Gear hobber (mechanical)

Hi Gustav, I've found some emails from 2003 to a chap called Michael Durkin where we tried to work out the hobber workings and some of the numbers around it. He kindly dug out some of the following pictures for me which he took at the Harrogate show. He also gave me the following explanation of how the differential worked. I'll have a deeper look for the spreadsheet as well, but with the picture and the description it should hopefully be enough to make your own differential. I'll try and dig out the calculations as well as that will get you to a point where you can cut the primes from non-prime gear sets.

"...I have figured out the differential. I noticed that the gear being cut has 36 teeth, so I tried a teeth counting exercise on the photos and tried to add everything up to arrive at 36 teeth on the blank.

I did so in this manner (you need to look at the photos at the same time!):-

Input from the motor drives the big gear behind the one with the differential, (I think they are on co-axial spindles, but not coupled), this has 60 teeth.

I think this gear is coupled to one of the differential wheels, the rear one, facing the one visible on the hob drive shaft. It also drives an idler (don't care about teeth on this) which you can just see 4 teeth of, and then onto a 60 tooth Acetal gear. This latter drives the worm and wheel, which seems to be a 36 tooth job.

The 60 tooth Acetal gear is coupled to a 30 tooth Aluminium gear and via another (Acetal 60 tooth) idler, drives the 30 tooth gear which also forms the differential cage.

I did the "mind game" involved with this differential and if the rear gear is driven at, for example, 100 rpm c.w., then this would drive the visible differential gear at 100 rpm a.c.w. if the cage was static. With the fitted gears, the cage is also driven at 100 rpm c.w., which would drive the visible gear at 200 rpm c.w. if the rear gear were static. The differential acts as a summer and the final speed of the visible gear (and the hob) is the sum of the two resultant input motions, 100 rpm c.w.

Of course, by playing with the ratios of the train between the big gear and the differential cage, you can play with the speed of the hob and the speed of the worm.

The worm acts as a 36 to 1 reducer and would have an input of 100 rpm, so the blank would rotate at 100/36 rpm. This is exactly right for the gear we look to be cutting.

I set this up in an Excel spreadsheet and played with the available change gears in a Myford set (maybe 2 sets for some ratios). It will generate lots of different outputs. If you change the worm gear, this adds even more.

I intend to analyse it further, because I think some interesting possibilities arise. This is because the differential is used to sum two speeds and end up with a third speed which can be prime. For example, if you input 200 rpm and 500 rpm (actually 250 rpm multiplied by 2 in the differential) you get 300 rpm...."

The above quoted bit was from Michael Durkin, following is the images that he used to go with the description above. More soon when I dig out the spreadsheets... Regards, Duncan.

.

Many thanks or sharing this, Duncan

I think that brief quote ^^^ shows why this mechanism should be of wide interest.

MichaelG.

The only thing I have managed to find on this subject so far is here

**LINK**

It seems that a helical gear cutting action is removed by using the none differential method against it to set the gear teeth angle back to zero or close to it.

John

-

Edited By Ajohnw on 29/02/2016 12:07:21

to Duncan Munro:

Many thanks for this long explanation. I must confess that my competence in English language sometimes comes to a border because I learned it fifty to sixty years ago. So I must have more time to understand what's about your chaps explanation. I will repeat reading what you have posted for some times until I will be back.

Hi Gustav, I now have the diagrams and spreadsheet which may be easier for you than the English text? There is a link at **LINK**

This now has a spreadsheet hobber_differential_2016_02_29.xlsx which has diagrams and text explaining how the gears interact, and also the calculations to allow the non-prime gears to be made. Hope this is of some help to you.

Hello, Duncan, Your spreedsheet seems to give a very good explanation. I will study it this evening. Another source with basic knowledge is this:

**LINK**

**LINK**

Many thanks!

Duncan & Gustav

There's some remarkably good reference material coming from what first looked like a simple practical question.

Thank you both.

MichaelG.

Thanks for the spreadsheet Duncan and a splendid bit of work by Michael Durkin.
Please can you correct the 'what if' page cell B11 to "3 Output driven Gear" instead of "4 Output....." to align with the diagram.

Translating the formula I gave earlier Target = ( 2 * differential train ratio) - (input train ratio) becomes

Target =( 2 * worm * output driver / cage) - (worm * output driven/input)

example 127 = (2 * 60 * 41 / 30 ) - (60 * 37 / 60 )

Clever choice of worm = input = twice cage means it reduces easily to depend on the two other gears. It looks like the 'Lathes.co' version uses cage = 40 so an adjustment can be made.

Thanks for spotting that, I've update the website page and the spreadsheet.

Regards,
Duncan

Duncan, another glitch. In 'what if' B10 and B11 need swapping over. driveR and driveN are the wrong way round. When changed the numbers for row 41 & column 37 which meet in 127 then agree with the calculations page. Difficult thing to have checked if you haven't got the machine but if left wrong people might not think to swap the gears and lose total confidence in what is a very clever spreadsheet.

Thanks again for spotting that. I made sure that bit had been double checked, so not sure how I managed to make a mess of it... Spreadsheet and web page should now be corrected.

I bought the CES Plans a couple of months ago and now I try to manufacture the different components of the machine. There is one problem with the feeding of the working head carrying the gear blank which I don't under-stand. The video "little hobbing machine" doesn't show this action.

My question is: How can I bring back the working head to the start position after the first feeding of the gear blank, to begin the next feeding with the condition to find the same cutting groove.

Hello. I have just seen this thread and it may have gone cold. The Lathes UK page on the Jacobs Hobber was from my input and the photos are of my machine. The purpose of the differential add-on is purely to allow the hobbing of spur gears with prime numbers of teeth without first having to have one in the index train. A 127T gear can be made in this way. The maths id fairly simple. If anyone would like to PM me I can send you details as I have written it up.

Edited By John Stevenson on 25/10/2016 11:17:06

Probably best not to post your e-mail address for every one(not in the forum membership) to see, and send you junk mail. Could a moderator sort Ian S C

Message for the moderator :Please remove my e-mail address from my previous post as advised

Thanks, Chris.

To Chris Robinson:

Thank you very much für your post. I'm very interested in the basics of your differential add-on. So I would be pleased to receive your informations. Perhaps you can give me some tips of the design of the add-on too.

For the record, from Neil’s spreadsheet index

“Using the Jacobs Gear Hobbing Machine to Cut Helical Gears Part 1”

MEW 2010 November vol 169 page 36 by Chris Robinson.

Not sure what it covered as I can’t find this issue.

Bazyl, This article, which I wrote was published in MEW in 4 parts. No.169 Nov 2010 pp36-41, No170 Winter 2010 pp 32-35, No172 Jan 2011 pp 44-48. It covered modifications to the Jacobs hobber to cut helical gears. I wrote a further article covering the fitting a differential to the index train, cutting worm gears using plunge feed, cutting worm gears using ACME taps as hobs with tangential feed and more but MEW declined to publish.

If you message me your e-mail address I can send this to you.

This diagram of the kinematics of a commercial universal hobbing machine (DEMAK) might be of interest.

It might not be obvious, but to cater for hob lead and helical gears, the hob axis can be tilted about the axis of gear pair H-I.

Hi All

Do you use Google translate to understand a thread? . Seems to work for me the few times I have tried it. Just a thought for Gustav.

Best wishes to all.

M