PDA

View Full Version : molded gear teeth



darryl
06-28-2014, 08:52 PM
Spent a bit of time today on my ring gear mold. This is for my solar panel tracker. I'm at the point where I need to finally decide what material to use. When I asked the question several days ago I was considering using glass filled fiberglass resin for the teeth themselves, molding the rest of the half-disc in regular fiberglass matt and cloth.

Today I'm just finishing up making a short test section using body filler, which I'll use to help me determine the profile for the pinion gear that will mate with the ring gear. When that's cured enough to pull from the mold, I'm pretty much ready to mold the ring gear. Now I have to choose the material to use for the teeth.

Options are the original idea, which is long and strong body filler- alternately it could be fiberglass resin with no glass fiber filler, it could be short fiber filled resin, it could be metal-filled resin, or it could be more or less any of those but using epoxy resin instead of polyester resin which is what's typically used in fiberglass layups.

Someone told me today that metal filled body filler is super tough stuff. When I looked at it, I saw that the filler was aluminum. There was no description on the can that said what resin it was, so I would assume polyester. I also saw on one can that the filler material was a combination of glass fiber and kevlar fiber.

The mention of kevlar got me thinking- I had played with the idea of laying a strip of fiberglass cloth into the teeth form, saturating it with resin, then pressing it into the form tooth by tooth using one of the fillers, backing that up with more strips of cloth, etc- but then the working surface of the teeth would be this layer of glass cloth. I have a feeling that this would be a disaster as glass is not what you'd call a wear surface. It would probably end up eating the pinion in short order, which some here have previously suggested. But what of kevlar? Would that make a decent enough tooth surface? I know it's tough and resistant to abrasion. Or- forget the cloth layer and just use one of the fillers. That would certainly be easier.

Just to go back a bit- this half ring gear is going to see the equivalent of one revolution per day, so there's no heat being generated as the teeth mesh. It will be totally exposed to the elements, although as the system works it keeps the solar panel oriented to prevent direct sunlight from ever seeing the gear. There will be no lube- or at least if I do put some on it will be gone after a time and I'm not climbing up there to replace it.

Just fishing for some opinions before I make the choice of material.

RichR
06-28-2014, 09:49 PM
How about making the teeth from a length of stainless threaded rod bent in an arc?

darryl
06-29-2014, 01:41 AM
Good idea, Rich. I've done that before and it works. It would have to be something like 3 tpi in this case though.

At this point the gear mold is complete, only needing a slathering of mold release before being put to use. I'm pretty much committed to using this method now. The test piece I made on it turned out well. It showed me that I had to clean up some flashing and fill some gaps in the mold, which is now done.

I'm going to see if I can refine the panel mount, making kind of a triangulated framework which includes the gear sandwiched centrally within it. This should stiffen the panels frame and lessen the amount of materials required, at the same time making it look somewhat streamlined.

This is a lot of work for one panel, but I'm hoping to get to the point where I can duplicate the mount much more quickly and efficiently, making it reasonable to build up a solar array at some point if I choose to. Most likely the drive system will be made as a slave, so several units can be driven by one brain. This is as much an exercise in creating a simple, reliable working system as anything else- a fun project for me.

As I sit here now I'm thinking about the mound of swarf sitting on the bandsaw. It would be great if it was all aluminum- then I might try making my own metal-filled composite material. But it's everything under the sun- steel, wood, aluminum, plastic, rubber- mixed with cutting fluid and wax. Still thinking that maybe aluminum-filled polyester resin would be my choice for the gear teeth-

The Artful Bodger
06-29-2014, 02:18 AM
This is a post I made on another forum...I think I posted it on this forum too but I cant find it!


I discovered (some time ago) a simple way to make plastic gears.

http://farm5.staticflickr.com/4123/4884491518_b4d0df1f96.jpg (http://www.flickr.com/photos/25239206@N06/4884491518/)
DSCN0196 (http://www.flickr.com/photos/25239206@N06/4884491518/) by aardvark_akubra (http://www.flickr.com/people/25239206@N06/), on Flickr

A plastic blank is mounted on the lathe spindle and a steel gear of the required gear 'family' is mounted on a spindle on the tool post. The size of the steel gear is not important.

The method is really simple, the lathe is started and the plastic blank brought into contact with the steel gear which begins to turn. Applying a little pressure causes a rumbling noise but after a minute or two that subsides. Apparently the friction of the steel gear running over the blank has heated the plastic which becomes softer. More pressure is applied and again held until the rumbling subsides. This sequence is repeated until teeth are formed of the required depth.

That's it, except for trimming the edges of the plastic gear.

A few notes, the number of teeth is determined by the diameter of the plastic blank which equates quite closely to the PCD. Any size steel gear will create a plastic gear with the correct tooth form regardless of the number of teeth on the steel gear or the diameter of the blank. It is important to withdraw the steel gear from contact before stopping the spindle, otherwise the plastic gear will be deformed where it cools against the gear. Not all plastics are suitable but it is not hard to experiment.

http://farm5.staticflickr.com/4143/4883913899_3bb39f38db.jpg (http://www.flickr.com/photos/25239206@N06/4883913899/)
cogs (http://www.flickr.com/photos/25239206@N06/4883913899/) by aardvark_akubra (http://www.flickr.com/people/25239206@N06/), on Flickr


John

darryl
06-29-2014, 03:12 AM
Hmm, interesting method.

Paul Alciatore
06-29-2014, 04:50 AM
If it truly is a tracking mechanism and not a clockworks, then why use gears? I would use a linear drive screw with an Evanut on the end of a lever arm that drives it. I doubt that you need a full 180 degrees of travel, probably more like 120 to 150. Let the servo take care of how far to move it.

The Artful Bodger
06-29-2014, 05:52 AM
This is for a solar panel not a telescope? Not so critical then? My thoughts swing towards a length of bicycle chain and a sprocket of the right size.

If you want to stay with your ring gear you could stretch a length of chain around your wheel and have a sprocket engage it like a pinion, but it would be a bit sloppy (I think).

Weston Bye
06-29-2014, 08:27 AM
Another vote for bicycle chain.

dian
06-29-2014, 09:58 AM
put some dry lube in you composite, like graphite, that you get in the small squirt bottles ewerywhere. and go with epoxy.

darryl
06-29-2014, 05:01 PM
I like the dry lube idea. If I go with molding the pinions instead of cutting them, I might do that. Likely I'll 'paint' the mold with a graphite impregnated epoxy, then fill the rest with straight epoxy. That should give a slippery surface without compromising the strength of the teeth.

I'm going to form a test pinion using aspects of ABodgers method, and I'll see if this is workable. I wanted to use 7 teeth, as that would give me exactly one rotation of the pinion per hour. This would make it easier to time the actuation of the panel, but also result in a weaker system as only one tooth would be in contact at any time. Now I'm thinking of making it 14 teeth, giving me one rotation of the pinion for two hours. I can still time that easily enough.

On the timing issue, it will be a clockwork mechanism. My plan is to start with a crystal controlled time base, then use that to control the rotation of a 24 hour disc. The disc will be marked out for an 11 hour cycle in 15 minute increments, then give a wait time of 2 hours, then 11 more hours of 15 minute increments in reverse, then start over. The panel mount will slave to those signals, and will have 8 points on the drive worm to signal the motor to stop at those points. So every 15 minutes, the motor gets a signal to drive the panel to the next stopping point. After 13 hours the motor gets triggered to reverse and drive back, stopping at each of the 8 points on the worm disc each time it's triggered. It will take about 6 seconds of motor running to move the panel each time. I'm estimating it will take about 2% of the average output from the panel to run the clock and drive the mechanism.

I think it will be prudent to put limit switches inside the drive mechanism. That way the motor cannot overdrive the mount, and it would result in a minimum of electronics in the mount itself. No electrolytic caps in this circuitry is one goal.

darryl
06-29-2014, 07:40 PM
Ok, crunching some numbers- it looks like the pinion will have a pitch diameter of 1.25 inches and major diameter of 1.5 inches, 10 teeth. The ring gear has roughly 3.4 teeth per inch and is about 21 inches diameter. Does that sound doable to anybody? I'm not up on gear tooth design and all the formulas- it's already hurt my brain to come up with this.

I have worked out that I can't use more than 11 teeth and I can't make the part properly with less than 10 teeth. The worm wheel that drives the pinion will now turn at one rev in 90 minutes, which still works out for the timing.

I'm hoping that the pinion will have ok teeth still-

The Artful Bodger
06-29-2014, 08:03 PM
Considering the application you could get by with just one tooth on the pinion!

darryl
06-29-2014, 08:40 PM
one tooth- ok. I presume you mean a kind of intermittent mechanism which advances by some fixed amount as the tooth comes around- can't remember the name of this now, but I like the idea.

darryl
06-29-2014, 10:25 PM
Well I used my setup to mold some teeth using epoxy putty. Basically I made a hub, then pushed some putty onto it, then rolled it around the test gear to form teeth. Turns out 10 teeth would work, but they turn out short, sharp on the ends, and only one tooth pushes at one time. The next tooth in line doesn't have a good angle against its mating tooth until the first tooth has already left its mate. So basically with this small of a pinion I would need more teeth, and what it boils down to is I need to use a larger pinion.

More math- 17 teeth would work. That will be my next test gear. Off to work that out now.

Sun God
06-30-2014, 12:02 AM
Small servo/stepper + Arduino in low power mode + Geneva wheel = job done in a couple of hours.

darryl
06-30-2014, 12:42 AM
Geneva mechanism- that's what I was trying to remember. I looked at that, and it would seem that it needs more of a positive lock before releasing control to the 'actuator arm'. The solar panel will be on a pivot which will be carrying the weight of it, but there will be constant force against the driving mechanism when it's facing east or west, plus considerable wind loading at times. It really does need to be capable of taking a beating- otherwise reliability will suffer.

This will be driven by a servo- something like a seat motor or windshield wiper motor- something like what I see at Princess Auto all the time.

I made another test gear sized for 17 tpi. I'm much happier with the tooth contact now, and the profile is better. I did NOT make the full 17 tooth gear, so I don't know if my pivot spacing is exactly right- that's my next step to arriving at a fully made gear to mold from. Alternatively, I'll grind a form cutter and set up on the mill to index a blank and cut the teeth in a rare return to standard machining practices :)