First attempt at gear cutting

[Steve CrowParticipant @stevecrow46066]

After what seemed like an age acquiring cutters and making arbors, a centering micrometer and other bits I finally got round to cutting some gears.

They are Mod 0.3 bevel gears, and spares, for a differential I’m making.

They still need parting and cleaning up.

There are 6 x 12 tooth mitre gears, and 2 x 40 tooth crown wheels with matching 16 tooth pinions.

It wasn’t entirely a success though. I lost the top of one tooth after turning the rotary table before I’d retracted the tool enough (back centre on photo). Stupid error but that’s why I made spares.

I used the parallel depth method from the Ivan Law book and everything was straightforward once I’d got my head around the maths.

A lengthy process though. The 10 gears I made needed a total of 552 separate cuts. (Each tooth gap is cut 3 times). This is my excuse for making the error – a mixture of tiredness and complacency.

They look fine under magnification and seem to mesh well (even the bit with the half tooth).

I will report back when they are finished and assembled.

Steve

[Steve CrowParticipant @stevecrow46066]

[Jim NicParticipant @jimnic]

They look good Steve. What a shame about the half tooth after all the work that went in to it. I look forward to seeing the assembly.

Jim

[Ian Johnson 1Participant @ianjohnson1]

Nicely done Steve I didn't realise how small they were until I got a five pence piece out of my pocket!

IanJ

[Nigel Graham 2Participant @nigelgraham2]

Fine work! Yes – pity about the damaged one, but that's why one makes spares. It's done in industry too.

One way to avoid the tiredness and complacency trap of repetitive work is to plan not to devote whole sessions to trying to complete the lot. It does mean the machine is set up so unavailable for other tasks for longer, hence "plan", but there are usually plenty of other parts of the project to work on. I.e. make one, perhaps, two of the gears one day, one or two the next, and so on.

I do not claim originality for this, but based it on recollecting an article in ME some years ago. The writer had the task of filing dozens of small, triangular ports in piston-valve liners, having pilot-drilled the holes first. As I remember he recounted getting up earlier in the mornings, and filing two or three ports before going to work. If he worked on the model in the evenings, it was on other components for it.

[HopperParticipant @hopper]

Awesome work.

Yes it always seems to be "the last one" that goes awry like the chipped tooth. Usually tap breaks off in the last hole etc. Nigel has good suggestion on this for sure.

[Howard LewisParticipant @howardlewis46836]

They look pretty god for a fist attempt.

As yet, I've never been brave enough to cut bevel gears, (Despite using Ivan law's book as THE reference ) so am impressed by your results

You can move forward with increased confidence.

If it is any consolation, you are not alone in producing an odd tooth. (founder member of the WHOOPS! club )

Howard

[Roger BestParticipant @rogerbest89007]

Awesome!

my favoured size too – small.

Did you take any shots of the set up Steve?

[Steve CrowParticipant @stevecrow46066]

I didn't take many photos of the set up – I always forget to.

Here is one showing the micrometer for centering the cutter.

And a couple from cutting the crownwheel.

Steve

[Roger BestParticipant @rogerbest89007]

Wow Steve, that is the machine for the job.

ta

R

[Pete RimmerParticipant @peterimmer30576]

That's going in at the deep end Steve. Nice one

[Steve CrowParticipant @stevecrow46066]

It took about 4 hours to set up.

I thought the only chance I had of doing these correctly was to have everything square and spot-on.

First I had rebuild the mill in horizontal configuration, something I hadn't done before. Then everything needed to be clocked on each axis before tightening. Then the angle plate had to be clocked square and the 8mm collet adaptor dead centred on the rotary table with a DTI.

Hours of tapping with a little rubber hammer.

After all that, when I went to centre my cutter I found out that the arbor was 5mm too short! (the angle plate screws fouled the column). So I had to remove the painstakingly clocked headstock, put it back on my lathe, set it over for morse taper turning (more clocking) and make a new, longer arbor.

This set me back the best part of a day so I didn't have much time to do the cutting as I had to give the kitchen table back!

In the end, I did all the cutting in one long session. It took about 6 hours including roll-ups, tea and sandwiches. That still works out one cut every 39 seconds.

As suggested by Nigel above, I would have liked to split it up over a couple of days but it wasn't possible.

I was getting quite tired by the end of it, hence the error.

Steve

[not done it yetParticipant @notdoneityet]

Kitchen table engineering! I used to rebuild car and motorcycle engines on the kitchen table about 50-55 years ago. I waited until Mum went to bed and tidied before she got up the following morning.🙂 Table was about 7 feet long and over 3 feet wide, so I only used one end of it – well covered in layers of newspaper….

They look good, btw. Well done. I can leave a mill set up for a job for several days, sometimes – but it does help to have 2 mills and a lathe to use.

[Dave HalfordParticipant @davehalford22513]

It wasn’t entirely a success though. I lost the top of one tooth after turning the rotary table before I’d retracted the tool enough (back centre on photo). Stupid error but that’s why I made spares.

I used my travel stops to tell me when I'd gone far enough to move the cut safely when doing my bull gear ring.

[Steve CrowParticipant @stevecrow46066]

Posted by Dave Halford on 31/12/2020 14:51:23:

I used my travel stops to tell me when I'd gone far enough to move the cut safely when doing my bull gear ring.

Dave, just after I finished cutting I realised I could have done that. Lesson learned.

Steve

[Author]

Posts