[Chris D]

Hello group,

I want to play around with some gear hobbing. Nothing fancy, but would like to tinker with it.

I want to cut two spur gears that mesh together, one having 100 teeth @1.9894 dia. and the other having 20 teeth @ .3979 diameter. This will give me a ratio of 5:1 and a tooth pitch of .0625. When I draw this up in CAD, I notice that the teeth will bump into each other as they rotate through.

For the hob, I plan on turning it on a lathe with a custom ground threading tool to match the tooth form. The actual hobbing will probably be done on my mill as there is more room on that machine for fixturing the blanks etc.

I am pretty sure that the shape of the threading tool is not going to be the exact shape of the tooth form. I suspect that the hob is going to alter the tooth shape to "automatically" provide the tooth shape to provide the clearances so that the points of the teeth don't rub as they rotate together.

Does anyone have any experience with home shop hobbing that can provide some pointers?

I have plenty of machining experience, but and pretty stupid when it comes to gear machining (except for making the blanks).

TIA!

Chris

[Mcgyver]

Every time I think of making a gear hobbing machine the difficulties in making the hobs, not the machine, talks me out of it. The form on the hob looks more like an acme thread as I recall, but the difficult is cutting the relief. Picture cutting a very coarse thread with compound feed disconnected and moving in and out on a cam several times per rev to create relief behind the teeth.

Did you arrive at those dia's using gear formula? I think you've done it correctly, as both have the same dp, but the dia you get is the circular pitch, not the outside dia. Maybe that’s why it’s not laying out well.

There were some excellent articles a few years ago in making hobs in Model Engineers Workshop magazine. There was also a series on making a hobbing machine from castings, of bench top dimensions. Well worth tracking down some copies – most of the theory and practical aspects are covered

[DareBee]

You could buy a gear form cutter (like you would run in a horizantal mill) and slice it int 4-6 (or whatever) pieces and you will have a handfull of "pre-shaped" hobs that you can make a toolholder for.
On the other hand... I assume you must have a rotary table the you will be using the index between teeth? It would save you many,many,many hours of hobbing if you would stand your rotary up, fixture your blank on an arbour, and just cut the gears with the horizantal cutter in an arbour.

[Mcgyver]

I know just enough about this to help out a bit, but if there is an expert around jump right in.

the profile of a tooth of a gear cutter is not the profile for a hob and the hob doesn't require any indexing, only the relative rate of rotation between the hob and the blank. I think the profile of the hob, which is like an acme thread, is determined by the pressure angle.

The advantage of hobbing is you only need 1 cutter for any number of teeth (for a given dp) and you can cut bevel gears etc. It's also more accurate as the gear is generated as opposed to the form cutter which is an approximation (that’s why each only cuts a range of teeth). Without a huge amount work, I don't know how you could hob in the mill as the basic principal is two rotating shafts connected via a gear train, one with the hob, the other with the blank and slide mechanism to control the distance between in two planes (depth of cut and traverse the hob across the blank).

I agree with the recomendation of just cutting the conventional way though


Here's a link for a video claiming you can easily do it for $3 http://technicalvideorental.com/rental_15.html

If thats the case, I'm guilty of overestimating the rigores of getting set up to hob gears!

[Mcgyver]

I learned some neat ways to make gears with cnc applicability - thought this crowd would find it interesting.

First of all, Peter Harrison's site gives a very good description of what I'll call faceted hobbing. Its not really hobbing but it is a very cool way to make your gears and cutters without requiring true hobs, a swiveling mill table etc. As this needs is a lot of back and forth stuff/index repetition it would be great with CNC. You double up the number of facets approximating the tooth form by offset y ½ and indexing to the ½ tooth. This holds real potential for us home shop guys – don’t forget expensive commercial form cutters are also only an approximation.

http://homepage.ntlworld.com/peter_h...ting/index.htm

For the Rolls Royce idea check out this photo John Stevenson’s site. True cnc hobbing. He’s encoding the spindle speed to control the blank speed. Absolutely brilliant! This idea does require a proper hobber. The table is turned by the pressure angle, and you can cut helix gears by subsequently setting the 4th axis over. DIY CNC helix and crown gears – that’s taking it to a new level imo

http://homepage.ntlworld.com/stevens...20indexer4.jpg
http://homepage.ntlworld.com/stevens...0indexer12.jpg

[Ball Screwer]

There are several books and web sites on alternative methods of gear generation.

[handlewanker]

Hi all, I know this is probably a bit late in the day now, but from my point of view anyone contemplating hobbing a gear using the knowledge that you don't have, is trying to use a sledgehammer to crack a wallnut.
Hobbing is a production process and no-one in their right mind would set up a machine just to cut one or two gears.
It is the simplest job to set up a horizontal mill with a gear tooth milling cutter and dividing head and get accurate results in a very short time, and all this with a bit of tuition and knowledge.
I have hand ground a flycutter to cut gears when the required gear cutter was not available.
The flycutter was checked on an optical projector against a 50:1 projection chart for form accuracy and worked well.
I wouldn't recommend this for some very sophisticated gear requirement but the job in hand was for a change wheel for a lathe and was still working years later.
For complexity I would rate it the same as for cutting threads in a lathe.
I've also cut spiral gears, the same way, for a right angle drive in a horizontal mill with a vertical head, and the dividing head geared to the table leadscrew, a bit more complicated.
On one other job I had to cut two small racks, about 4" long with teeth at 1/8" pitch.
This was done using a Bridgeport mill and the cutter was a flat end mill type cutter ground to the tooth form.
The pitching of the teeth was done by the leadscrew of the X axis, and cut by traversing the cutter across the rack blanks,held in the vice, with the Y axis.
I still have the cutter, after 30 years, as a souvenir in case anyone doubts me.
Ian.

[el_tony33]

Hey there people!

I´m designing a gear making machine for a school proyect, and I'm really interested in any information that you can provide me please, about gear manufacturing

I already know there is a lot of books about this, but in the web? do you have a good secret place? jajaja?

I´m really in love with gears, and I know some automation concepts, all that I need is the step by step of the process, or a table of speed feed rates, speed recomended for the hob, etc...

In abstract the technical data of the process...

I promise that I will reply the advance of the proyect, that is intended to manufacture plastic and aluminium gears, because is going to be a "table size" machine..

Ok... well,... thanks for reading!

[RICHARD ZASTROW]

Tony, Do you intend to make a true gear hobbing machine or a milling machine using gear form milling cutters?

I'm in the process of designing a commercial gear hobbing machine. This is much more complicated than a milling machine.

If it's a CNC machine you will need to think about a few things like;
1) a work spindle drive (C axis)
2) a hob spindle drive (A axis)
3) a longitudinal drive (Z axis)
4) a cross feed drive (X axis)
All the above must be able to move under simultaneous control to produce a crowned and/or helical gear. It is possible to hob a plain spur gear by positioning the X axis for depth of cut and using C, A and Z axis' to cut the gear.

In addition, the hob spindle must be tilted to the correct angle to allow for the hob lead angle and if required gear helix angle. This can be done manually or as an additional axis 5.) (B axis).

Also, to spread wear on the hob itself, another axis is employed as the hob shift which can also be manual or under CNC control 6) (Y axis). This is also used to employ more than one hob on a common arbor to cut more than one pitch gear on the same workpiece.

If a CNC machine, you will need some pretty fancy software to co-ordinate all these motions.

Good Luck!!

Dick Z

[RICHARD ZASTROW]

Tony, Add. to post #10 above.

If you intend to machine gears by the milling method rather than hobbing, a far less complicated machine will work. Straight spur gears are the easiest and a good starting point.

A milling machine with an accurate indexing device is all the machine needed. An accurate arbor to hold the gear form milling cutter and an accurate arbor to hold the gear being cut is all the durable tooling required. The reason for all the "accurate" in this paragraph is to avoid the compounding effects of any inaccuracies.

For expendable (cutting) tools you will need gear form milling cutters. These can range from high speed steel, coated high speed steel, coated powder-metal high speed steel, carbide through coated carbide form cutters. There are indexable insert type cutters available.

There are different cutters for every pitch of the gear and different cutters for the number of teeth within a given pitch.

The speeds and feeds for these are available from the cutter manufacturers. Different speeds for different cutting tool materials and more different speeds for materials being cut.

One source for these cutters is Ash. I will try to locate their web page.


Dick Z

[el_tony33]

Thanks RICHARD!! For the info!

And...oooh wow, it is going tobe more challenging that I've been thinking!! jaja

well thanks for the help, and about the inclination of the hob for cutting is something I don't contemplate from the beginning, thanks!

I have been thinking in that axis of movement but not for that purpose, it was for making the inclinated tooths.

I´m working on the interface for the computer-program-driver-motor, I got a really fancy motors jaja and drivers too, I´m using some VEXTA, 5 Volts, 1Amp, And the driver is a Centent CN0162, for each axis.

I´m trying to control it from a program, using a laptop via RS232, the only inputs that need the centent drive is the direction and step pulse.

Becaue some inconvenients of time I can't make this divers work, RICHARD, have you or any one use it before? I found the manual but is not really good about explainig, and have a very few images and squematics

Ok back to the machine, I´m contemplating 4 axis:

1.- Z axis: The one that carries the motor coupled to the hob
2.- A axis: Rotation of the Z axis respect to the horizontal
3.- Y axis: Movement to get near the "work disc" to the hob to make the calibrations depending of the size of the work disc
4.- B axis: Rotation of the coupling that carries the work disk, for the spin feeding of the material, having the angle of rotation in his centre

I don`t know if I get explained :O?

Well, The angle that RICHARD say and I mention before, the one of the gear hob, it can be compensed with the A axis I suppose?

The milling methos the inconveinence that I see on it is the number of milling tools that you need for make a game of diferent number of tooths an module :S... So I was thinking that, yes, is going to be hardy, but at the end is much more better and "special" machine, and thats the goal of the proyect, to help to the design of the new generations of ingennering in my school, because in my generation we take a lot of time looking for the right gear and that kind of things that end with your designing dreams... So if you want to use gears, let's use the TONY'S MACHINE jajaja.... :P

Well I'm still working on it!!

Is really late, but in another post I will put some drawings that I made...

welll thanks again RICHARD!!

[RICHARD ZASTROW]

tony, Get some prices on gear hobs. I think you will find that you can buy many gear form milling cutters for the price of one hobbing cutter.

I believe your motors/drivers will be a bit "light" for this job. The smallest hobber I ever worked on made gears smaller than 1" (25mm) in diameter. The spindle motor has 4.3 Nm continuous torque.

Remember, the hob spindle (A-axis), the work spindle (C-axis) and the hob directional feed (C-axis) must be ACCURATELY synchronized or you will produce a corn cob rather than gear.

People who create this kind of software can spend years developing it. Spur gears are the least complicated. In it's simplest form it's 3-axis. Include crowning and it becomes 4-axis. Add helical gears to the mix and it becomes 5-axis, all must be synchronous.

The hob shift is a 6th axis but not necessarily synchronized with the other 5 axis.

If you add multiple hobs to the mix, the hob shift an be employed in the CNC program. Adding more than one gear pitch, gear helix angle or hob helix angle and the hob head swivel angle will have to be included in the CNC program as well.

I'm not trying to discourage you, but trying to inform you of someof the "complications" you may/will run into.

Good Luck!

Dick Z

[handlewanker]

Hi all, if you're going to "hob" a gear then that is purely a mechanical process, nothing to do with CNC, and totally not even capable of being CNC'd.

CNC'ing a gear would mean you would have to use a 29 deg shaped end mill in a vertical head, that is not having parallel sides but ground to a 29 degree included angle.

This would then cut the gear with the blank mounted on a stub arbor in the 4th axis using a back and forth action to initially get the bulk of the material out of the centre of the tooth form, which would leave you with a 29 degree straight sided slot like a rack, after which the true form for the number of teeth and DP of the gear would be achieved by incremental rotation of the gear while the cutter was also incrementally move in the Y axis direction for each side of the gear tooth, which as has been said would leave you with a small amount of facets.

To remove the facets would require a regular parallel end mill, no bigger in diam than the root width of the gear, working parallel to the gear face and just shaving the face as the cutter traversed the profile under CNC control.

You can imagine this if you drew a gear on a piece of paper and pointed your finger down at the gear tooth form which represents the path the end mill would have to take, under CNC control, to shave the gear.

Gears are usually about four times the tooth depth in width, so an end mill this small in diam and length would have to mill from both sides to get full width of the gear tooth form accurately.

If anyone wants the magazine numbers for making a gear hobbing machine and hobbs, as shown in the UK publication Model Engineer back in the 70's I'll dig out the magazine numbers and dates and let you have the details, but this is purely for a mechanical set-up.
Ian.

[RICHARD ZASTROW]

Ian, I think you must be referring to "CNC'ing" as an end milling process rather than hobbing.

Nearly all current gear hobbing machines are CNC's. Personally, I've been designing them since 1995. Not a DIY project.

However, you can do a reasonable job of making some gears on manual or CNC milling machines. 1-off's etc. but not commercially.

The cost for tooling etc. is quite high and really only makes sense for gear production shops.

Easiest and least expensive DIY and prototype gears not requiring high precision is form cutters on milling machines with indexers or rotary 4th axis.

A modern gear inspection machine is well over $250,000. A coated carbide hob 1 1/2" dia X 2" length can easily cost $2,000, some much more. Precision hob arbors can cost another $1,500.

No offence intended, just free info. which is worth what you paid for it. LOL

Dick Z

[Mcgyver]

Nearly all current gear hobbing machines are CNC's. Personally, I've been designing them since 1995. Not a DIY project.

cnc is I suppose a matter of degree in that the following isn't a full cnc gear generation machine, but it is a diy hobbing set up using cnc (to coordinate spindle and work) on a universal mill (John Stevensons work)....hobbing on a universal mill sans gear trains

http://homepage.ntlworld.com/stevens...0indexer12.jpg

[handlewanker]

Hi, thanks Richard, no offence taken.

Yes I was referring to the end mill as a "generation" aspect under CNC control, which I assumed was the reason why anyone would want to get away from buying an expensive hob.

Some years ago, 1975, I had to cut a number of small racks for a one off job we were working on, and I resorted to regrinding a slot drill to 29 degree included angle to cut the rack teeth in a Bridgeport mill by holding the rack blanks in the mill vice and traversing the end mill back and forth for each tooth using the X axis leadscrew for pitching.

It's a slow process, as you have to progressively cut downwards as you traverse forward and back across the blank to prevent side deflection from the slender tooth form cutter. (I've still got the cutter)

Having succesfully made a number of racks in Nickel Chrome alloy steel, I also cut one more gear wheel later on using the same type of hand ground cutter, and ground the tooth form using a shadow graph to check the form shape for the number of teeth in the gear.

Once again the cutter needs to be traversed forward and back, to cut on both sides of the gear tooth alternatively, otherwise you get side deflection if cutting on one side only.

Using this method the CNC control would really lend itself to using an end mill ground to the 29 degree shape for 14 1/2 degree pressure angle gears which is rack form, and then using 4 th axis rotation to get the tooth flank shape for the various teeth numbers required etc, all from one simple cheap cutter.

You do of course end up with a number of small facets on the tops of the gear tooth flanks but thes can be burnished smooth by running it fairly tightly meshed with another gear for a short while.

I was under the impression that to "CNC" meant literally that the tool was generating the form by point contact, and the surface of the job was being traversed by plotting the travel with X, Y & Z movements.

Thinking about the hobbing process, I came to realise that normally, in order to utilise a hob, the blank must first be gashed, and then the hob can follow the gashes in the freely rotating gear blank to generate the teeth reqd, so if the 4 th axis is under CNC control, gashing is not required, as the hob under CNC control can now cut straight into the gear blank and work in unison with the rotating gear blank, so saving a lot of time and not having to set up gear trains etc, that's really cool.

We live and learn....LOL.
Ian.

[RICHARD ZASTROW]

I refer again to #14, line 3 above.

The commercial CNC gear hobbing machines usually have (6) axis and still rely on the cutter for tooth form accuracy.

The machines I work with typically have all axis capable of simultaneous interpolation.

Often more than one hob is mounted on the cutter (hob) arbor to cut more than one pitch gear on the same workpiece. This also requires the hob head to be swiveled under CNC control to adjust for hob lead angle and possibly gear helix angle change.

The X-axis might move toward and away from the workpiece centerline to crown the teeth.

This happens while the cutter (hob) can be spinning merrily away at over 750 sfpm and all axis stay in synchronous motion. (high speed dry machining)

This is what I mean by a "commercial" gear hobbing machine. They are very pricey!!!

More cheap info. LOL

Dick Z

[handlewanker]

I reckon you wouldn't want to miss a step somewhere along the line....LOL.
Ian.

[RICHARD ZASTROW]

If you miss ANY steps, especially if you repeat the misdeed, I'm sure you will not have the opportunity to do it again. (at least with that employer) LOL

Dick Z

[Charles Lessig]

I am a retired wind tunnel machinist and technician. My home
shop is where I do my tinkering now.

Here is a device I made for machining hobs.

YouTube - Charles Lessig's Hob Reliever Eureka Balzer Gear Cutting


I started out with Ivan Law's Gears and Gear cutting of the
Workshop series number 17. I followed his plans as closely as I
could because I had no idea how it worked. Only after making it
could I see what was going on. I think it is amazing that he
reverse engineered it from an old advertisement.

At first I just extended some of the features to hold a hob blank.
Since the lead is a factor of pi , I used 22/7, the half nuts had to stay
closed. Because it drives through a ratchet it loses place when
reversed. I was able to relieve a hob but the tool had to be reset for
each pass. It was very slow going. After a couple years the pattern
screw idea came to me and that was the answer.


Here is some more information about this device.

http://tinyurl.com/bv6vkh

This has pictures of the various parts or the set-up.

Here is the best description of Balzer's device I have found so far
from 1896. The later references have his name spelled wrong.

http://tinyurl.com/b3hpgc

This is the type that he manufactured but not the one he patented.

http://www.google.com/patents?id=PblfAAAAEBAJ&dq=535127

That one used gears. This shows the odd gear,

Here is a 1905 paraphrase of Balzer's Ratchet backing off device
patented by Taylor.

http://www.google.com/patents?id=ouBNAAAAEBAJ&dq=817885

I adapted a small mill to be a hobbing machine and am now working on
a similar set-up for a Van Norman 12 to have more capacity for bigger
gears.

I have made a set of gears, 71 and 113 which give 1/5 pi as
close as could ever be wanted for this work. The error is 85 parts per
billion, around 5 or so thousandths of an inch per mile.

I showed the device to my metal workers group but it was still hard
to visualize how it worked so that's when the video was posted.

This device can't do worm gear hobs with no center bore. For
that you still need a backing-off lathe. The time to stop is when
it gets more complicated than that. The second neat thing about
the Balzer-Eureka is the built in speed reduction.

A sharp eye will spot a couple of mistakes that don't affect
the Eureka operation. The lead screw modification was done
in a fit of fury.

This was done without using CNC which would have made it easier.

Best regards, Charlie