Results 1 to 16 of 16
09-06-2008, 12:58 AM #1
Set-up question for making a worm gear
My little bandsaw has completely wore out my worm gear and I would like to make anther. I know I could probably buy one from Browning, but would rather take a stab at it myself sometime this winter.
What set-up is needed for cutting this gear? If it were a spur gear, that would be obvious. With the worm gear being a bit helical, how would the dividing head be set on the knee mill? Would tilting the head be all that is needed?
I also have a question about measuring this gear. My worm is cast iron (?) and in fine shape, the worm gear, however is almost gone except for a couple slivers of material on one out side edge. What is the trick for measurement on the worm if I can not figure it out off of the worm gear? Thanks
09-06-2008, 01:46 AM #2
You'll need to figure out the diametral pitch of the gear. This can be ascertained by a measurement of the pitch of the worm thread. Diametral pitches are related to diameters of gears, so that means that the tooth interval repeats in Pi inches. If your worm thread was that long, you would probably get a whole number of thread "teeth" in exactly 3.141". You can estimate with a shorter measurement spanning fewer threads. But it won't work out to even inches in most cases, although I have seen one worm gear from an ancient box that was cut on circular pitch and it had 1/2" circular pitch and the worm thread was 2 tpi.
You can only rough the teeth on the wormwheel with an ordinary milling cutter. The true helical form is generated with a hob. For a power transmission gear, you may not be as fussy, but it takes a lot of trial cuts and hacking to do it with a regular mill minus the hob. I have done this with manual mill in the past.
I used to grind up a simple flycutter resembling an Acme thread tool but ground according to the width on the gear charts I used.
I would swivel the table of the mill to the helix angle of the worm. Take a cut in the middle as for a spur gear. Go all around roughing the teeth this way. Now, offset the cutter by one thread pitch endwise, and raise the knee until the tool is at the same depth as it was before. This simulates the worm rolling out of engagement with the wheel.
You can interpolate this way as many times as you like, but it will still not generate that true helical twist of the tooth. But, then you can blue the gear and roll the worm around it as if it were a spur gear, to rub off the blue and see the contact spots. Then, swivel the universal mill table a bit one way, then the other, and take a light shaving off the high spot.
What you end up with is a facetted tooth. Surprisingly you can make it fit eventually, although you need to know the true center distance between the worm and the wheel so that you can measure across and tell when its fitting deep enough.
That is a cabbage method. The better way is to finish with a hob but you won't want to buy one of those.
You can also use a lathe and a boring bar, or a cnc mill with a rigid tapping cycle to move the bar, and get a true helical path for the tool. However, this is still incomplete generation because the tool tip only occupies one of a large number of possible positions that a multi-tooth hob occupies at any one instant. That is why you have to take 3 or 4 cuts, turning the wheel a 1/4 or 1/2 turn and rerunning the tool in the new position.
Complex? You bet. Makes spending $100 at Browning a bargain, but you won't learn anything from buying a gear.
09-06-2008, 08:41 AM #3
It was common to cut worms on lathes. My '54 20" L&S even has the gear chart for cutting diametral pitch worms.
09-06-2008, 12:00 PM #4
I made this one a few years ago
using the method Hu outlined above.
In retrospect, it would've been easier overall to have made a hob by the method John Stevenson outlined at the time, and set it up between centers in the horizontal mill for hobbing.
I made a fixture for indexing, which I mounted on the turret of a CNC lathe, and a flycutter mounted on a bar held in the chuck. This generated the helical path, but as Hu says, it takes multiple iterations to generate an approximation of the shape that's correct as the worm wheel rolls into and out of contact with the worm. The hob will generate the true shape that my method only approximated, and my experience says it can be done in less time than I spent doing it like I did. Any way you go about it, you'll learn a lot.
09-06-2008, 07:42 PM #5
Here's some pictures of a worm gear with internal buttress thread I made. Did the gear on a manual BP with dividing head and the thread on a CNC. Martin
09-06-2008, 09:35 PM #6
how to gash it and hob it
To do it right you need a hob. Store bought hobs cost a lot. There are a lot of pitches out there, worm sizes, etc. no job shop has it all so here is our workaround for 1 shot gear fabrication in bronze.
You will need a lathe, a bridgeport or equivalent, a dividing head with tailstock, a boring bar that holds a tool bit that will fit the bridgeport, a torch or oven.
To make the hob get a steel you can harden I always go cheap on this cause its a one shot tool. I use 1144 stressproof, easy to cut, gets hard as glass if water quenched from red heat.
The hob will be a shank hob, the shank end fits your bridgeport collet, the worm end will do the work. Turn the worm od .314 x the diametral pitch larger than the existing worm and a inch longer in length than the old worm. Its probably 29 degree single start so grind you up an acme style thread bit to fit the old worm good and chase the threaded section maintaining the root diameter as a tooth thickness control unless you have a set of gear tooth verniers to measure the pitch thickness.
The hob should now look like the old worm on a stick so to speak. Now you have to flute it. put it in an indexer in the bridgeport and set the end mill side on the centerline and cut straight flutes on it, I usually put 3-4 on a 3/4 od worm thread. Remove any burrs but remember these edges are your cutting teeth so NO chamfers just get the hangers off there.
If you used stressproof heat the worm section red and drop straight into a bucket of lukewarm water, NOT COLD WATER, obviously there is a risk here of splashing hot water, shattering steel, etc, etc. Be Cautious if its your first time heat treating. You can always go the high dollar road and use A2 or D2 and air quench if the water quench shakes you up. We just want it hard, how you get there is up to you.
Now for the worm wheel, turn up the blank with provisions for putting it between centers on a pressed in mandrel or built in shank depending on how its made. Get it in the bridgeport, with the indexer set up for the number of teeth. At this stage you could hold the mandrel in a chuck or collet or on centers with a clamped dog.
Take your lathe tool bit from chasing the worm and put it in a boring bar in the mill spindle. Try to get the tip of the tool so it would bore a hole the same size as the hob od, if thats too small of a circle go as small as you can. Tilt the bridgeport head to the helix angle of the worm and gash the teeth as other posts show, 1 pass around, gash depth should be about 1.7-1.9 x diametral pitch to leave stock for the hob.
Now take out the boring bar and put in the shank hob, set your spindle rpm relatively slow 100-200 rpm, put the mill head back vertical. Get the part between centers, no dog, grease on centers, it must turn free but not sloppy. Bring the hob into mesh with the gashed blank with the spindle OFF until there is about .015 slop or clearance. Make sure you start the spindle in the proper direction.
Feed it in slow by hand, there will be a lot going on with the cutter turning and the worm turning and your probably nervous cause its scary looking. The total depth of a standard tooth is like a spur gear 2.157 x dp. Don't be afraid to stop the mill and put the real worm in mesh and measure the center distance, take it out and try it in the saw if you have to, you can always put it back in and shave off some more.
I cut bronze dry or with water soluble coolant if its a big one with more heat. Stress proof steel cutters don't like heat and they are very brittle. Once you learn how to do this a bronze worm gear under 8" od and 1" thick becomes about a 10 hr project if you tinker around with it or about 4 hours if you are in money making mode. I have done and seen done 30-40 of these in my life using this method cause its cheap, effective, and easy. After all laziness is the true mother of invention.
09-06-2008, 10:16 PM #7
worm gear replacement
09-06-2008, 10:51 PM #8
When I think of a worm gear, I think of the drive gear for the above pictured driven gears.
I would interpret Lukes question as John Odor did as I too have a lathe that is set up to cut Diametral Pitch gears, but I realize he is referring to the driven gear.
Are they both referred to as worm gears?
09-06-2008, 11:16 PM #9
09-07-2008, 12:15 AM #10
I have to agree with Glenn. That is a worm wheel, though you may call it a gear.
There are various worms that drive worm wheels, up to double enveloped. The worm wheel itself is nothing spectacular, with a straight worm drive. Not as strong, you are not going to move 100 tons of horizontal mill column with it.
OP has a little saw, so a little wheel, but it did wear out.
It IS, today, going to take a CNC to make a double enveloped worm.
09-07-2008, 09:15 AM #11Are they both referred to as worm gears?
If you are making a hob as suggested, naturally the lead on the hob will need to be in terms of diametral pitch - which requires special threading gear set ups on a manual lathe.
I.E. a hob for making an 8DP worm wheel will need a lead of .3927"
We are talking single start hobs of course.
09-07-2008, 09:34 AM #12
At my age I don't have enough time left to plant an apple tree and wait for it to grow and produce a large quantity of good fruit.
I feel the same about many of the things I do around the house. My lawn mower won't start after setting in the garage all winter and after spending the whole day trying to fix it and finding the carburetor is full of parts that can not be fixed, replacement parts cost almost as much as a new lawn mower, the lawn mower repair shop wants more to fix it than it cost to buy a new mower, so I push the mower out behind the garage and go buy a new lawn mower. Now I have 6 old lawn mowers behind the garage, 4 weed eaters, 2 chain saws and several other things that are not worth fixing, that is MY opnion.
On the other hand if you get enjoyment out of spending your last days on earth fixing things then go for it. I am a firm beleiver that people should do what they like doing best.
09-08-2008, 11:18 PM #13
THANK YOU for all the information guys! I'm still trying to wrap my brain around this one. First off, I need to take some measurements of what's left of my existing gear. I ought to post a pic and show just how little is left. I still cant figure out how it cut in this condition.
I am between mills right now and will be picking up my new one next month. Dads big American Pacemaker is where the mill is, and if I can get some ducks in a row and time at the lathe, I may be able to make the hob if I go that route.
I'll be back later for more questions.
09-09-2008, 08:39 AM #14
09-09-2008, 10:21 AM #15I'm still trying to wrap my brain around this one. First off, I need to take some measurements of what's left of my existing gear.
If OD is mostly gone we can come close with a tooth count and the diameter across two opposite root grooves.
The first method adds two teeth to the tooth count and divides the OD into that number (example: 32 teeth plus two = 34 teeth divided by 2.125" OD = 16 DP)
The second method takes two teeth away from the tooth count and divides the root diameter into that number. (example: 32 teeth less two = 30 teeth divided by 1.875" root dia = 16 DP)
The root dia. will be a little less than nominal since the dedendum is always more than the addendum so there will be clearance, but we are smart enough to allow for that.
11-17-2008, 01:34 PM #16
Has anyone had any experience cutting a steel worm wheel with a hob as described above?
I can see cutting bronze with an unrelieved hob, not sure about doing the same operation in something like 1018 or stressproof.
Redlinx liked this post