complicated gear cutting write up – using taps

[BWParticipant @bw]

This may be of interest to some folk – I know some people have succesfully cut gears with taps without going through all the additional work this chap has undertaken.

I never really thought about a metric tap as being a hob with a 30 degree pressure angle before ………….. I suppose a whitworth tap is a 27.5 degree pressure angle.

http://www.ijeijournal.com/papers/Vol.4-Iss.8/C0481422.pdf

[BWParticipant @bw]

[Brian WoodParticipant @brianwood45127]

BW,

How interesting, thank you for making the link available. I've printed a copy since I really can't digest that sort of information fully by scrolling screens about.

Regards

Brian

[Thor 🇳🇴Participant @thor]

Thanks for the link, it resembles to some degree Joseph Noci's solution using a rotary table, but no stepper motors, electronics or software needed.

Thor

[Joseph Noci 1Participant @josephnoci1]

Interesting way to get motion through 90 degrees, but the windup in the flexi-drive would bother me..Most (inexpensive) rotary tables do not exhibit constant friction all around the 360 degree circle, and so wind up in the flexi shaft will vary over a rotation and this will be reflected in the gear being cut. The Writer does reflect on the wind up issue, but not as exhaustively as I think may need be..

Although it is not a 'specially built hobbing machine', it is made of specially built parts, and works as a hobbing machine does – you need all sorts of pulley ratios to the the job as you would need on a hobber anyway.

At least with the electronic/stepper version ( more or less the same effort to build, I would say) you have 'infinite' ratios at hand, and you can use the table as an automated indexor as well!

BW, your comment :

I know some people have successfully cut gears with taps without going through all the additional work this chap has undertaken.

Do you mean the method where the blank free-wheels and is rotated by the 'screw' doing the cutting ( the tap..)?

If so, the problem with that is that the process is not synchronised and the resultant teeth are often not correctly spaced, etc. I have seen chaps do this when making a worm and wheel to thumb/forefinger rotate the worm driving an azimuth/elevation mechanism on a telescope. It worked fine, but I suspect not for accurate positioning of such a mechanism when, for example, automatically tracking a planetary object.

Horses for courses..

Joe

[Brian WoodParticipant @brianwood45127]

Joseph,

I too have seen these freehand hobbing results and the real problem often is that the gear finally produced is one tooth less than was intended.

The two motions have to be linked, preferably in a rigid relationship, for the system to work as well as it can; I agree with your thinking, but the idea is interesting and it does explain the logic in a simple straight forward way

Regards

Brian

[BWParticipant @bw]

Posted by Joseph Noci 1 on 02/11/2018 11:46:27:

BW, your comment :

I know some people have successfully cut gears with taps without going through all the additional work this chap has undertaken.

Do you mean the method where the blank free-wheels and is rotated by the 'screw' doing the cutting ( the tap..)?

If so, the problem with that is that the process is not synchronised and the resultant teeth are often not correctly spaced, etc. I have seen chaps do this when making a worm and wheel to thumb/forefinger rotate the worm driving an azimuth/elevation mechanism on a telescope. It worked fine, but I suspect not for accurate positioning of such a mechanism when, for example, automatically tracking a planetary object.

Horses for courses..

Joe

Thanks for the clarification Joe – I had assumed that they had got the blank diameter wrong and that caused the wrong number of teeth – I've seen a few posts now where people say they had a few goes before getting the number of teeth correct. Didn't realise it was due to uneven spacing of the cut teeth. Does this mean that even if the correct number of teeth is achieved then they are not necessarily equally spaced ?

If you gash the blank first does that help or is the tap unable to grip on the gashes on the major diameter , ie they are too far apart ?

Bill

[Neil WyattModerator @neilwyatt]

It is possible to achieve accurately spaced worm gears by free hobbing, although it takes care and (usually) experience to get a good result.

The problem is more usually that the number of teeth comes out wrong for several reasons.

A recent example a chap making a 450mm gear with a 16 TPI ended up with six extra teeth. The error per tooth was only of the order of 0.0005 and had no effect on fit, the numbers could be dealt with in software.

A greatly respected telescope maker said that in the day they used to make such large worm wheels using sections cut from from 30m bands of jubilee clip material.

Neil

[Joseph Noci 1Participant @josephnoci1]

Bill, both problems occur, but assuming the number of teeth was calculated correctly, the problem lies in the blank having to be 'pulled' along by the 'thread' of the tap and that is indeterminate, especially at the start of the process. Friction of the blank 'bearing, tap edge keeness, how well the blank runs true, how rigid the blank is on its axis, all mess you around. As Neil said, with care, it may work, but most times you swear a lot and then learn to make two or three blanks so you can throw a few attempts in the bin.

If you gash the teeth first, you have a better chance of success, which if you have no hobbing equipment is a way out – however, you still need a means of indexing the blank to gash the teeth, so may as well add a stepper, some software, etc…

Here a chap makes a worm wheel – from first principles..casts the brass blank, machines it, gashes it, and cuts the final with a hob, not a tap, but the principal remains. The blank is free turning.

**LINK**

Joe

Edited By Joseph Noci 1 on 02/11/2018 14:39:18

[Joseph Noci 1Participant @josephnoci1]

Neil, just a comment on the error you mention (0.005) – Often a gear cut in this fashion has teeth that are not evenly spaced – I would presume that would be a problem in telescope use when the gear is driven for auto-tracking of a celestial object while taking long exposure photos? I suppose it could be mapped out by the smart software in some of these telescope drivers, although not sure how you would reference it..

Sorry, hijacking the 'thread' here – nothing to to with what the initial subject was.

Joe

[Tim StevensParticipant @timstevens64731]

Anyone thinking of using this method might avoid the flexible drive bit by using a (longer) toothed belt with intermediate (toothed or flat) pulleys so that the drive from spindle to handle was direct. It would need a support for the pulleys, of course, but fairly easy to make up.

Cheers, Tim

[Neil WyattModerator @neilwyatt]

Posted by Joseph Noci 1 on 02/11/2018 14:34:36:

Neil, just a comment on the error you mention (0.005) – Often a gear cut in this fashion has teeth that are not evenly spaced – I would presume that would be a problem in telescope use when the gear is driven for auto-tracking of a celestial object while taking long exposure photos? I suppose it could be mapped out by the smart software in some of these telescope drivers, although not sure how you would reference it..

Hi Joseph,

The calculated pitch error is 0.0005 not 0.005, I worked it out as part of a demonstration to the chap who made the gear that very little initial slip is needed to add those extra teeth, apart from the fact that cutting starts on the blank diameter, not the PCD, immediately trying to add two teeth. My experience is that gashing is needed to guarantee the correct number of teeth.

If you look here you will see than approach taken to achieve sub-arc-minute accuracy with free hobbing.

http://www.mini-lathe.org.uk/making-telescope-worm-wheels-gears-mini-lathe.shtml

I can't say that I feel my own free-hobbed gears are of that quality.

In practice active guiding can get errors down to levels far beyond what it is feasible to achieve in making gears anyway. This is because the errors are relatively gradual – the accuracy of the worm has the greatest influence. Also, when imaging below about 30 degrees diffraction effects mean the object won't move exactly at sidereal speed – a subtle manifestation of the way the sun lingers at the horizon as it sets and the effect is at its extreme. Last night I had poor guiding and was only achieving 1.68 arc-seconds RMS, both axes combined, on a good night I can hit 0.6 arc-seconds. Accuracy wise, that's hitting a football pitch in New Zealand from here in the UK – while it's moving.

For unguided high-precision mounts typically a combination of high-accuracy encoders and closed loop servo motors are used, together with a sky model made by 'plate solving' images to correct for the air conditions (I think density and humidity both affect diffraction effects).

I need to go through a periodic error correction routine – that won't improve my accuracy (0.6" is pretty much the limit with my setup) but will make it easier to achieve by mapping the cyclic error in my RA worm gear.

The impact of guiding and stacking of images means that today's amateur scopes a few inches in diameter can produce images that put old film images by the largest old scopes to shame. In fact, they are less affected by the seeing (unsteady air conditions) so large telescopes almost always need adaptive optics to benefit from their greater aperture. Now, of course, some people have adaptive optics on their hobby scopes…

Neil

[Neil WyattModerator @neilwyatt]

Just to add, the 'classic' description was on Chris Heapy's legendary lost website, these days he's gone to the 'dark side' (astronomy).

Neil

[Joseph Noci 1Participant @josephnoci1]

Hi Neil, Sorry, was supposed to be 0.0005 , just a typo…Interesting what you are doing, and also how environmental modelling is playing a role now – Sort of Photo-Shop for telescopes…Adaptive optics and all. Amazing stuff.

What happened to your foray into meteor scatter and the like?

Maybe I should not diverge here..Sorry Chaps.

Your comment :

apart from the fact that cutting starts on the blank diameter, not the PCD.

of course hits the nail on the head. Obvious, once you realise…

Joe

[Michael GilliganParticipant @michaelgilligan61133]

Posted by Neil Wyatt on 02/11/2018 16:15:17:

Just to add, the 'classic' description was on Chris Heapy's legendary lost website …

.

… and is archived, under 'Workshop Projects'  here: **LINK**

http://web.archive.org/web/20050313085821/http://easyweb.easynet.co.uk/~chrish/homepage.htm

MichaelG.

 

 

Edited By Michael Gilligan on 02/11/2018 16:42:45

[Neil WyattModerator @neilwyatt]

Posted by Michael Gilligan on 02/11/2018 16:40:26:

Posted by Neil Wyatt on 02/11/2018 16:15:17:

Just to add, the 'classic' description was on Chris Heapy's legendary lost website …

.

… and is archived, under 'Workshop Projects' here: **LINK**

http://web.archive.org/web/20050313085821/http://easyweb.easynet.co.uk/~chrish/homepage.htm

MichaelG.

This seems to be the key page:

web.archive.org/web/20050313085821/http://easyweb.easynet.co.uk/~chrish/homepage.htm

Neil

[Michael GilliganParticipant @michaelgilligan61133]

"seems to be"

You evidently got caught the same way as me, Neil

Trying to copy the URL for a specific page doesn't work

… That's why I edited my post to include navigation

MichaelG.

.

Edit: Let's try this instead …

http://web.archive.org/web/20050404033016fw_/http://easyweb.easynet.co.uk:80/~chrish/worms.htm

Edited By Michael Gilligan on 02/11/2018 20:01:22

[Neil WyattModerator @neilwyatt]

Well done Michael.

[BWParticipant @bw]

Thanks for the comments and tips and links.

I've only cut a few spur gears and a couple of bevel gears out of delrin, using single point cutters and indexing each tooth individually so am very much an amateur at gear cutting.

When would you use hobs rather than gear cutters and why ?

If you have to gash each tooth you might as well leave the blank on the jig and then run a gear cutter or single point around, although I guess for lots of teeth hobbing is easier ?

Am off to read Ivan Law's book again.

EDIT – I think I get it – if I am going to do a one off job I make a gear cutter or single point tool, if I am going to cut a heap of spur gears for my lathe its worthwhile making a hob and gashing before hobbing ?

Edited By BW on 02/11/2018 22:38:45

[Joseph Noci 1Participant @josephnoci1]

Bill, there are three processes in your mindset, and I think they are becoming slightly mixed up..

First, using a hob ( or a Tap) to cut teeth in a blank that free wheels is one process and, as described , is tricky and inaccurate. Pre-gashing the teeth with an involute or single point cutter will help, but requires proper blank indexing.

The second is using an involute gear cutter, or a suitably profiled single point cutter, with the blank held fast while completing the cut on one tooth, and then indexing the blank to the next tooth and repeating – The same as gashing above, except the tooth is cut fully, not just gashed.

The third under this forum topic is continuous hobbing – Using a helical hob, or a tap. The blank in this case is forced to keep rotational sync with the cutter, related to the 'pitch' of the hob helix, and the desired number of teeth on the blank. The blank must be mounted at the appropriate angle in relation to the hob helix angle as well

Your comment:

if I am going to do a one off job I make a gear cutter or single point tool,

This is the second process above and is fine for a few gears – is also accurate, assuming the cutter profile is accurate, and the indexing is correct.

if I am going to cut a heap of spur gears for my lathe its worthwhile making a hob and gashing before hobbing ?

I guess it depends what a 'heap of gears' is..for 5 or 6 gears , if I really did not wish to go down the road of making a 'proper' hobber, I would cut them using method 2 above. Use a proper involute cutter and the job should not be too taxing. First gashing, then removing the index table and mounting the blank so it freewheels firmly, and fitting the hob/tap and then free-hobbing, etc, seems a heap more work. While you are gashing, you may as well finish the job! Remember that when you use a helical hob, you also have to mount the blank at the helix offset angle to the hob axis, else the teeth will not be axial to the blank..and hat angle must not exist when you are gashing the blank, so all in all a big pain in setup – just use process 2 all the way for a half dozen gears, keep you wits about you so you dont lose tooth/index count, and its done in a jiffy.

I have seen some folk mix hob description terms as well –

I believe, but may be corrected, that a hobbing cutter is normally a helical cutter, ie, it has a 'thread' the pitch of which is related to the DP of the gear being cut. I have seen folk making a cutter that looks more like a stack of involute cutters side by side, and using that to cut a single tooth at a time, indexing to the next tooth as required. I think this profiles the adjacent teeth in a manner similar to a helical hob in continuous hobbing. This cutter seems also to be referred to as a hob, but I think that is not correct? Or maybe a hob is just a cutter..

Some videos on this- Skip close to the end to see what the cutter looks like – the chaps do go on a bit…

**LINK**

**LINK**

Joe

[Ian S CParticipant @iansc]

I have not as yet done so, but I think using a spiral tap would be better than one with straight flutes.

Ian S C

[BWParticipant @bw]

Posted by Joseph Noci 1 on 03/11/2018 07:09:01:

I believe, but may be corrected, that a hobbing cutter is normally a helical cutter, ie, it has a 'thread' the pitch of which is related to the DP of the gear being cut. I have seen folk making a cutter that looks more like a stack of involute cutters side by side, and using that to cut a single tooth at a time, indexing to the next tooth as required. I think this profiles the adjacent teeth in a manner similar to a helical hob in continuous hobbing. This cutter seems also to be referred to as a hob, but I think that is not correct? Or maybe a hob is just a cutter..

Joe

Thanks for the explanations and links Joe.

I believe, but may be corrected, that a hobbing cutter is normally a helical cutter, ie, it has a 'thread' the pitch of which is related to the DP of the gear being cut

That's what I understand too – I have cut a wormscrew in the past to match the 18 DP of my spur gears (just for fun) and I believe that I could make a hob from that wormscrew and it would be able to cut any size gear at 18 DP – saves me having to make/buy several different cutters to cover all of the range of gear sizes at any one given DP.

I have seen folk making a cutter that looks more like a stack of involute cutters side by side

Making that cutter and why it works is documented at the link below, I'd like to try this someday.

Gear Cutting

[Author]

Posts