PDA

View Full Version : Gear OD, how do I cheat?



gary hart
02-05-2009, 12:14 AM
Need a 15 tooth 24 DP spur gear with a 7/16 bore and a 3/32 keyway. The gear is for an old dividing device. When the gear is cut the keyway is too close to bottom of the teeth and easily breaks.

How does one cheat and make a 15 tooth gear that will mate with other 24 DP gears and be a larger diameter or less tooth depth?? Center distance of gears is adjustable and there is no way to make bore or keyway shallower and still work.

Gary

wierdscience
02-05-2009, 12:24 AM
Make the keyway in the gear a half round slot and radius the top half of the key.No corners=no stress risers.

tony ennis
02-05-2009, 12:33 AM
How does one cheat and make a 15 tooth gear that will mate with other 24 DP gears and be a larger diameter or less tooth depth??

Assuming we're talking standard involute spur gears, you can't.

Can you braze/weld/solder/pin/glue on a collar to the 15 tooth gear, and put the keyway in the collar?

gary hart
02-05-2009, 12:45 AM
Can't cheat on the keyway, It has to inner change with the other 23 gears that fit on a keyed shaft.

tony ennis
02-05-2009, 12:56 AM
Is this a new gear, or are you replacing an old one?

Roy Andrews
02-05-2009, 12:57 AM
better metal? only option i see.

torker
02-05-2009, 01:15 AM
The gears that break...was the keyway cut right under the center of the tooth or was it off to one side or the other a bit?

gary hart
02-05-2009, 01:23 AM
The keyway was centered under a tooth gullet so the sharp corners were under the teeth.

Steel used was stress proof.

oldtiffie
02-05-2009, 01:24 AM
Drill the gear and the shaft as an assembly - about mid-way across the gear width - and insert a roll-pin. That should not only do the job quite well but it will eliminate the need for a key and a key-way.

gary hart
02-05-2009, 01:29 AM
It has to be removable and one of the two shafts that it can be used on does not have a removeable key. This is a dividing device that uses gears to get the different divisions.

winchman
02-05-2009, 01:34 AM
Make a gear slightly longer than you need, and shrink a collar around the end of the gear on the unused portion of the teeth.

The collar will reinforce the gear, and it probably won't interfere with anything.

Roger

oldtiffie
02-05-2009, 01:35 AM
Gary,

it would be a lot easier and a lot less guess-work if you were to post some pics to assist.

gary hart
02-05-2009, 01:50 AM
Roger, maybe collar might work. Won't be able to try tomorrow.
Don't pic of broken gear but here is dividing device.
http://img.photobucket.com/albums/0903/ghart3/SB%20gear%20mill%20attachment/P3130526.jpg

Evan
02-05-2009, 04:14 AM
You can cheat slightly, probably enough to make the difference you need. Involute gears have the property of producing the same ratio even with changes from the nominal center distance.

Make the gear the same OD and pitch circle but change the tip profile of the cutter. The way to do this is to modify the cutter so that it still produces the same depth but cuts the bottom of the tooth profile as round. By grinding the outer tip of the cutter with a rounded tip profile it will produce a gear that will still mesh just fine at a slightly increased center distance. It won't take much to strengthen the gear enough to prevent breakage.

It also looks like you can make the gear at least 25 percent wider than needed. That will increase the strength of the gear as well.

If this doesn't seem like a good idea then try making the keyway so the top is actually round. Even though it will remove more material than a squared keyway it will be stronger as long as it is centered under the root of a tooth.

John Stevenson
02-05-2009, 04:14 AM
First thing is file the key down to 1/2 hight and make a shallower keyway in the gear.
The other gears will still fit the shaft.

Next kludge is to make a new gear with the OD equal to a 16 or even 17 tooth gear and then have 15 teeth HOBBED onto it.
It will look weird but it will be geometrically correct and give a stronger tooth form.

http://www.stevenson-engineers.co.uk/files/hob%20indexer8.jpg

Three gears, all the same OD and only the 22 tooth is correct but they will all run in any order.

.

tony ennis
02-05-2009, 09:10 AM
You can cheat slightly

Indeed, no gear is perfect. By pushing the 'imperfections' in a specific direction I am sure you can get a different result.


make a new gear with the OD equal to a 16 or even 17 tooth gear and then have 15 teeth HOBBED onto it.

Perhaps this is the extreme edge of 'pushing imperfections into a certain direction.' I am surprised this is possible. Does it cause excessive wear on the mating teeth? Noisy? Why did you make those gears have the same diameter?

It looks like the 15 tooth gear will go on top of the cone stack. Perhaps he'll have enough wiggle room to make something work out.

Carld
02-05-2009, 10:16 AM
How about brazing the washer under the nut to the gear??? How about not cutting the tooth depth as deep. Most gear sets have a gap at the bottom of the mated teeth. You would have to grind a cutter to cut the gear or modify a gear cutter by taking some of the OD off.

How about doing both of the above for the new gear.

Evan
02-05-2009, 12:27 PM
If you don't cut the tooth to full depth then it won't mesh properly. It will leave the teeth wider than they should be by effectively increasing the pitch diameter. The way to reduce the depth is to modify the cutter as Carl mentions so the tip is either a different shape or even ground off a bit. Then it is used to cut to the recommended depth as if the cutter had not been modified. This produces the correct tooth profile but doesn't cut as deep leaving more meat at the bottom. It will require running with a larger center distance and the gears will run with extra lash but they will mesh.

There is another approach that I haven't bothered to mention that does work. You could, in theory, modify all the other gears to reduce the tooth height by a small amount. This would leave room to increase the amount of material at the root and would make all the gears run with no additional backlash.

Not going to happen? I didn't think so. :rolleyes:

John Stevenson
02-05-2009, 01:15 PM
Perhaps this is the extreme edge of 'pushing imperfections into a certain direction.' I am surprised this is possible. Does it cause excessive wear on the mating teeth? Noisy? Why did you make those gears have the same diameter?


Not only is it possible but it's done all the while. Measure a set of gears up out of an auto transmission ans I'll be very surprised if you find any standard gears.
Cutting less teeth on a larger blank is a sure way to get a stronger tooth.
They even designed a stub tooth
e form to achieve just that where the gears have two numbers like 10/12 DP they are cut to 10DP form but only 12 DP depth.

It's also often done on low count gears to actually improve the mating by cutting out the undercut in low count gears.

Those gears were done just as an example of what a hobber can do with just the one standard hob.

gellfex
02-05-2009, 05:04 PM
Assuming we're talking standard involute spur gears, you can't.

Can you braze/weld/solder/pin/glue on a collar to the 15 tooth gear, and put the keyway in the collar?

I concur with Tony to attach a collar. I've also seen gears (from WM Berg I think) where they simply didn't cut the teeth all the way across, leaving an intact collar. I guess it depends on your cutter diameter.

lane
02-05-2009, 06:26 PM
Make up a 1/2 dozen of them out of 4140 pre heatreat material . Then when one brakes you will have extras. You probably are not going to use that one gear all the time any way . With change gears some hardly get used.

gary hart
02-06-2009, 03:42 PM
"Next kludge is to make a new gear with the OD equal to a 16 or even 17 tooth gear and then have 15 teeth HOBBED onto it.
It will look weird but it will be geometrically correct and give a stronger tooth form."

Checked the 16 tooth gear in this set and it is done as Sir John suggested in quote above, by Hobbing on larger blank. The 16 tooth gear is .070 larger diameter then it "should" be.

Another question. Can I fake a hob with a standard cutter by making three passes?

First pass on axis and less then normal depth of cut.

Do a second cut having cutter axis off set to one side and work rotated so many degrees opposite way.

Cut all teeth again and change for third cutting for other side of teeth.

If this would work, how do i do calculations? Along with my other dumbs am also math dumb.

Gary




http://img.photobucket.com/albums/0903/ghart3/SB%20gear%20mill%20attachment/x16t70thouoverod.jpg

John Stevenson
02-06-2009, 07:29 PM
OK get a coffee it could be long and contains some maths but more pretty pictures than maths.

First off lets get the maths out the way. Given Gary's problem we will address the basic gear which is 15 teeth at 24 DP and for this exercise I'll use 20 degree pressure angle as it's easier to show the profiles.

Using simple gear formulae this gives a pitch circle diameter of 0.625" and an OD of 0.708"

If we were to cut 15 teeth in a 16 blank the PCD would be 0.667" and the OD is 0.750"

This gives us this pretty picture.

http://www.stevenson-engineers.co.uk/files/15_15.BMP

The black gear is the correct 15T wheel and the blue gear is a corrected gear cut on a 16T blank.

Now Gary asked if it was possible to cut a corrected gear by fudging it using 3 cuts a bit like you do bevel gears, I dare say you can as what you are looking for is to produce a gear with the correct radii in the correct place.

Now you won't find this in any gear books but if you think about it there are 8 cutters in a set of any same DP gears. Also given you have probably two close choices either side of the required DP, in this case 20, 22, 24, 26 and 28 that's a choice of 40 cutters that can give you a very close approximation without the need for 3 passes.

Time for pretty picture number 2

http://www.stevenson-engineers.co.uk/files/15_15part.BMP

Just one tooth space of this same gear blown up. Black gear has green construction lines and blue gear has red lines.

First off we draw the two button process on the black gear as we know it's true shape.

Article here on the button method.

http://www.metalwebnews.com/howto/gear/gear1.html

This is calculated for a number 7 cutter which does 14 to 16 teeth.

Now if we switch to our red pen and find the centre of the blue arc we get a centre distance and a button diameter which allows us to draw the red arcs.

Now if we run these figures thru the table in the button article which is based on 1 DP in excel we can play what if.

And what if in this case is a 24DP number 5 cutter which should be for 21 to 25 teeth infed 90 thou.

This will give a very close approximation of a corrected and generated gear without needing a special cutter.

http://www.stevenson-engineers.co.uk/files/15_15mesh.jpg

Picture of both gears in mesh.

gary hart
02-07-2009, 12:01 PM
Thanks John for the good description and all your effort into doing it.

Made on paper the cutters using information from your article on button method of making gear cutters.
http://www.metalwebnews.com/howto/gear/gear1.html

Made the gears on paper as you had just described.
Really appreciate the answer but mostly your explaining method to get there.




http://img.photobucket.com/albums/0903/ghart3/large15t.jpg

Paul Alciatore
02-07-2009, 03:32 PM
Great information on gear making here. But I wonder if you really need a 15 tooth gear. Dividing heads are all about prime numbers. You can usually get any number of divisions that can be represented by multiplying the prime numbers present in the various gears and division plates used.

15 breaks down into the prime factors of 3 and 5. 3 X 5 = 15. These are fairly low and common prime numbers and can easily be present in other gears in your device. For instance, a 30 tooth gear contains both of these in addition to 2 (2 X 3 X 5 = 30). So the 30 tooth gear would probably allow you to do anything that the 15 tooth would unless you need a second 3 or a second 5 or both (225 = 3 X 3 X 5 X 5). But even then, you could just use a second 30 instead of a 15. And it would be a lot stronger.

With the 30 tooth you would probably have to go twice or half as far between divisions, depending on the setup.

Another choice would be to fit a 30 division plate to the head. This would instantly provide you with many additional divisions as you could add other plates with other prime numbers fairly easily. Numbers like 53 or 101 could be added. I have explained the procedure for making such plates with great accuracy in other posts. Use the search function.

I would also wonder what you are doing that is putting so much stress on the gear. Normally when I do indexing, I loosen the locks, rotate to the new position, and lock down to drill or cut. I even have to slow down when approaching the new position to prevent my table from coasting past the proper position - that is how little drag there is on the table. If your table has a lot of drag, perhaps it needs attention. Or if you are stressing it while making a circular cut, then just use a larger gear (16, 18, 20, etc.) or a slower feed. But perhaps you are doing spirals. But even then, you could use a slower feed.

gary hart
02-07-2009, 09:08 PM
Paul, this dividing device has chart on it showing different gear combinations for getting the divisions wanted. It is true that all combinations maybe could be done without the 15 tooth gear. But there is over a dozen listings on the chart that use the 15 tooth gear and that along with having a complete set would be nice if not necessary. The gear set allows making all divisions on the chart with a full turn of the handle, no partial turns or index plate needed.

Your also right in that no real pressure is needed on the gear as it is only used for indexing. But, the bore and keyway was too weak for a standard gear. Nice thing is lesson learned from gear guru Sir John Stevenson.

Gary