PDA

View Full Version : gear cutting tools



aero
04-25-2001, 03:49 PM
1) Who sells a decent selection of vertical mill gear cutting tools (not hobs)? Or - who sells them at all??? Looking for metric as well as inch.

2) I am going to be purchasing a mill soon (and dividing head assuming a rotary table won't be accurate enough for gear cutting) - it seems from reading recent posts that the Grizzly machines are decent?

SGW
04-26-2001, 07:37 AM
Several companies sell gear cutters; MSC (http://www.mscdirect.com) is probably the most well-known although others may have a better selection, I'm not sure.

A rotary table would be perfectly adequate for use in cutting gears, although it may be somewhat inconveient to set up the work. You need to "somehow" hold the gear blank so you can get at it with the gear cutter, and on a rotary table that may take some imagination. You would also need a rotary table that can be set vertically (a horizontal/vertical model). A dividing head would probably be a lot easier to use, but a rotary table would certainly be accurate enough for anything you're likely to be doing.

I've never seen a Grizzly mill; I have seen Jet, and think they are pretty good. I have seen Enco, and I am unimpressed.

Ben Shank
04-29-2001, 10:54 PM
If you are going to do any serious gear cutting you really need a dividing head and a horizontal mill, a verticl mill isn't ridgid enuff and is going to be a pain to setup. The arbor in a vertical mill doesn't have any support on the outboard end allowing it to flex. For very light ocassional work you can do it but if you have very many to do a horizontal is the only way to go,,

John Stevenson
04-30-2001, 05:43 PM
<font face="Verdana, Arial" size="2">Originally posted by Ben Shank:
If you are going to do any serious gear cutting you really need a dividing head and a horizontal mill, a verticl mill isn't ridgid enuff and is going to be a pain to setup. The arbor in a vertical mill doesn't have any support on the outboard end allowing it to flex. For very light ocassional work you can do it but if you have very many to do a horizontal is the only way to go,,</font>

Don't forget though that Bridgeport do a horizontal attachment for their verticals.
Chances are it also fits the clones.



------------------
Regards,
John Stevenson,
Nottingham, England

Ben Shank
05-01-2001, 01:03 AM
John,
You could probably buy horizontal over here cheaper than you could buy a Bridgeport attachment, if you could find one. Just my $.02 worth

John Stevenson
05-01-2001, 03:34 AM
Ben,
I was just making a point. I fully agree that a horizontal is far better but I was just trying to point out an alternative bearing in mind that most people have a space problem and may have a vertical mill already.
We have J&L in the UK as you do, they often list these as specials for about 250 UKP, thats about $360. Possibly a lot to spend but taking something like distance and shipping in the US even a free horizontal will possibly cost the same.
Another plus that a lot of people don't realise with the horizontal attachment is that it's actually a right angle drive and will take the standard R8 collets. Very handy if you need to work on the side of large pieces.

------------------
Regards,
John Stevenson,
Nottingham, England

Ben Shank
05-02-2001, 12:04 AM
John,
I agree that the attachment is a nice thing to have and if one is short on space or just wants all the gadgets that came with the mill it's nice to have. But seeing manual horizontals selling at auctions for less than $100.00 us, they are dirt cheap right now.

John Stevenson
05-02-2001, 03:18 AM
<font face="Verdana, Arial" size="2">Originally posted by Ben Shank:
John,
I agree that the attachment is a nice thing to have and if one is short on space or just wants all the gadgets that came with the mill it's nice to have. But seeing manual horizontals selling at auctions for less than $100.00 us, they are dirt cheap right now.</font>

Ben,
AGAIN I'm not stating that one is better than the other - just pointing out alternatives.
Hell if you have the room go for it. They are real work horses. I have a Victoria U2, bed size about the same a Bridgeport. I make toolholders for North Sea oil rig boring tools. This machine puts a 7/8"wide by 7/8" deep slot in the holder in one pass.
As regards cost though work out the full cost. Take the auction price, add the sales percentage, usually 10%, add shipping and add your time off work going to the auction.
Another thing to watch is that horizontals were bought in industry for heavy work and usually show signs of such. If you get a chance get one with a vertical head so you have the best of both worlds but beware not many have a moveable quill so and angle work like boring motorcycle valve seats still require you to set the job at an angle as opposed to setting the head at an angle.

PS. They are that cheap over here that you can't give them away. Peter Forbes over on RCM was trying to give a small Adcock and Shipley away about a month ago. He got no offers and dumped it off on me so I could rob the main ISO 40 spindle out of it for using as a dummy to machine special holders. Once someone else collects the motor it's going in the scrap.

------------------
Regards,
John Stevenson,
Nottingham, England

[This message has been edited by John Stevenson (edited 05-02-2001).]

aero
05-02-2001, 02:56 PM
Thanks for the replies - I will mostly be cutting small steel gears and also plastic gears - very very low quantities (hobby work). I think a vertical mill will work fine for this. My problem is finding a distributor for metric gear cutters that will sell to the home user (non-commercial sale).
Anyone have any experience with the rotary table / dividing head combinations? Grizzly sells one (they sell dividing head add-ons for their rotary table) and for what I need it for (gear cutting as mentioned above above) it seems like it should work fine.

Ben Shank
05-03-2001, 12:34 AM
John,
I don't have room for another machine, but if one comes along, I'm sure it will be crammed into the shop somewhere, I think I'm running an "old folks home" for vintage machines and other "orphans" that need a home!
No, I'm not arguing the vert vs horiz, just wanted to point out that horizontals are cheap rite now, and may be a better deal than a horizontal attachment. I have a B&S horiz'l with vertical atachment and it has no moveable quill in it.

Ron LaDow
05-03-2001, 07:47 PM
Aero,
I just got a 'super spacer' from Enco. I got it because I needed it in the past, didn't have the money, and had to work around it. I have the money just now.
Point is, I haven't used it yet. But it certainly looks like it will suffice for both rotary table work and dividing head work. And it does come with both a chuck and a face plate. A two-fer...

Ken S
05-03-2001, 08:27 PM
I purchased some from Travers Tool. I do not use a dividing head - rather I adapted the excellent method described by Rudy Kouphout in HSM some time back. If you have a "master gear or gears", you do not need a dividing head. I have some pix of the one I made on my website, and have found it to be very useful. see profile for web address.

Ken

SGW
05-04-2001, 07:25 AM
I think for some things you do need a dividing head or rotary table. I don't think there is any way, for instance, to use a "super spacer" to make a 47-tooth gear.

John Stevenson
05-19-2001, 04:06 AM
<font face="Verdana, Arial" size="2">Originally posted by Ben Shank:
If you are going to do any serious gear cutting you really need a dividing head and a horizontal mill, a verticl mill isn't ridgid enuff and is going to be a pain to setup. The arbor in a vertical mill doesn't have any support on the outboard end allowing it to flex. For very light ocassional work you can do it but if you have very many to do a horizontal is the only way to go,,</font>

Ben and list,
Just spotted this is a magazine. Thought I'd share it. Looks very cheap and simple to make.
http://homepage.ntlworld.com/machines/vertical%20support.jpg



------------------
Regards,
John Stevenson,
Nottingham, England

artificer in metal
05-20-2001, 11:05 PM
Depending on the pressure angle of the gears you can find some machine tool places that will sell cutters. Problem with this a set is eight cutters and thats for each size of tooth (diametral pitch) and as far as I know only really available for 14 degree pressure angle (I would want at least 20 and better 25 degree pressure angle). It depends on the material you are cutting the gear in but your solution may be to free hand grind a form tool in a HSS blank(use a gauge to assist you and good luck - its a challenge!).

SGW
05-21-2001, 10:18 AM
While you need a complete set of 8 cutters to cut the full range of teeth of a given diameteral pitch (from about 12 teeth to a straight rack), for any given number of teeth you need only one cutter, which in the MSC catalog seems to be about $27 if memory serves. Unless you're doing a lot of gear cutting, it's highly unlikely you'd need a complete set of cutters.

MSC sells both 14 1/2 and 20 degree pressure angle cutters, I believe.

An alternative is to use a circular approximation cutter, as described by Ivan Law in his book "Gears and Gear Cutting." For any given number of teeth, you can pretty closely approximate the "correct" involute form with a circular segment. In fact, a correct circular segment will be closer than what you get with a gear cutter used at the ends of its range. See the book for details.

John Stevenson
05-21-2001, 10:47 AM
<font face="Verdana, Arial" size="2">Originally posted by SGW:
An alternative is to use a circular approximation cutter, as described by Ivan Law in his book "Gears and Gear Cutting." For any given number of teeth, you can pretty closely approximate the "correct" involute form with a circular segment. In fact, a correct circular segment will be closer than what you get with a gear cutter used at the ends of its range. See the book for details.</font>

Similar article can be read at :-
http://www.metalwebnews.com/howto/gear/gear1.html




------------------
Regards,
John Stevenson,
Nottingham, England

Rob
02-01-2006, 11:23 PM
Where are the $100 horizontal mills at? I don't have the room but Im sure I can find a place for it.

Jim Hubbell
02-02-2006, 02:40 AM
I have made a number of 16 and 32 DP gears which I am now using as spare change gears on my Atlas 10 in lathe. A 6in rotary table and homemade tailstock with homemade mandral does just fine. I got my cutters from VME that advertises in HSM. My Jet Mill/drill does the job.

phil burman
02-02-2006, 03:45 AM
An easy way to make a gear cutter is to cut five or six groves in a bar of silver steel (drill rod) with the form of the equivalent rack for the gear DP you wish to make. This is easy because the rack form has straight flanks. You then gash it longitudinal say 6 times at equally spaced distances around the circumference and this forms a series of cutting teeth. Harden and temper then stone the cutting faces and you are ready to go.

The method of use in a vertical mill is to set the center cutting tooth on the gear blank center height then take say a 5 thou cut across the blank. Index to the next tooth and repeat. Once you have been all the way round add another 5 thou and repeat. Repeat until at full depth. This method is good for light machines as you only nibble away small amounts all the way to full depth. With the normal form cutters sooner or later you have to take a full width cut.

This method worked fine for making gears in silver steel. I then hardened and tempered the gears and cleaned up the gear tooth flanks by running the gears together for a short period with a fine grinding compound. This method produces a very nice matched pair, but possibly not so good for interchangability with other gears.

I’m surprised that this method is not discussed in Ivan Laws excellent book on gears and gear cutting.

Best Regards
Phil

franco
02-02-2006, 05:07 AM
Aero,

I have a 6" Vertex H/V rotary table with the optional dividing kit. I have used it on a mill drill to cut several gears, including some 16DP lathe change wheels. The Vertex tables have a MT socket in the table center, 2MT on the 6" tables or 3MT on the 8" tables. I made an adapter to fit the socket which will accept the 4" 3 jaw chuck or 6" 4 jaw chuck from one of the lathes. Also made a tail stock to suit.

For the gears I have made I ground form tools and used these in a fly cutter. It is slow, but cheap, and the gears turned out OK.

If you go the rotary table way, be aware that it quite possibly will not fit the slots in your mill table, so an adapter plate may be required.

As a matter of interest, I eventually bought the Vertex tailstock to suit the 6" rotary table, and was disappointed with it - the fits and finish on the one I got are not up to the standard of the rotary table itself. I also had to make an adapter plate for it so the height matched the R/T when mounted on its adapter plate. I still tend to use the home made tailstock for this job.

Regards,

franco

phil burman
02-02-2006, 05:32 AM
I forgot to say that one of the big advantages of this gear cutting method is that you only need one cutter for any number of gear teeth for a given DP.

Phil

bpsbtoolman
02-02-2006, 07:52 AM
I have been interested in gear cutting for many years. ( I'm 81 ) and up until a few years ago have been getting gear cutters in ones and twos and auctions for a few bucks each. So I now have most cutters for 14-1/2 PA in DP of 20,18,16, and a few bigger and smaller. Since Boston Gear does not sell 18 PD used on my 9A, I make my own gears on a Sheldon horisontal mill that matches my Sheldon 12" shaper. For my Toolroom Heavy ten I have made a set of 100/127 gears for metric cutting. I bought the Shelon mill from a guy that had it listed as a horizontal mill, not.
I have always been interested in hobbing gears and considering putting a hobbing extension on my H mill. A very interesting event occured a few years ago when the Western Machine Co that made Steptoe shapers in Holand, Michigan a few miles from me sold out after being shut down for many years. They were selling or almost giving away just about everything a large tool company. I saw straight edges 5' and shorter sell for peanuts. Drills , reamers, both square and acme taps. Me and the guys were in hog heaven.My friend and I were picking up castings and steel stock for future projects. My horizontal mill came with a 1" No. 9 B & S arbor. Some of my gear currers were 7/8" bore. They had a pile of these in 7/8" and I bought one for cheap. I told my 2 friends about it and the next day decided to get a couple more, But then the only one left was one without coulars that I bought anyway.
The machines were the only thing they had a bid auction on. I looked at a small gear hobber that used conventional hobs and wondered what it would cost. I even bought a new 16 Dp hob in anticipation. You could call in a bid and a few days later I called and asked what the bid was. The girl said they had none. I jokenly said one dollar and she said $ 100 would do it. My friend said he would loan me his truck to haul it home to my walkout basement shop that has a 2 ton I beam just outside the door.. I gave it some serious thought, but I was pushing 80 years old and thought I should be cutting back. I am still thinking what a deal I missed. Bet you know, water under the bridge.
Anybody got a true story to match mine?
Walt

J Tiers
02-02-2006, 08:17 AM
The difference between the rotary table and the dividing head is two-fold.

One, the dividing is simple with the head, and unless the rotary table emulates the dividing head, you have to count degrees etc. With the DH you simply count full crank turns and move the index arms to keep track of fractional turns.

Two, the rotary table is big, and gets in the way. The dividing head with T/S uses a mandrel to hold the blank clear of itself, and so allows easier access by the cutter.
It also presents the blank in a way that either a horizontal or vertical mill can get the cutter to the blank easily.

The horizontal mill is sturdier, more rigid, and, unless you have a universal mill, always in tram.

The ram-type vertical mill like bridgeport allows doing helical gears, which needs a universal mill for horizontal. But it is otherwise easy to have off tram, or get knocked off tram in the cut, unless it is a heavier type.

For cutters, Victor Machinery has cutters at reasonable prices for about any pitch and angle.

Gears may be non-critical, but if inaccurately made, will be noisy or even prone to lock up.

[This message has been edited by J Tiers (edited 02-02-2006).]

EddyCurr
12-19-2015, 01:46 AM
I've enjoyed good results with some inexpensive gear cutters
purchased from CTC.

.

mattthemuppet
12-19-2015, 02:24 AM
Sorry eddy, should have checked the dates, i think this is the oldest so far. When did this forum start?

Spin Doctor
12-19-2015, 09:55 AM
There is one guy on youtube with the site name of hobbynut. He's has videos on what you might call "approximate hobbing". He makes his own cutters. Basically you have a cutter that is a circular rack. When cutting a gear the cutter cuts more than one toototh at a time. You wind up with a tooth form on the gear with a series of flats instead of an involute curve. Probably they are a little noisy but at low speeds I don't think they would be to bad. One advantage is one cutter does a whole range of teeth from say 20T to 80T or more.

Here's a site that explains the process

http://www.helicron.net/workshop/gearcutting/method/

Man, did this one get dredged up from the depths.

Hopefuldave
12-20-2015, 08:03 AM
Following on from http://www.helicron.net/workshop/gearcutting/method/ If you can make the "almost hob" longer and arrange for the gear to rotate as its mounting (the carriage?) moves it'll cut a proper generated involute form - one method that occurs to me is to have the gear blank rotated by a pitch-diameter pulley and wire(anchored to the tailstock and tensioned by a very strong spring at the headstock end?), this has been done on shapers with a rack-form cutter and generates near-perfect gear profiles given a slow enough rotation of the gear with the shaper table travel. This method may be pretty close to gear shaving...
On a lathe, you'd run into flex of the cutting "hob" and its shaft, a travelling steady would be chewed up pretty quickly (unless it was fitted with roller bearings?) so you'd probably want to make the "hob" fairly sturdy and/or fairly short and re-index the blank and cutter a few times?

Dave H. (the other one)

Mcgyver
12-20-2015, 12:38 PM
There is one guy on youtube with the site name of hobbynut. He's has videos on what you might call "approximate hobbing". He makes his own cutters. Basically you have a cutter that is a circular rack. When cutting a gear the cutter cuts more than one toototh at a time.

I call it faceted hobing and have done it and written on a bunch of times. It produces an ok gear and for non demanding operations is, well, ok.

Really, anything short of generation produces just ok gears....and the HSM heros of generation innovation have to be John Stevenson and Texax Turnado (who i believe is sadly no longer with us)

John for electronically coupling the cutter and work so hobing can be done without complex gear trains....and TT for adapting it to a vertical mill which is so much more common than the universal mill otherwise required

http://bbs.homeshopmachinist.net/threads/45314-Gear-Hobber/page9

Zahnrad Kopf
12-20-2015, 01:44 PM
I call it faceted hobing and have done it and written on a bunch of times. It produces an ok gear and for non demanding operations is, well, ok.
Really, anything short of generation produces just ok gears....and the HSM heros of generation innovation have to be John Stevenson and Texax Turnado (who i believe is sadly no longer with us)
John for electronically coupling the cutter and work so hobing can be done without complex gear trains....and TT for adapting it to a vertical mill which is so much more common than the universal mill otherwise required http://bbs.homeshopmachinist.net/threads/45314-Gear-Hobber/page9

I do not understand the effort for the shortcomings. If one is to accept "less than" considerations, then why expend the effort if something equal to ( or actually better ) is more easily achieved by using Involute Mills? Am I missing some point?

Mcgyver
12-20-2015, 02:03 PM
I do not understand the effort for the shortcomings. If one is to accept "less than" considerations, then why expend the effort if something equal to ( or actually better ) is more easily achieved by using Involute Mills? Am I missing some point?

because different things have different performance demands, you make it to adequately serve its function. For example, the pic below shows the 80dp gears in my cut knurling tool. They operate at hand speeds with almost no load; faceted hobing work perfectly for this and I neither had nor wanted to buy a couple of 80dp involute cutters...and it wasn't required given what they are called on to do. The number of teeth with this method also heavily affects how well the gears turn out - more teeth they get closer to a generated shape. Anything short of generation is a "less than" compromise; it's up the to builder to determine where the cut off points are relative to the performance required, equipment at hand, etc

http://i20.photobucket.com/albums/b201/michael0100/knurl/DSC_3765-large.jpg (http://s20.photobucket.com/user/michael0100/media/knurl/DSC_3765-large.jpg.html)

PS4steam
12-20-2015, 02:10 PM
Hi

I have used this outfit for cutters they have inch and metric:

http://www.msdiscount.com/catalog_list.aspx?cat_id=7112&category_site=STARTOOL&referer=&compidcookieset&sessioncookieset

McMaster has a limited selection.

http://www.mcmaster.com/#gear-cutters/=10bo7ci

I actually could not find any at MSC, tried a number of searches.


Travers has some

http://www.travers.com/gear-cutters/c/297742/?keyword=gear%20cutter#keyword=gear%20cutter


Bob

Hopefuldave
12-22-2015, 04:06 AM
I do not understand the effort for the shortcomings. If one is to accept "less than" considerations, then why expend the effort if something equal to ( or actually better ) is more easily achieved by using Involute Mills? Am I missing some point?

Ah, but involute cutters (assuming the usual set of 7 or 8 to cut 12 teeth to rack) are a compromise too and only cut the correct involute form for the smallest number of teeth for the cutter!

A generated gear produced by a rack-form hob, given enough facets, approaches the proper involute form whatever the number of teeth, if enough facets (i.e. by rotating the blank while advancing along the hob) the difference between the GENERATED gear and the ideal involute form is vanishingly small. With an infinite number of facets (in reality 10 or so upwards counts as infinite here! For example for a 36T 16DP gear the advance would be 0.01963" per degree of blank rotation) the difference is small enough that it's hard to measure with inspection-grade, let alone workshop-grade, measuring equipment. And one rack-form hob will cut any number of teeth and is easily made compared to a set of 8 approximate "involute" cutters...

With a slow feed (and it will seem VERY slow!) of 0.005" / rev and a 6-flute hob, the gear generated will have 24 facets on the faces, the actual deviation from involute form will be in the hundredths of thou" - probably better than an involute cutter at its optimum tooth count?

Dave H. (the other one)

J Tiers
12-22-2015, 09:05 AM
I don't get it......

Milling with cutter is straightforward and simple. It gets an adequate gear. It uses a simple dividing head and process.

Faceted hobbing requires carefully co-ordinated rotation and table movēment. Actually most of the way toward what you do when hobbing, as the movements are the same. Just not on a universal mill, and not continuous. And it will be more passes than you use for milling with a gear cutter.

That's a lot more effort to get a gear that is "somewhat better", or to avoid buying a cutter. Getting it "near-hobbed" quality takes even more effort. And you get to decide how to make the "near hob" so as to get good clearance, so even that is not straightforward.

But, yes, you "can do it". I guess.

Mcgyver
12-22-2015, 09:26 AM
Faceted hobbing requires carefully co-ordinated rotation and table movēment. Actually most of the way toward what you do when hobbing, as the movements are the same. Just not on a universal mill, and not continuous. And it will be more passes than you use for milling with a gear cutter.

not so, you proceed exactly was you would with a involute cutter. Index to a tooth, cut, repeat. There is no rotation of the work other than indexing.

The big step from it or involute cutting to a 'proper' generated gear is the coordinated rotation of the work and cutter which JS and TT have made more accessible, but it still requires building additional tooling. Had I enough gears to do that buying became expensive and they had to be really good, I think i'd pursue hobing with TT's set up

J Tiers
12-22-2015, 11:29 AM
not so, you proceed exactly was you would with a involute cutter. Index to a tooth, cut, repeat. There is no rotation of the work other than indexing.

The big step from it or involute cutting to a 'proper' generated gear is the coordinated rotation of the work and cutter which JS and TT have made more accessible, but it still requires building additional tooling. Had I enough gears to do that buying became expensive and they had to be really good, I think i'd pursue hobing with TT's set up

OH?

Does that not then give a tooth with perhaps 2 facets per side (possibly 3)? And is that not actually worse than a form tool cut gear, even though it has then 2 or three points that are on a true generated curve?

Ant the smaller the gear, the worse the number of facets will be. At least the large gear may touch more cutting sections. A small gear might hit two..

I guess I'd like to see a good large photo of the teeth on a fresh 14 tooth (or so) pinion done with the "near-hobbing" process.

Zahnrad Kopf
12-22-2015, 11:34 AM
Ah, but involute cutters (assuming the usual set of 7 or 8 to cut 12 teeth to rack) are a compromise too and only cut the correct involute form for the smallest number of teeth for the cutter!

Yes... I am familiar. :)


A generated gear produced by a rack-form hob, ...


I think this is one point where things start getting muddled by improper use of words. A Hob is a Hob. Period. By definition, a Hob _IS_ composed of a rack form. Period. End of story. ( I.E., - a rack wrapped around a cylinder at an angle to the axis of the cylinder )

As such, it appears to me that what you and others are describing is NOT a Hob. It is a Form Mill, bearing a single tooth profile and space. Whether that form is repeated or not is another matter, and bears little impact upon its nomenclature.

Continuing...

Now, if I am understanding your description correctly you are indeed describing a Multi Toothed Form Mill that happens to correspond to a Rack Profile. This cutter has no lead component, as well.

Is this correct?


given enough facets, approaches the proper involute form whatever the number of teeth, if enough facets (i.e. by rotating the blank while advancing along the hob) the difference between the GENERATED gear and the ideal involute form is vanishingly small.

Okay... Again, this seems a bit vague with respect to the detail of the description. To be sure, I am not picking nits for the sake of the exercise. I am picking nits because I believe it is important to be specific and correct in the description of something for exactly the reason exhibited right now - the ability to convey an idea or thought that it is easily understood by others, and without causing misunderstanding or ( worse ) misinformation.

So, your descriptive -


... (i.e. by rotating the blank while advancing along the hob) ...

This is the definition of Hobbing.

Can you further clarify your statement? Do you mean to instead say that the blank is rotated before each milling pass? And if so, by how much? And how is this value being calculated by you?



With an infinite number of facets (in reality 10 or so upwards counts as infinite here! For example for a 36T 16DP gear the advance would be 0.01963" per degree of blank rotation) the difference is small enough that it's hard to measure with inspection-grade, let alone workshop-grade, measuring equipment.

I think ( owing to the confusion stemming from your above statements ) that you may be referring to the chordal arc length of a 1 rotation there, but am not sure because while a little close the number is actually different than what I get for such. So, can you clarify that? For a standard 36T, 16DP gear, I come up with a chordal arc length of .001612" at the Pitch Diameter and .001702" at the Outside 9 or Tip ) Diameter.



And one rack-form hob will cut any number of teeth and is easily made compared to a set of 8 approximate "involute" cutters...

With a slow feed (and it will seem VERY slow!) of 0.005" / rev and a 6-flute hob, the gear generated will have 24 facets on the faces, the actual deviation from involute form will be in the hundredths of thou" - probably better than an involute cutter at its optimum tooth count?

I am still not "seeing" what you are trying to describe. Your math seems a bit off to me, and I remain very doubtful that you are experiencing the accuracy levels that you claim to be able to.

I am willing to be completely off base and very wrong on this, though. So, in the interest of learning something new, I will offer to actually qualify your assertions on actual Gear Inspection equipment, calibrated, and able to discern single millionths of Tooth Form Error ( as well as just about every other measure of quality as it pertains to Gear Inspection ).


not so, you proceed exactly was you would with a involute cutter. Index to a tooth, cut, repeat. There is no rotation of the work other than indexing.

Okay, this is very dissimilar to what I read from the explanation that Hopefuldave gave.

So, if I am understanding your description correctly, there is a flat, angle sided tooth ( bearing an included angle matching the Pressure Angle of the gear desired ) being fed from Face to Face of the gear. Once through the blank, the tool is withdrawn or cleared of the work, the work indexed an amount corresponding to one tooth space of the gear's total tooth count. ( I.E., using Hopefuldave's example 36T, 16DP gear - 1/36 or 10 ) Is this correct? Am I understanding correctly?

The big step from it or involute cutting to a 'proper' generated gear is the coordinated rotation of the work and cutter which JS and TT have made more accessible, but it still requires building additional tooling. Had I enough gears to do that buying became expensive and they had to be really good, I think i'd pursue hobing with TT's set up

I am not sure what you are referring to, above there.

Can you guys clarify this stuff for me? As it stands right now, the idea appears dreadfully inaccurate and incorrect. Thanks.

Mcgyver
12-22-2015, 12:44 PM
OH?

Does that not then give a tooth with perhaps 2 facets per side (possibly 3)? And is that not actually worse than a form tool cut gear, even though it has then 2 or three points that are on a true generated curve?

Ant the smaller the gear, the worse the number of facets will be. At least the large gear may touch more cutting sections. A small gear might hit two..

I guess I'd like to see a good large photo of the teeth on a fresh 14 tooth (or so) pinion done with the "near-hobbing" process.

the smaller the number of teeth the closer it moves towards useless.

Mcgyver
12-22-2015, 12:52 PM
Yes
Can you guys clarify this stuff for me?

http://www.helicron.net/workshop/gearcutting/cutting_gears/

on the right hand sidebar, under gear cutter, the pages explain it.


As it stands right now, the idea appears dreadfully inaccurate and incorrect. .

so is an involute cutter, but that doesn't mean you can never make a workable gear without a hobing machine. Whether it is adequate for the task at hand, depends on the task at hand. I gave an example above where it worked perfectly. The technique works better with larger numbers of teeth for sure.

I agree on the sloppy nomenclature, a hob is a hob, and I'm taking some chances calling it faceted hobing however it is kind of what is happening.....but I would never call the cutter a hob or the process hobing because its not

J Tiers
12-22-2015, 12:53 PM
As I understand it.....

Rack circular form tool, near-hob as I called it, form mill as you more correctly say it. Circularly positioned teeth, no lead.

Use as if it was a single toothspace form cutter.

Whatever part of the rack still intersects with the blank (if any) will cut another facet on the blank when blank is indexed by one toothspace. This is stated to be "like hobbing" but with fewer facets. (Hobbing does give what might be called facets, but there are a lot of them).

It seems obvious that a larger blank will have more intersection with the "secondary teeth" of the cutter (ones not in the primary indexed location) than a small tooth count pinion will. And the smaller the toothcount, the farther from the pitch line the "secondary teeth" will cut their facet(s).

It should produce a primarily rack-form tooth on the blank, with a varying amount of secondary facets, in varying locations. One point on each facet will be on the true perfect form of the finished gear, so the accuracy will vary with toothcount.

Yes, it seems likely that the pinions cut will be the least accurate, although they actually require the most accuracy of form.

Now, if this procedure was modified to do fractional toothspace indexing, with proportional movement of the table, a pretty good gear could result. It would approximate with a rotary cutter the "generating" action of a reciprocating gear shaper with a gear shaped cutter, just a lot slower, and probably with fewer facets if the operator wanted to get done sometime this decade.

Zahnrad Kopf
12-22-2015, 01:33 PM
As I understand it.....
Rack circular form tool, near-hob as I called it, form mill as you more correctly say it. Circularly positioned teeth, no lead.
Use as if it was a single toothspace form cutter.
Whatever part of the rack still intersects with the blank (if any) will cut another facet on the blank when blank is indexed by one toothspace. This is stated to be "like hobbing" but with fewer facets. (Hobbing does give what might be called facets, but there are a lot of them).
It seems obvious that a larger blank will have more intersection with the "secondary teeth" of the cutter (ones not in the primary indexed location) than a small tooth count pinion will. And the smaller the toothcount, the farther from the pitch line the "secondary teeth" will cut their facet(s).
It should produce a primarily rack-form tooth on the blank, with a varying amount of secondary facets, in varying locations. One point on each facet will be on the true perfect form of the finished gear, so the accuracy will vary with toothcount.
Yes, it seems likely that the pinions cut will be the least accurate, although they actually require the most accuracy of form.
Now, if this procedure was modified to do fractional toothspace indexing, with proportional movement of the table, a pretty good gear could result. It would approximate with a rotary cutter the "generating" action of a reciprocating gear shaper with a gear shaped cutter, just a lot slower, and probably with fewer facets if the operator wanted to get done sometime this decade.

We are on the same page. Complete agreement.

What has been described is next to useless. But! ... ...


http://www.helicron.net/workshop/gearcutting/cutting_gears/
on the right hand sidebar, under gear cutter, the pages explain it.

Not coming up, here.



so is an involute cutter, but that doesn't mean you can never make a workable gear without a hobing machine.

???

I am not quite sure where you are getting that sentiment. I have not read that anywhere in the thread, until just now. :confused: :confused: :confused:



Whether it is adequate for the task at hand, depends on the task at hand. I gave an example above where it worked perfectly. The technique works better with larger numbers of teeth for sure.

Which is partly why I called it nearly useless. For the effort expended, one is far better off using Involute Mills. The results being much more like a gear than what you are describing and proposing.


I agree on the sloppy nomenclature, a hob is a hob, and I'm taking some chances calling it faceted hobing however it is kind of what is happening.....but I would never call the cutter a hob or the process hobing because its not

( Emphasis added by me )

No. No it is not. Not even close, nor by a long shot. One would be deluding themself to believe so. For so many reasons. I don't think I really need to point them out. ( Or, do I? )

It is a shame, really. For just the slightest bit more effort and consideration, you would actually have something commonly usable, widely suitable, and accurate. It is a shame that Nick Mueller does not come 'round any longer. You would benefit immensely from studying his own efforts.

Think about it some more. You are literally this ->| |<- close to doing it.

Zahnrad Kopf
12-22-2015, 01:45 PM
(Hobbing does give what might be called facets, but there are a lot of them).

I wanted to address this separately so that it did not get lost in the discussion. You raise a great point with that, and one so many people overlook. Hobbing does not create facets. If one actually considers the conjugate action of the two cylinders rolling together it becomes obvious ( ? ) that there is no straight line cutting action. The result is closer to a trochoid and is part of the generative process that is so desirable in consideration of Hobbing.

Back to the "almost" discussion. :)

J Tiers
12-22-2015, 02:39 PM
I am assuming that the whole point of the procedure is to avoid spending north of $300 per DP of gear one might want to cut. The multiple ring form mill CAN be just one part for the whole range, particularly the higher tooth counts. And I think a series of index amounts and table movements might be developed to produce a reasonably good gear, for the expenditure of time only.

That is in line with typical higher class home shop practice. Trade time for money.

The original procedure described, straight indexing one per tooth, sounds particularly evil for any smaller gears. I am not sure just where to call it for "smaller", but somewhere between 135 and 12 there is going to be a rapidly increasing error of form that wants correcting. Otherwise the "gears" so formed will likely just lock up when engaged at their nominal depth.

The good news is that there are fewer teeth per gear in the high error area, so a few more index points per tooth may not be so bad.

Yeah, I didn't want to get involved with the hobbing "artifacts", they have a complex shape... I just said they might be called facets and left it at that. They are technically errors of form, but can be rather low importance. They are visible on hobbed gears, though. Shaping also may create small facet-type artifacts, but generally the angular advance is chosen to minimize them. You have fewer choices once presented with a hob, it has "X" many teeth per rev, and that controls what you get..

Mcgyver
12-22-2015, 02:44 PM
No. No it is not. Not even close, nor by a long shot. One would be deluding themself to believe so. For so many reasons. I don't think I really need to point them out. ( Or, do I? )

It is a shame, really. For just the slightest bit more effort and consideration, you would actually have something commonly usable, widely suitable, and accurate. It is a shame that Nick Mueller does not come 'round any longer. You would benefit immensely from studying his own efforts.

Think about it some more. You are literally this ->| |<- close to doing it.

sigh, its no where near close to hobbing, but it is forming a approximately of faces tangential to the curve you'd get from hobbing unlike involute cutters which cut an approximated curve.

Your 'you're almost there' and 'you're this close' is condescending and mildly offensive - I'm exactly where I intended to be and made gears perfectly adequate for the task at hand in the most expedient manner. Expending more time and energy would be stupid is would buying 80 dp cutters for one use when it was not required.

Plus you couldn't be more wrong about 'almost being there'. A hob is hugely more complex to make as is coordinating/synch'ing the rotation of work and cutter. No, this is a technique that has its place based on results required and effort invested.....if I was trying to get 'there' i'd be there.

A skilled man values having many techniques in his satchel and applies the most appropriate and does not scoffing at ones that aren't perfect. Skill and good workmanship is a matter of knowing what to apply when. Nick would agree on that I'm sure . Your position that this technique has no place is as ridiculous as saying involute cutting is worthless because its less perfect than hobbing.

J Tiers
12-22-2015, 02:48 PM
A change of technique with same tool would do a lot towards giving a good generated form.

I see the technique described as being the main limit.

Form cutters are BEST at the lower end oof the range, just where they NEED to be. The rack form cutter is WORST at the important low end of toothcount, when used as you describe. Not comparable.

But do as I suggested, and you can get as good as you have time for. You will have essentially a Sunderland gear shaper, with a rotating cutter.

Paul Alciatore
12-22-2015, 02:52 PM
Understand:

One set of involute cutters = 7 X $$.cc. Going to be well over $100 for even the least expensive cutters.

One hob = 1 X $$.cc. Going to be around 1/3 or 1/4 the price of the set of involute cutters if at the same or similar quality level.

This is a hobbyist board and cash is often in short supply. So some members are often looking for the least expensive choice, not the best or the least labor intensive.




I do not understand the effort for the shortcomings. If one is to accept "less than" considerations, then why expend the effort if something equal to ( or actually better ) is more easily achieved by using Involute Mills? Am I missing some point?

The Artful Bodger
12-22-2015, 02:53 PM
Many moons ago I showed how plastic gears of apparently involute form could be produced by rolling against a steel master gear and I have been told that a similar system is/was used in manufacture of car transmission gears although in that case the gear blanks were red hot.

I am wondering if it would be practical for a home shopper to cut gears to near final form then finish by heating the teeth (oxy/acetylene ?) while rolling against a master gear?

J Tiers
12-22-2015, 03:00 PM
Understand:

One set of involute cutters = 7 X $$.cc. Going to be over $100 for even the least expensive cutters.

One hob = 1 X $$.cc. Going to be around 1/3 or 1/4 the price of the set of involute cutters if at the same or similar quality level.

This is a hobbyist board and cash is often in short supply. So some members are often looking for the least expensive choice, not the best or the least labor intensive.

Well you can get form cutters for $40 or $50 from some vendors. Travers, is one, Victor had them for a long time also.

Hobs may be available for less than the total, but a true hob needs accessory equipment, at a considerable expense above the cost of the hob, which itself is not chaep and often must be ordered. Not practical for most.

The rack form milling cutter can, properly used, do well if a lot of time is taken.

The plastic gear thing is fine............if you need plastic gears. And only if you need plastic gears.

The Artful Bodger
12-22-2015, 03:09 PM
The plastic gear thing is fine............if you need plastic gears. And only if you need plastic gears.

You may find that just about every gear we have is plastic to some extent or another which may be fortunate as non-plastic gears might tend to loose their teeth! BTW, my comment was to the potential of making steel (still plastic) gears by the pressure forming method.

Zahnrad Kopf
12-22-2015, 03:25 PM
sigh, its no where near close to hobbing, but it is forming a approximately of faces tangential to the curve you'd get from hobbing unlike involute cutters which cut an approximated curve.

And I will respond with the fact that ( as described thus far ) the result of which is a higher number of flats the result of which is almost completely contradictory to what Involute gears are so desirable for. ( I.E., smooth and constant velocity motion transmittal )


Your 'you're almost there' and 'you're this close' is condescending and mildly offensive - I'm exactly where I intended to be and made gears perfectly adequate for the task at hand in the most expedient manner. Expending more time and energy would be stupid is would buying 80 dp cutters for one use when it was not required.

You are welcome to take offense as you see fit. I hope you don't choose to be offended. None of what I wrote was intended to be. I think if you take the time to read what I have written and take it at face value, purely for the text of it and subtract any over tone of assumed intent you will find that I am actually cheering you on.


Plus you couldn't be more wrong about 'almost being there'. A hob is hugely more complex to make as is coordinating/synch'ing the rotation of work and cutter. No, this is a technique that has its place based on results required and effort invested.....if I was trying to get 'there' i'd be there.

Then you have completely misinterpreted my comment's intent. Maybe I was not clear enough. Another clearly got it, though. :)


A skilled man values having many techniques in his satchel and applies the most appropriate and does not scoffing at ones that aren't perfect. Skill and good workmanship is a matter of knowing what to apply when. Nick would agree on that I'm sure . Your position that this technique has no place is as ridiculous as saying involute cutting is worthless because its less perfect than hobbing.

No. I won't belabor the concept. You've taken offense where none was meant and seem to exhibit a lack of care regarding the actual intent. Sorry to have bothered you.


A change of technique with same tool would do a lot towards giving a good generated form.
I see the technique described as being the main limit.
Form cutters are BEST at the lower end oof the range, just where they NEED to be. The rack form cutter is WORST at the important low end of toothcount, when used as you describe. Not comparable.
But do as I suggested, and you can get as good as you have time for. You will have essentially a Sunderland gear shaper, with a rotating cutter.

Ding! Ding! Ding! Winner, winner! Chicken Dinner! Give that man a cee-gar!

Exactly. THIS is what I meant by "You are literally this ->| |<- close to doing it."

Take the already existing Form Mill, make a pass, rotate the gear blank X and raise ( or lower ) the Form Mill the matching distance that is the chordal length of X, repeat X times ( as one sees fit for their own purposes ***) and you will result with a very nicely done, and relatively simple approximation of the Sunderland process.

And you have spent ZERO dollars ( or Quid ) extra for the effort, yet have gained MUCH better gears for all the "trouble". I like to call it the Mueller/Sunderland method. :)

But what do I know? Not like I've ever designed or made any. :p :cool:

Have fun.

*** - It will take a surprisingly small amount of these moves to make a rather nice gear.

And the offer still stands to objectively interrogate any samples you might desire in effort to illustrate empirically.

Paul Alciatore
12-22-2015, 03:33 PM
First, I had to look up "trochoid". Here it is:

https://en.wikipedia.org/wiki/Trochoid

If you visualize the real hobbing process you see that both the gear blank and the cutter are constantly moving. This means that as a tooth of the hob passes the blank, it can only be on the perfect gear surface at one point in that pass. That is one, single, mathematical line. A line with absolutely zero area. All other (parallel) lines generated in that pass by other points on the cutter's tooth will be at a short distance from that actual, perfect gear tooth surface. Then the next tooth of the hob comes around and cuts a similar path at a slight offset from the first one. And it also only touches the perfect gear tooth surface along a single line which is also offset from the first tooth's line. All the area between these perfectly cut lines will be off by a very slight amount. This is what is being referred to as facets in the gear hobbing process. And there will be one facet on each tooth for every row of teeth on the hob. So, if a hob has ten rows of teeth, it will generate ten facets on each tooth.

But the hob will also be advancing along the gear's axis so after one rotation of the gear blank, it will be advanced by some feed distance. And the next rotation of the blank will produce a new set of facets that are axially displaced from the first pass. This displacement will also have an effect on the amount of error in the hobbed gear.

Here I am talking only about proper hobbing, and not about the process called near hobbing in the posts above. That process seems to have a lot more problems with gears of small tooth count. Notice I said small tooth count, not small gears. That process would produce a very nice gear of 0.100" diameter with 100 teeth (OK ridiculous example). It is the tooth count that determines the problems, not the diameter.

True hobbing is an approximate process but that approximation can be very close. The more rows of teeth there are on the hob, the less the errors. Also, the smaller the axial feed the smaller the errors and you, the machinist, can control that factor.




I wanted to address this separately so that it did not get lost in the discussion. You raise a great point with that, and one so many people overlook. Hobbing does not create facets. If one actually considers the conjugate action of the two cylinders rolling together it becomes obvious ( ? ) that there is no straight line cutting action. The result is closer to a trochoid and is part of the generative process that is so desirable in consideration of Hobbing.

Back to the "almost" discussion. :)

Zahnrad Kopf
12-22-2015, 03:34 PM
Understand:
One set of involute cutters = 7 X $$.cc. Going to be well over $100 for even the least expensive cutters.
One hob = 1 X $$.cc. Going to be around 1/3 or 1/4 the price of the set of involute cutters if at the same or similar quality level.
This is a hobbyist board and cash is often in short supply. So some members are often looking for the least expensive choice, not the best or the least labor intensive.

I get that Paul. I honestly do. And I appreciate it. That's why I would rather see someone use such a Form Mill just so very slightly differently, yet able to achieve so much better. Zero additional cost, with improved results orders of magnitude better. Seems a win-win to me.