thick. For bigger sizes, 3/16 to 1/4" thick.

For back clearance use a tapered end mill, 5 deg. min, 10 deg. max taper per side, depending on how much back clearance you need for a given situation. That will give you back clearance in all directions while still being able to cut the part in a horizontal orientation to preserve the CAD geometry. You will need to start a little oversize, cut and measure, then adjust the cutter comp to get the exact size. I tend to go for more back clearance on soft materials (ali or softer) and less on harder materials (mainly different steels).

Okay thanks, I thought 1018 was tool steel. Sorry, with all the responses I missed your specific recommendation against 1018 and your suggestion of O1 and W1. I'll get some of those.

All my tapered end mills are between 1 and 3 degrees so I am going to attempt to machine this using the ball mills I have.

I'll have to draw it up in SW and then see what ball mills will be able to do that taper.

So I take it that the cutting surface of the cutter remains totally flat, correct? And the taper would be on the rear and symmetrical?

There is no "relief" on the front cutting surface? Can you explain why that is for my education?

Should be interesting!

But I MUST get some work done today. So once I've done my fair share of machining stuff I should be machining, I am going to jump on this like white on rice.

I have been waiting FOREVER for a project that combines CAD, CAM, gear making, cutter making, steel, and hardening of steel..and now I have one that is combining all of them lol!!! This is going to be so much fun!

Toolguy: do you have any pictures of cutters you have made and/or hardening of cutters you've done? I think you may have posted some before so if so just let me know which thread it was please.

doctordoctor
10-10-2014, 12:29 PM
okay so pricing from mcmaster is pretty great:

3 feet of O1 1/2" diameter rod: = $7.16 (p/n 89905K11)
3 feet of W1 5/16" diameter rod: = $5.50 (p/n 8890K232)

Since I am not absolutely sure yet what stock size would be most appropriate, I'm not going to order anything. Looks like tooth to tooth with the calipers is about 0.275" on the Atlas change gears, so I would imagine 5/16" would be big enough for stock..but I'll wait until I'm sure.

Mcmaster gets stuff to me in 1 day with $5 shipping so I'll just wait until I have the toolpath and setup totally done before I order stock. But its good to know how cheap it is, even from McMa$$$steal.

Toolguy
10-10-2014, 01:18 PM
Doc-

I don't make too many cutters out of tool steel, but I make and harden a lot of parts. Those are mostly for customers, so have gone out the door. I do sharpen drill bits and endmills and make some of my own carbide insert tooling.

I do suggest some top rake rather than just flat for the gear tooth cutters, maybe 5 degrees. This would be done as a last milling operation before heat treat by setting the cutter in the vise on an angle. Also, the tool steel can be had in square and rectangular cross section as well as round.

SGW
10-10-2014, 06:21 PM
Have you seen this? http://homepage.ntlworld.com/peter_harrison/workshop/gearcutting/index.htm

I haven't tried it, but it looks like an excellent approach.

doctordoctor
10-10-2014, 06:26 PM
Have you seen this? http://homepage.ntlworld.com/peter_harrison/workshop/gearcutting/index.htm

I haven't tried it, but it looks like an excellent approach.

Hmm that is interesting...

The thing is my ultimate goal is to make truly accurate spiral bevel gears for something else, so I need to establish a foundation of methods which are as close to correct as possible. In other words, only limited in precision by the precision of the machine, not by the actual design of the method.

doctordoctor
10-10-2014, 06:28 PM
Okay I just bought two 3 foot sections of W1 and O1 half inch rod. Mcmaster says its shipping today by USPS priority mail so there is a strong possibility I'll get them tomorrow!

In the meantime, I can start getting the CAD and toolpath going..

mickeyf
10-10-2014, 07:25 PM
I try to get away from manual processes as much as possible And yet, it seems you're clearly doing this largely for the process, not the end result?

I'm now going to display my total ignorance of CNC, and my delusion that CNC means, "if you can describe it mathematically your machine can produce it". Why not just make the gears themselves on your CNC mill?

George Seal
10-10-2014, 08:28 PM
Doc
since you have ordered O-1 & W-1
get into the habit of marking as it comes into the shop. Six months from now you wont have a clue what the metal is.

I use 5 different colors
1018 = green
ss = red
O-1 = yellow ( I only buy O-1 )
12L14 = blue
4140 = black
Use what ever works for you. Mark one end and always cut from other end

doctordoctor
10-10-2014, 09:03 PM
And yet, it seems you're clearly doing this largely for the process, not the end result?

I'm now going to display my total ignorance of CNC, and my delusion that CNC means, "if you can describe it mathematically your machine can produce it". Why not just make the gears themselves on your CNC mill?

Its for both. I want a controlled, precision process, that results in a precision object, and reduces labor and inconsistency as much as possible.

Actually I could just machine the gears from scratch. But as I said before, this way of doing it is like a 6-way win of many different things Ive wanted an excuse to learn, all rolled into one.

So I have the first shot at a toolpath ready. The process went like this:

1- first I needed an accurate model of an involute tooth gear that will mate with the Atlas gear. Best guess at this time is 16 DP..but I'm a little suspicious of that. Not critical at this stage, its easy to tweak. I look around for free spur gear geometry generators and the easiest one I found was actually a little tool in the free emachineshop.com CAD package. It spat out a 2D DXF that went right into solidworks.

2- once I had the spur gear sketch in SW, I extruded it to form a solid gear. Then I put it into an assembly, and inserted a new part on its flat face. I projected the tooth shape onto this new part, and now I have a "negative" of the tooth shape as a solid. I then added the 5 degree taper Toolguy said was required. I did not add a relief to the cutting edge, because I'm not sure how you can add a relief to a form tool like this without entering into the taper, and thus causing the form to be smaller as the cross section goes deeper into the relief. Am I missing something?

3- I then generated a simple z level toolpath. This will not create a smooth finish on the taper, but I realized, that doesn't actually matter, because the taper doesn't do any cutting. The only surface that matters is the front, and that will be razor sharp since it will be done it one continuous cut.

4- The simulator shows that with a 1/4" diameter end mill, 100% of the geometry is cut. Which is a little surprising given that this tooth shape has a small undercut. NOTE: I re ran the sim with much higher resolution and it shows that there is about 1 thou remaining in the undercut area as expected. I'm not concerned with this right now because its easily remedied by adding a finish pass with a smaller end mill. So I'll leave it for later.

Now all I need is to rough some steel to size and stick it out of the vise and hit go. I'll have to experiment with feeds and speeds for the 1/4" carbide 2 flute I'll be using, but I think its worth a shot just for fun.

http://i242.photobucket.com/albums/ff197/acannell/gearassy_zps358700c6.jpg (http://s242.photobucket.com/user/acannell/media/gearassy_zps358700c6.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/toolpath_zps0530d297.jpg (http://s242.photobucket.com/user/acannell/media/toolpath_zps0530d297.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/sim_zpsd9b4b6f8.jpg (http://s242.photobucket.com/user/acannell/media/sim_zpsd9b4b6f8.jpg.html)
3-

doctordoctor
10-10-2014, 09:05 PM
Doc
since you have ordered O-1 & W-1
get into the habit of marking as it comes into the shop. Six months from now you wont have a clue what the metal is.

I use 5 different colors
1018 = green
ss = red
O-1 = yellow ( I only buy O-1 )
12L14 = blue
4140 = black
Use what ever works for you. Mark one end and always cut from other end

will do..actually Ive been doing that for awhile now. Whenever I buy any material now I write the full alloy name on it with a paint pen.

Too bad I already have quite a supply of "mystery" stuff. At least its good for fabbing/welding stuff but not so much machining.

garyhlucas
10-10-2014, 09:57 PM
I've been wanting to try shaping tools on my CNC mill too. I have SolidWorks, AutoCad, Rhino, and CamBam so the software part is simple enough. I'm thinking a CBN grinding pin in my router spindle to do the cutting. Lots of fine passes with mist cooling, and the vacuum taking away any abrasive dust. I often make shaped tools for plastic parts we use, but getting the tool ground accurately has always been tough for me.

doctordoctor
10-10-2014, 10:10 PM
I've been wanting to try shaping tools on my CNC mill too. I have SolidWorks, AutoCad, Rhino, and CamBam so the software part is simple enough. I'm thinking a CBN grinding pin in my router spindle to do the cutting. Lots of fine passes with mist cooling, and the vacuum taking away any abrasive dust. I often make shaped tools for plastic parts we use, but getting the tool ground accurately has always been tough for me.

I dont see why you couldnt do exactly what Ive laid out above. You would use the "waterline finish" as they call it in Cambam. Can you cut steel with your router? I would imagine a small bit at very high speeds? In that case, you could probably do literally exactly what Ive shown so far. It sounds like we may be in the same boat. Have you tried getting your cutter drawn up in SW? Whats your cutter for?

Toolguy
10-10-2014, 10:11 PM
Doc-

You are on the right track. With my tapered endmill method all you need is one toolpath of the cutting edge rather than many steps of different toolpaths. If I were doing that part, I would cut it first with a straight 3/16 endmill, then come back in with a tapered endmill to where it just meets up with the cutting edge made by the straight one. You would have a finished cutter in a couple of rough and finish passes of each tool and only one 2 dimensional tool path for the finish cut.

doctordoctor
10-10-2014, 10:20 PM
Doc-

You are on the right track. With my tapered endmill method all you need is one toolpath of the cutting edge rather than many steps of different toolpaths. If I were doing that part, I would cut it first with a straight 3/16 endmill, then come back in with a tapered endmill to where it just meets up with the cutting edge made by the straight one. You would have a finished cutter in a couple of rough and finish passes of each tool and only one 2 dimensional tool path for the finish cut.

I would concur that would work, but let me ask you this, the finish on the tapered section doesnt really matter, does it? Its always in air right?

Also, I want to accomplish this using generic tools like straight ball and flat end mills. That way I can tweak angles as needed. Not so important for the taper, but for the relief it might be, and also for other tools I may get involved making later.

What do you think about the no relief on the cutting face? How do you put a relief there if its tapered behind it? That would make the cutting cross section too small wouldnt it?

lakeside53
10-10-2014, 10:58 PM
Download the demo copy of Gearotic 2.0. It will let you easily model what you are going and generate the g-code to cut the gear with a straight end mill if you want. If you like it, buy it ($120).

http://gearotic.com/

From their FAQ section.

What outputs can I get for my gears?

-- It varies for each gear type. For example, Spur gears will output as DXF, Gcode - both 2d and 4th axis output, 3d DXF's, STL's and as paper printouts at 1:1 scale.

doctordoctor
10-10-2014, 11:15 PM
Download the demo copy of Gearotic 2.0. It will let you easily model what you are going and generate the g-code to cut the gear with a straight end mill if you want. If you like it, buy it ($120).

http://gearotic.com/

From their FAQ section.

What outputs can I get for my gears?

-- It varies for each gear type. For example, Spur gears will output as DXF, Gcode - both 2d and 4th axis output, 3d DXF's, STL's and as paper printouts at 1:1 scale.

I already have a model of the spur gear in solidworks, as I showed. If I wanted to just cut it out I could. Getting a gear made for the atlas lathe is probably the least important goal of all this. Have you read any of this thread?

lakeside53
10-10-2014, 11:51 PM
I did. You can use gearotic to help make your single tooth cutter.

doctordoctor
10-11-2014, 12:01 AM
I did. You can use gearotic to help make your single tooth cutter.

I already have the cutter modeled, what advantage would gearotic offer?

lakeside53
10-11-2014, 12:15 AM
Code to cut it? but, if you already have that it doesn't matter.

rcaffin
10-11-2014, 12:57 AM
I want to have a controlled , parametric process from CAD to machined cutter that lets me make arbitrary cutter shapes and arbors to hold multiple custom cutter inserts. Ill be working on lots of other gear types and some of them will need many cutters and they will all need to be the same.
Sounds like a rotary table with a trunnion - that is A & B axes, plus a diamond wheel on the mill head. On a solid 3-axis mill of course.
OK, you could get away with a single A axis tilted up at 5 degrees and an alox wheel on the mill.
Programming it will be ... FUN!

But with this (5 axis) you can make most any sort of cutter you want. That includes end mills, ball end mills, ...

My T&C grinder is manual, but I dream...

Cheers
Roger
PS: why a flat face to the cutter, unlike the reliefs on the faces of carbide inserts? Becasue it is hell to program and cut out any concavity!

rcaffin
10-11-2014, 01:04 AM
let me ask you this, the finish on the tapered section doesnt really matter, does it? Its always in air right?
NOT NECESSARILY!
If the tool leads the axis of rotation you are OK, but if it trails then the heel can rub on the workpiece - badly.

Cheers
Roger

doctordoctor
10-11-2014, 01:24 AM
Sounds like a rotary table with a trunnion - that is A & B axes, plus a diamond wheel on the mill head. On a solid 3-axis mill of course.
OK, you could get away with a single A axis tilted up at 5 degrees and an alox wheel on the mill.
Programming it will be ... FUN!

But with this (5 axis) you can make most any sort of cutter you want. That includes end mills, ball end mills, ...

My T&C grinder is manual, but I dream...

Cheers
Roger
PS: why a flat face to the cutter, unlike the reliefs on the faces of carbide inserts? Becasue it is hell to program and cut out any concavity!

I dont know what sort of CAD/CAM package you are using, but programming concavities is really trivial. Any 3 axis face is trivial actually. Are you talking about writing G code by hand or something?

With regards to the taper, to be clear, I'm talking about the taper BEHIND the cutting face not the relief on the cutting face. That surface (the taper) should never touch the gear surface. By finish quality, I mean 1 thou scallops. Would that really matter? Its equivalent to the outside surface of an end mill behind the cutting edge. If that touches the work, something pretty bad is going on,yes?

So I just HAD to do something..I didnt want to wait until the O1 and W1 rods I ordered got here. I had a 1/2" 41L40 rod laying around, so I decided to whip something up just for fun.

It wont work as a cutter, because the cutting side isnt machined. But I wanted to see what the tapered section look like, because thats the part which has some actual 3 axis going on. And I dont have a clear idea of what the relief geometry on the cutting face is supposed to be, so its sort of pointless to cut it at this time. Once I know it shall be done!

I roughed the rod down to a little over the resultant cutting width, which is .164"!

Then I let the toolpath go to town. First, roughing with a 1/2" 4 flute carbide hog, then a 1/4" 4 flute carbide finisher. The tools werent even trying and I was running very conservative feedrates. All went quiet and smooth. A whopping 3 minutes later, out pops the tapered geometry. The scallops between cuts are so smooth I can barely feel them. There is nothing in the toolpath even attempting to remove them. If I changed the stepover and used a ball mill finish pass they'd be toast.

http://i242.photobucket.com/albums/ff197/acannell/20141010_215651_zpspeo5cpx7.jpg (http://s242.photobucket.com/user/acannell/media/20141010_215651_zpspeo5cpx7.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/20141010_220829_zpsoan9ovef.jpg (http://s242.photobucket.com/user/acannell/media/20141010_220829_zpsoan9ovef.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/20141010_220958_zpsncqwmbd2.jpg (http://s242.photobucket.com/user/acannell/media/20141010_220958_zpsncqwmbd2.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/20141010_221023_zpsklcwqgql.jpg (http://s242.photobucket.com/user/acannell/media/20141010_221023_zpsklcwqgql.jpg.html)

doctordoctor
10-11-2014, 01:24 AM
http://i242.photobucket.com/albums/ff197/acannell/20141010_221221_zpsek1bko4w.jpg (http://s242.photobucket.com/user/acannell/media/20141010_221221_zpsek1bko4w.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/20141010_221241_zpssqgdzgjp.jpg (http://s242.photobucket.com/user/acannell/media/20141010_221241_zpssqgdzgjp.jpg.html)

http://i242.photobucket.com/albums/ff197/acannell/sim_zpsd9b4b6f8.jpg (http://s242.photobucket.com/user/acannell/media/sim_zpsd9b4b6f8.jpg.html)

PStechPaul
10-11-2014, 03:43 AM
Looks almost like my square thread cutting bit:

http://enginuitysystems.com/pix/tools/Square_Threading_1297_800p.jpg

J Tiers
10-11-2014, 11:02 AM
Depending on the type of cutter, the finish on the part behind the edge may matter quite a bit.

For the cutter shape you pictured, not so much. But for a regular multi-tooth round gear cutter, the finish on the "gullets" matters because the chip touches those. You want the chip NOT to "catch on" anything, but instead, to slide away easily and fall out after exiting the cut.

rcaffin
10-12-2014, 12:26 AM
Hi Doc

Are you talking about writing G code by hand or something?
Full 3D CAD I have (Alibre, SolidEdge, ...) but not CAM. For the sorts of things I machine, I would need to spend $50k and buy another 2 axes - and the production rate would be lower than I have now. Yes, I hand code, and I set up production in batchs, in jigs, in parallel. It works just fine. Mainly 6061 and Fortal.

However, I learn something new every day. Your method, of tipping the workpiece upside down to machine it and doing it in multiple passes like that to get a taper ... CUTE! Yes, a carbide ball end cutter would be nice, or even a square end mill with rounded corners.

Cheers
Roger

doctordoctor
10-12-2014, 02:00 AM
Hi Doc

Full 3D CAD I have (Alibre, SolidEdge, ...) but not CAM. For the sorts of things I machine, I would need to spend $50k and buy another 2 axes - and the production rate would be lower than I have now. Yes, I hand code, and I set up production in batchs, in jigs, in parallel. It works just fine. Mainly 6061 and Fortal.

However, I learn something new every day. Your method, of tipping the workpiece upside down to machine it and doing it in multiple passes like that to get a taper ... CUTE! Yes, a carbide ball end cutter would be nice, or even a square end mill with rounded corners.

Cheers
Roger

Its pretty basic 3 axis machining, or more accurately, 3 axis CAM. All the geometry can be accessed from one side so a z level finish does it pretty easily. Technically its not really even 3 axis. Only 2 axis were moving simultaneously. For it to be 3 axis the toolpath would have needed to be a slice or UV and rotated 90 degrees but it wasnt required for this toolpath so why bother. Why would it cost you $50k for 2 more axes? And why 2 more axes? You only have 1 axes now and 3 axis would be $50k more?

Paul Alciatore
10-13-2014, 12:36 AM
Understood, knock yourself out.



Just buying the cutters doesnt do me any good as far as learning about making cutters or creating a process I can use later to make arbitrary cutters.

rcaffin
10-13-2014, 12:52 AM
Its pretty basic 3 axis machining, or more accurately, 3 axis CAM. All the geometry can be accessed from one side so a z level finish does it pretty easily. Technically its not really even 3 axis. Only 2 axis were moving simultaneously.
True. It's a little bit above a router though.


Why would it cost you $50k for 2 more axes? And why 2 more axes? You only have 1 axes now and 3 axis would be $50k more?
Not quite what I meant. As far as I can see, and my vision may be limited, there's 3-axis CAM and there's 5 axis CAM. The 3-axis stuff I can program in my head, and yes, I do mean 3 axis. But once you add in an A axis and a B axis ... my head starts to ache :p

For what you can do with a 5-axis machine and the CAM SW to match, check out Boris the Spider at
http://www.delcam.tv/delcam-video.asp?VideoId=302
I think the machine cost around $1M and the CAM SW cost around $100k - from memory. He took a couple of days to machine.

But Boris is kinda cute.

Cheers
Roger

doctordoctor
10-13-2014, 12:43 PM
Ive noticed one thing that happens alot on internet forums is when people feel threatened they resort to citing accomplishments of third parties, to try and equalize things, since they cant do so with their own accomplishments. Happens alot in person too, at first I thought it was just young guys, but I notice alot of older guys (baby boomer generation) do it.

I dont see guys over 75 doing it though. I think that generation never got the hurt ego virus that infected the later generations.

And then you have anyone born after 2000. They think the idea of an "accomplishment" is an offense altogether.

Axkiker
10-13-2014, 12:51 PM
I cant tell you the best way to make the cutter but I can tell you what I did to get around the whole issue. I used UHMD plastic to make the gears out of. That way I didnt need any hardened cutter to cut the gears. They are going on 2 years and dont show the first sign of wear.

rcaffin
10-14-2014, 12:28 AM
when people feel threatened they resort to citing accomplishments of third parties, to try and equalize things, since they cant do so with their own accomplishments.
Was this aimed at my comment about Boris? I dunno. I thought he was cute, but I sure as hell couldn't machine him myself!

Maybe I should point out that a) all Australians are thick-skinned, and b) I am thick-skinned for an Aussie.
Mind you, I am not that far off 75 myself - sigh.

I agree with axkiker re UHMW plastic: it goes great.

Cheers
Roger

lakeside53
10-14-2014, 12:36 AM
I'm ignoring DD's rhetoric. He's what the other forums call "a troll...".

RichR
10-14-2014, 01:21 AM
From the original post:

... And then..can you do the same with aluminum? What happens if you try cutting aluminum with something softer than HSS? ...

To answer that, check out the dovetail cutter some joker made using some mystery steel from some old ankle weights:
http://bbs.homeshopmachinist.net/threads/63200-Dovetail-mount-for-telescope

J Tiers
10-14-2014, 08:27 AM
Ive noticed one thing that happens alot on internet forums is when people feel threatened they resort to citing accomplishments of third parties, to try and equalize things, since they cant do so with their own accomplishments. Happens alot in person too, at first I thought it was just young guys, but I notice alot of older guys (baby boomer generation) do it.

I dont see guys over 75 doing it though. I think that generation never got the hurt ego virus that infected the later generations.

And then you have anyone born after 2000. They think the idea of an "accomplishment" is an offense altogether.

How to win friends and influence people........

What if I mentioned that when a person's argument goes south, some folks resort to attacking the personality, tactics and motivations, etc of the person who shows them up? It has zip to do with the facts, but may possibly allow the attacker to feel "powerful" and "right" again.

Could that be true of you? Only you know for sure.

I can say this here, since I have no "dog in the fight".