Help regarding internal gear

Home Model Engine Machinist Forum

Help Support Home Model Engine Machinist Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.

Don Pittman

Member
Joined
May 18, 2018
Messages
10
Reaction score
1
Hi I've done some research and forum searching but having difficulty finding needed info.

I'm wanting to attempt to cut an internal gear on a small lathe using a slotting device attached on the carriage. I want to use a single point cutter bit. I can't seem to find Info regarding how to get the right profile for this task. I know Info regarding an external gear tooth profile is easy to come by, but for the matching internal tooth profile I am lost. I want to cut a Module 0.5 72 tooth internal gear.
I have the book by Ivan Law "Cutting Gears" but it doesn't address my problem.
Can someone direct me? Probably any advice will be helpful.

thanks
 
Using software you can determine the shape of the cutter. I used a free add-in for Fusion 360 from FM Gears that lets you draw very easily an internal (or external) gear. You could then model your cutter from the resulting gear. If you don't have Fusion 360 the same author of the add in offers a stand alone version for making DXF drawings https://www.forestmoon.com/Software/GearDXF/ I didn't try it.


Hope that helps.
Mike
upload_2019-11-25_21-12-52.png upload_2019-11-25_21-15-20.png
 
Thank you for the info and link, I will study and check it out.
Much appreciated
 
You may also be able to find one on something like Grabcad, even if teh MOD is different that can be scaled provided the tooth count is similar. It's then just a case of grinding a tool to fit the print out and cutting away. I would suggest drilling most of teh waste away to help reduce what the tool has to remove. How I did one in this post and carries on down and over the page
 
Just for kicks, I wondered if an internal gear could be cut by generating the tooth shape like is done with a pseudo hob in spur gears.
Take a spur pinion made in any way you know how and hollow grind it a bit so it can cut mowing axially.
Gash the internal gear, an chuck it on the lathe.
Mount the pinion like a boring bar, then push the pinion to shave 3 or 4 teeth a few thousands.
Rotate and shave the 5 tooth and a bit more of the teeth N0 4, 3, 2 wile the No 1 recede out of engagement.
Go all around
Move the cross slide to shave the teeth a bit deeper all around.

After thinking of this method for my Forest Edwards Radial I spent $113 for the gear from SDP. It was a cope out but would have been interesting to try.
 
So I've managed (with help from friends) to get a print out of the gear I want to make and now see that grinding a single point cutter accurately that small would be dang near impossible! It's a 0.5 module gear.
I'm wanting to broach on my lathe, a brass internal 72 tooth gear. Do you suppose if I cut a few teeth with an involute cutter in a steel partial blank I could use that as a broaching tool?
 
Maybe I should cut just one tooth in the cutter blank and use that as my single point cutter broach???
 
Unfortunately that won't work

As the small gear rotates it needs gaps in the outer ring that are wider than it's teeth so if cut to the same profile as the internal gear it will just lock up. You can see it here for the gear I cut in the thread I linked to but that may not have shown up as it was an attachment.

What is the gear set going to be used for? if just a slow running lightly loaded application then it would not matter if the exact shape was a bit off and if you do it as I showed the gears can be tested for mesh before being removed so easy to take a bit more off if needed.
 

Attachments

  • xtype gear.JPG
    xtype gear.JPG
    30.2 KB · Views: 446
Don, Doing your gear is not impossible
There is so much good information available to help you in your endeavor
I think the best is work articles by the late "great" John Stevenson
Go to post 6 on this thread where I list the resources

https://bbs.homeshopmachinist.net/forum/general/76512-john-stevenson-s-gear-cutting-writeup
You do not have to "grind a cutter
You can form it as John does, or
Go to Mikes workshop where mike developed another forming method

http://mikesworkshop.weebly.com/making-gear-cutters.html

http://mikesworkshop.weebly.com/update-on-gear-cutters.html

Happy Chips !
Rich

PS By the way , when you cut an internal gear , because the cutter is on the inside of the gear circle, it automatically makes the ring gear tooth narrower to give clearace to your pinion gear.
 
Thanks for the helpful info and links. Although I'm thinking it will be near impossible to get good enough accuracy, I am still leaning to attempting to grind a single point cutter I can use for broaching.

The gear is for an "Elmers #5 geared engine". It's a big project for my skill level and tools and I am starting with the gears! I may eventually have to "bail ship" on this but I thought it would be fun trying, I've already learned a lot.
 
I looked at the GearDXF program and it is really a spur gear pattern used to cut an internal gear. The reason for saying this is that the root radius is not in the correct location but where it would be on a spur gear.

I use FreeCAD and it has the exact same approach but can export step, igus and many others. On the new 18 version their is a work bench for gears that designs a profile without specifying the root radius. The spur gear choice lets you add other features like clearance and shift in the profile which may be needed for a gear be it a spur helix or internal. The GearDXF exact same approach on the part design work bench by looking at the table for picking the list on top for that bench.

Want to see the product I have entered a design in the GrabCAD contest Grundfos Challenge. 50 and 51 mod 1mm internal gear and the mating 46 and 47 spur gear are in step files and in the mating locations.
 
If you are finding 0.5MOD to hard to see when grinding the tool you could probably get away with 1MOD and 36/18T, just check that there is enough clearance around the screw holes.
 
I looked at the GearDXF program and it is really a spur gear pattern used to cut an internal gear. The reason for saying this is that the root radius is not in the correct location but where it would be on a spur gear

Just to be clear, are you saying the resultant internal gear shape that the software yields is not quite a correct tooth form with the various clearance factors designed to run with a conventional spur gear inside it? Are we talking the same software that Mike mentions in post #2? Like maybe the external/spur gears are correct & the internal/ring gears are kind of a graphical transpose as opposed to going through the calculation gyrations to yield the tooth shape? I read the documentation but it didn't really reference anything, but that in itself doesn't mean much.
https://www.forestmoon.com/Software/GearDXF/Help/Gear DXF Help v3_1_1.pdf

Mike post #2> Using software you can determine the shape of the cutter. I used a free add-in for Fusion 360 from FM Gears that lets you draw very easily an internal (or external) gear. You could then model your cutter from the resulting gear. If you don't have Fusion 360 the same author of the add in offers a stand alone version for making DXF drawings https://www.forestmoon.com/Software/GearDXF/ I didn't try it.
 
petertha: Gears with rolling contact, involute, doesn't care if the gear is external or internal. However, to mate two gears the tip of one should not hit the root of the other. When the internal gear model has a root radius on the tip clearly the software does take into account the additional depth needed. What else is not included in the model? And what features are provided to correctly form of a gear to conform to the standard parameters we find in gear design books that are standard in the industry? I didn't see the program Gear DXF offering some of these standard parameters to adjust gears. I pointed out that FreeCAD 18 work bench for gear and specifically the spur gear/ internal gear (basically the same as Gear DXF) has added some. I looked at the other two gears the work bench offers, and I do not understand what they make.

petertha and Mike: I am throwing up the "user be where sign". The models I have pointed to in my GrabCAD are examples of what you get from FreeCAD18. I have a MathCAD7 calculation sheet for those standard gear parameters that I applied to the work bench model. Also in some comments which include me saying that I have no idea the accuracy of the model. The zoom capacity of my laptop and the FreeCAD18 software does not let me zoom in sufficiently, stops increasing the screen flats use to represent the curves of the mating gears. I can not verify the contact angle or the clearance. If you have a better program to load the step files maybe you can tell me.

Some math can be used to make good gears from the Gear DXF or any program that creates involute gears, there are many. I could use the parameters from MathCAD7 for example and change the contact angle to the programs calculated value ~ 34.2 deg. and adjust the module off 1 mm to get the base circle diameter. And by hand cut the root radius into the internal gear. So good gears can be produced. I have done this to make bevel gears from two sketches using the gear option in FreeCAD part design work bench, see the GrabCAD model of the Atlas Lathe.
 
Thanks for your reply TSutrina, that's exactly what I was wondering myself. I think 'user beware' is appropriate. No disrespect to the software/app because it looks useful, but the absence of some standard parameter inputs that influence gear shape should probably be verified before trusting the outlines 100%.

I want to better understand your methodologies. I have never used Mathcad but now you have me interested. Are you saying you use a Mathcad app to modify an existing involute shape? Or you have a more start-to-finish app where one can specify gear design input parameters & it calculates resultant tooth form? Then what - does it output x,y coordinates from which you load into CAD program and spline curve intercept or something? Or maybe it exports in DXF?

I did a Google search on internal gear + Mathcad. Not sure it was a direct hit re internal but for sure some useful gear related worksheets.
https://community.ptc.com/t5/PTC-Ma...rning-spur-gear-calculation/m-p/117478#M45971
 
MathCAD is a program to do math. I basically typed in the equations found at this web site or the handbook from Stock Drive Products metric gear book. https://khkgears.net/new/gear_knowledge/gear_technical_reference/calculation_gear_dimensions.html

Looks nice but it just doing math. A few functions in the program let you fit data with a curve and find a value that solves an equation or group of them. Lucky I have the last version that does not require a code to use. It is not open source. It is a decade old. There are a few other programs that are nice but hard to read: Scilab, Octave, and Sage.

I used MathCAD do calculate the calculation given in the gear hand book. By using software I can change the values until I like the results. I do not change the involute equation. I change the base circle diameter in the involute equation. The rest is using that curve. Pick the average pressure angle, 14.5 and 20 are standard. The diameter associated is the pitch diameter. The mode is determined by where the distance to the same involute curve that forms the other side of the tooth. The root and tip are just different diameters. The handbook defines standards for determining them. And means of changing the pressure angle choice, thus change the base circle. and the width of the gear is also changeable. Thus clearance is created by moving the curve closer to the opposite side curve.

When I was at Electrolux vacuum cleaner company an engineer in sketcher input the involute curve as a table of points and copied and mirrored it until he had a gear. He forgot about clearance and I ended up adjusting his model by moving the mirrored curve. Thank god there is a repeat after moving the angle between teeth.
 
MathCAD is version like 20 so I can not use anything published in version 7. However it is a very popular program for people to use. A lot of masters papers are done using mathCAD for the same reason I like it. Great place to find the equations and method of solving them. You could use them to write programs in an open source program.
 
I've never used MathCAD but have used Matlab quite a bit. It's also an expensive program but Octave is a completely free 'clone' which is very close to the real thing (code written on Octave will mostly run on Matlab and vise versa). On first glance they look like 'just' math solvers but in reality will handle many complex tasks including things like image analysis and complex simulations of many variables.
 
Does Fusion allow for parametric design / global variables / equations / functions etc. to design gears right inside the CAD file similar to this Solidworks example?
https://blogs.solidworks.com/tech/2014/07/how-to-create-spur-and-helical-gears-in-solidworks.html

I didn't look too hard at the SDP link that TSutrina provided, but if its basically a step-wise sequence of operations based on some defined input values, then possibly a true gear of any design with all the appropriate compensations could be entirely modeled entirely within the CAD app itself. That would cut out middle men, the in-between calculation program. And from from what I can see of discussion thus far, still requires some manipulation to turn into actual tooth curves. Even DXF import can be fiddly, there are resolution parameters & things to fart with or else you get weird things like facets or disconnected geometry.

Developing a gear inside CAD would not be button click automatic, it would likely require intermediary drawing steps so one would likely have a procedure sheet to follow. I assume there is no iteration or convergence type math involved - that would be the only show stopper. Fusion is free for hobbyists I believe?
 
I believe that all the gear programs do a good job creating the involute curves. What they seem to not do is to make the small changes that experts gear designers do. And that is expected because the coders are not gear designers. So let us say your making a planetary gear set, design three gears. The centers can not be moved to create clearance, but ideal involute curves for the three gears have no clearance and will not work. This is what the earlier mentioned person did. So who can clearance be added if the gear generating program doesn't do it? One approach is to copy the gear and rotated at the pitch diameter the clearance wanted. The gear wanted is the combine volume they have together. Roots of the gears is also a cutting or added or the tips of the mate lowered.

For the gear set of an internal gear and a spur gear where the teeth count are close or small pinion gear with undercut. The solution is to change the pitch angle which the programs do well. But you will have to determine that angle with the gear hand book. And also to make a bevel gear from the generated sketch for gears. Scale the standard gear inputs. to the two ends of the gear.
 

Latest posts

Back
Top