cutting spur gears on a mill

[brian jones 11Participant @brianjones11]

Fantastico JB whats not to like?

so make another gear and see how they run together

you can smooth them out by running them through a shallow bath of fine abrasive powder of some sort

Its exactly what I intended to do

Obtw

the ultimate insult to the purists is to use my sliced bolt method, ie take a CS HT bolt , hold it in a vice and use an angle grinder with 2mm blade to make 30 deg slices longitudinally across the thread, rotate at 90 deg and repeat

then you have an approx to a spiral flute cutter

That was to be my last ditch attempt cos spiral flute taps are not easy to come by and are v expensive if large say 1"

So it may not be a nasty knurl at all, just a rough fix for those without expensive complex gear hobbers who need to knock up some gears for lightweight use.

I didnt have time today but your vid has really kicked my butt

I hope other members appreciate this

[Anonymous]

Thanks to JasonB for an interesting experiment.

Not sure if it was a feature of the video but there seemed to be a few noises coming from the cutting that made me wince. I would have expected a cut in aluminium to be fairly benign? Another strange thing is that the 'teeth' on the gear seem to be a smaller included angle than the tap, which can't be so. Could be an optical illusion due to the chamfer?

The method works well for worm wheels but I won't be rushing to use it elsewhere.

Andrew

[DiogenesIIParticipant @diogenesii]

Yes, another

..I guess the squeals are because a tap won't have the same backing-off along the flanks as would be 'nice' in something sticky? (or should I have said 'plastic'..)

Edited By DiogenesII on 13/09/2021 10:47:27

[SillyOldDufferModerator @sillyoldduffer]

Posted by brian jones 11 on 12/09/2021 22:01:21:

Fantastico JB whats not to like?

so make another gear and see how they run together

you can smooth them out by running them through a shallow bath of fine abrasive powder of some sort

Its exactly what I intended to do

Obtw

the ultimate insult to the purists is to use my sliced bolt method

…

So it may not be a nasty knurl at all, just a rough fix for those without expensive complex gear hobbers who need to knock up some gears for lightweight use.

…

I hope other members appreciate this

I get the impression Brian sees himself as a brilliant maverick leaving the rest of us plodders in the dust. Anyone who points out his method has disadvantages must a dimwit conservative failing to recognise progress. In the film, Brian will be played by Mel Gibson as a thrilling success!

Actually, this approach is far from new, and there are good reasons why it's never caught on. The method is unreliable, which much reduces the number of applications for gears made this way:

• Cannot guarantee how many teeth will be be cut on a blank. No good for applications requiring fixed gear ratios.
• The teeth are an unknown profile, neither hypo-cycloid or involute. So whatever the profile is, no good for low friction, power transfer, or low noise applications.
• Difficult to make two gears that will mesh together from fixed centres. Can the same cutter be used to make both gears, or not? If two different cutters are needed, what are the design parameters? I don't think anyone knows.
• Difficult to make a replacement gear for an existing gearbox.

The method works in the sense gears can be bodged, but they aren't good gears. And the method wastes time because it might take several attempts to make a meshing pair, and even more time to grind them in. I've tried it in the past and wasn't impressed: Jason has demonstrated the method in this thread, whilst Brian has yet prove his purist critics wrong by delivering.

I suggest the method isn't popular because it has major theoretical and practical shortcomings compared with repeatable, accurate alternatives such as filing teeth by hand, hobbing, casting, and generating. Rather easy to make Sunderland rack cutters if you own a lathe and mill and they produce predictable involutes. Even easier to buy mass produced gears in standard sizes. Brian's method is interesting and occasionally useful, but in the real world it has many disadvantages and faces hot competition.

In Lethal Weapon 2, Mel Gibson saves his burly partner from a booby trapped toilet by pulling him into a bathtub just before the bomb explodes. Mel's maverick solution works well in the film because the scriptwriter rewards his bravery for our amusement. Unfortunately, maverick tropes cut no ice in the real world where both men would have been hospitalised or worse…

Dave

 

Edited By SillyOldDuffer on 13/09/2021 11:19:40

[brian jones 11Participant @brianjones11]

Well if you havent got a proper hobber you can always go for a dividing plate and cut the individual teeth , its what proper modellers do innit, probably take all morning.

Ive watched that vid several times (is this like machinists prawn?)

would like to see another go with something to blow away the swarf so we can get close to the dirt – fascinating.

Others wading in this mucky subject have reported that they seem to get less teeth than was calculated given OD x tpi

So yours in 3.14" x 19 (1/4bsf) = 60 teeth

how many teeth did you actually get JB?

[Michael GilliganParticipant @michaelgilligan61133]

Posted by brian jones 11 on 13/09/2021 11:27:33:

[…]

probably take all morning.

.

Whereas, this thread has run for four weeks … so far

Not a criticism, just an observation … I’m actually enjoying it.

MichaelG.

[brian jones 11Participant @brianjones11]

Mountebank Millionaire Mel Maverick here with the magnificent flying Myfords, star of the silver suds tray and adored by lady lathe turners

That boy Pike is one stoopid mudder

Q

• Cannot guarantee how many teeth will be be cut on a blank. No good for applications requiring fixed gear ratios.

Doesnt matter, I got 59 when I aimed for 60, but my model will still work

• The teeth are an unknown profile, neither hypo-cycloid or involute. So whatever the profile is, no good for low friction, power transfer, or low noise applications.

Going thyroid was never the intention

• Difficult to make two gears that will mesh together from fixed centres.

Drill your centers afterwards

• Can the same cutter be used to make both gears,

Yes anything from ca 12? teeth upwards whats not to like

• or not? If two different cutters are needed, what are the design parameters? I don't think anyone knows.

parameters shmarameters just suck it and see – like all talented bodgers. FYI they fixed a leak in the ISS space station with chewing gum or blutack?

• Difficult to make a replacement gear for an existing gearbox.

So go the traditional route may you live so long

UQ

Can you hear the silent roar of thunderous hooves all rushing to EB to sources spiral flute taps meh

BFN I must go and see my Doreen and make swarf – how bad can it get

 

OBTW

this approach is far from new, and there are good reasons why it's never caught on

the method never caught on because it wont work with straight flutes, but then came JB and his spiral flute

ROTFLOL

Edited By brian jones 11 on 13/09/2021 12:35:04

[Martin KyteParticipant @martinkyte99762]

Posted by brian jones 11 on 13/09/2021 12:30:19:

this approach is far from new, and there are good reasons why it's never caught on

the method never caught on because it wont work with straight flutes, but then came JB and his spiral flute

ROTFLOL

Edited By brian jones 11 on 13/09/2021 12:35:04

Actually it does work with straight flutes. Used it myself to generate a curved knurling wheel for making brass terminals nuts on my Synchronome. I think I used a 4BA straight flute tap cutting a steel blank. I cannot remember now if it was silver steel or mild and then case hardened. The thumb nuts came out OK. I think I turned a circular profile groove in the blank first.

regards Martin

[John PParticipant @johnp77052]

Posted by Michael Gilligan 13/09/2021 11:40:54

Whereas, this thread has run for four weeks … so far

Not a criticism, just an observation … I’m actually enjoying it.

MichaelG.

Michael
Quite an interesting observation since the third line in BJ's 11
opening post is

"OMG what a painfully slow process as you mill each one"

Four weeks in ,still no gears cut ,for the price of a couple
of gear cutters about £12 each or maybe in the time elapsed
could have made the Eureka device and made some of his own.
A couple of dozen gears could have been made by now by
a painfully slow process.

As time passes us all by the most valuable thing that you have
is time ,probably not the best thing to do is to waste it in some
futile quest.

John

[Martin ConnellyParticipant @martinconnelly55370]

It's a bodge method for a bodge job by a bodger. If that's what they want then there's no point discussing it further.

Martin C

[brian jones 11Participant @brianjones11]

Eureka tool vid

If thats what floats your boat?

 

correction

the method never caught on because it wont work with straight flutes,

It works for worms but I meant NOT for spur gears

Edited By brian jones 11 on 13/09/2021 13:27:56

[JasonBModerator @jasonb]

Andrew,there were a few high pitch (Chatter ? ) noises as the work was being fed into the cutter but that quietened down once the blank had gone full circle.

Also a couple of clunks where I think it jumped a tooth. I did try a staraght flute and that jumped a lot more so the theory that a spiral tooth will work better as there is always some engagement would seem to be true

[Martin KyteParticipant @martinkyte99762]

Posted by JasonB on 13/09/2021 16:16:36:

I did try a staraght flute and that jumped a lot more so the theory that a spiral tooth will work better as there is always some engagement would seem to be true

A fine pitch and a reasonable sized blank works better with straight taps at least in my knurling wheel case.

regards Martin

[JasonBModerator @jasonb]

Yes that would seem logical and being you were cutting a worm wheel type profile there will be more wrap around.

At the opposite end of the scale if you were doing a fairly thin gear say 1/8" or 3mm wide then a large tap would have flutes wide enough to swallow that so could chew things up a bit if it looses mesh

Not sure if I'll get to play with this anymore as out on site a lot the next few weeks as a Client is away and in need to run amok get a lot done before they return and I do like having an empty house to work in.

Edited By JasonB on 13/09/2021 16:47:57

[brian jones 11Participant @brianjones11]

How many teeth did you end up with JB? 59?

What happens if you tilt the work at 45deg to a straight flute tap? – yes you will get a helical gear but does it work?

I am assembling my stuff on Doreen AWS going straight for the sliced bolt

[Anonymous]

Posted by JasonB on 13/09/2021 16:16:36:

……couple of clunks where I think it jumped a tooth……

Specifically it was the clunks that made me wince. Good to know there's a reason for them.

Andrew

[JasonBModerator @jasonb]

Does it mesh yes it does

As I said I have not got time to machine another gear but the quickest way to get two gears when you only have the one is to use a hacksaw! One gear on the machining arbor in a collet chuck and the other half retained but free to rotate on another arbor, you can never have too many odd arbors hanging about.

I'd set the mesh prior to starting the video just adjusting the spindle position until it sounded reasonable. It runs and also transmits quite a bit of power, I was applying reasonable force to the bit of MDF.

 

Tooth count was indeed 59T which is about right for an OD of 1" as the PCD will be a bit less than that.

 

 

Andrew asked about the tooth profile so I took a couple of shots up against a Whit thread gauge. Interesting that the gear teeth don't fill the gauge and even though a 20tpi gauge seemed to fit better there were still spaces. 22 was too small and there is not a 21T pi in my set. Click for larger image

 

 

Edited By JasonB on 13/09/2021 18:39:08

[Pete RimmerParticipant @peterimmer30576]

Posted by JasonB on 13/09/2021 16:16:36:

Andrew,there were a few high pitch (Chatter ? ) noises as the work was being fed into the cutter but that quietened down once the blank had gone full circle.

This can happen with a proper hobber too. Mine is often noisier when starting off a pass but quietens right down as the cutter engagement increases.

[brian jones 11Participant @brianjones11]

Well well Mel the Magnificent confounds his armchair grumps, Stand back from success

I was backward in forthcoming as I didnt have any suitable blanks and ordered from EB

the pix are turned cw cos i forgot to landscape my phone – as you do

These are simple acrylic discs 6mm x 45mm gives circum = 141mm

i got 71 teeth

the bolt was 1/2" unc give 13 tpi – 1.954mm pitch so should give 72 teeth

so you get 1 less

Also note that my teeth are slightly slanted as I didnt bother with the helix angle just now

Note the crude bolt with the slits cut at 45deg, 6 in all to match the bolt head which i sliced off later. As a cutter it worked really well for what it was. I was very surprised. Even under a loup the teeth were respectable

I will have to try it in Aluminium and Brass before the naysayers go ape and spit the dummy

I will also try factoring the OD 71/72 x 45 = 44.37mm

Maybe try a slant of the blank to make a helix

I also located a pair of RH – LH threads might go for a herribone

I did make a video but I dont know yet how to put this on YT as per JB. Its so quiet and unhurried I shall add something symphonic

Edited By brian jones 11 on 13/09/2021 19:30:54

 

well I did manage a vid on YT

https://youtu.be/rcXPWMTVNig

its a lot easier than it used to be, i remember a daunting process which deterred me

Sorry no happy housewife muzak, this is my first YT attempt, lots of mistakes but I hope it shows the method

have ordered some al and br blanks and also the means for my herribone rude attempt. Who knows, will this lead to a rough hyperthyroid crown and pinion gear set. get your pooh sticks ready.

Edited By brian jones 11 on 13/09/2021 20:30:38

[Anonymous]

Posted by JasonB on 13/09/2021 18:37:42:

Andrew asked about the tooth profile so I took a couple of shots up against a Whit thread gauge. Interesting that the gear teeth don't fill the gauge and even though a 20tpi gauge seemed to fit better there were still spaces.

Looking at the MDF there's an idea; who needs milling cutters? Simply push the workpiece into a home made gear and hey presto, job done.

It's interesting that a 19tpi gauge isn't a good fit. It's as if the cutter slips initially leading to a smaller circular pitch than one would expect. I'd surmise that the internal angle of the cut tooth being larger than the included angle of the tap is due to a form of undercutting.

Andrew

[Martin KyteParticipant @martinkyte99762]

So explain to me why it should not be an involute. Regular hobs have triangular teeth and generate involute forms. OK Free hobbing introduces errors but should be ay least an approximation to an involute. Pre gashed blanks will do better and give the correct number of teeth.

regards Martin

[brian jones 11Participant @brianjones11]

Regular hobs have triangular teeth and generate involute forms

I didnt know that?

But look at the method

if i want two gears say 2:1 ratio, stick the blanks of 2:1 OD in your jig and 5 mins later ish you got a result

no fancy hobber linked to spindle ££££, just slice up an HT bolt to suit blank OD/pitch = teeth minus 1

Cheap and cheerful for non ferrous mtl

[Anonymous]

Posted by Martin Kyte on 13/09/2021 20:55:31:

So explain to me why it should not be an involute……..

It probably is an involute. But since the tap was unified thread form the pressure would nominally be 30°. That leads to a more triangular tooth shape. So while the involute is probably there the divergence from a straight line is too small to be visible.

Andrew

[Martin KyteParticipant @martinkyte99762]

I was commenting on tooth form not on the method brian.

That confirms what I thought Andrew. Which leads to another reason why free hobbing with taps is not terribly suitable for making gears as it limits you to quite high pressure angles. Nothing to stop you making your own hob but that makes the process more involved.

regards Martin

[SillyOldDufferModerator @sillyoldduffer]

Posted by brian jones 11 on 13/09/2021 21:06:20:

…

if i want two gears say 2:1 ratio, stick the blanks of 2:1 OD in your jig and 5 mins later ish you got a result

no fancy hobber linked to spindle ££££, just slice up an HT bolt to suit blank OD/pitch = teeth minus 1

Cheap and cheerful for non ferrous mtl

Cheap and nasty rather than cheap and cheerful! Although it produces teeth, Brian's method is unreliable. It's a good way of producing unreliable gears, which most of us don't want at any price!

Look at Jason's example:

The green circled teeth show the free method 'working', but the nearby red circled teeth highlight a serious problem – malformed teeth. Jason's example wouldn't mesh consistently with itself! Not because Jason is unskilled – it's because the method is flawed.

Having made one plastic gear by this method Brian bravely asserts the number of teeth is OD/pitch = teeth – 1. Maybe, maybe not. The number of teeth produced depends on slip during the cutting process, and it's unpredictable. Half teeth are possible.

The claim that the same cutter will produce 2:1 ratios is untested, and engineers really shouldn't jump to conclusions.

Brian seems keen to sell his idea as a practical man versus Armchair thing. It's not! I suggest others try the method and decide for themselves. Bear in mind the flaws only become obvious when many gears have to fit together. Asking several gears at particular spacings to transmit power or step up with low friction will reveal problems galore. No one has to believe me or Brian, because the method can be tested.

Dave

[Author]

Posts