source of largish gears?

I have two largish (15 cm) surplus Pittman motors, which I'm trying to cram into a circular shell 24 cm across in order to power the two
primary wheels (for a differential-drive robot). Obviously I can't just lay them out side by side; I'm going to have to have one motor in the front, and the other in the back, which means connecting to the wheels via gears, chains, or belts.
Of these options, I like gears best and chains least. But now I'm at a loss -- where would I find reasonably-priced gears beefy enough to suit these motors? I'm expecting this to be a fairly heavy robot, which is why I'm glad to have such strong motors. The axle-to-axle span of these gears only needs to be about 4 cm, but it's no good if they're skinny little axles that introduce an obvious weakness compared to the rest of the system -- the gears built into the motor assemblies are all 5 mm thick or more.
So, in brief: I need (for each wheel) a pair of gears with a total radius of 4 cm or so, and a thickness of 5 mm or more. (I'll also need fitting shafts and bearings, but presumably those are easier to find than the gears themselves.) Any suggestions?
Thanks, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Hey Joe. The best source I have found so far is Stock Drive Products, found at: https://sdp-si.com/eStore/ . They're the *only* place I have managed to find both matching timing pulleys and belts so far.
McMaster-Carr is also a good place to look & get educated, if nothing else: http://www.mcmaster.com/. So is Stock Drive.
Hope that helps ! JCD
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Pogo wrote:

SDP is more for small stuff. As you move up in size, you go to Boston Gear (www.bostongear.com).
Take their "Gearology" on line course:
http://www.bostongear.com/pdf/gearology/all_gearology-chapters.pdf
If you take the quizzes and pass, they send you a "Gearology" certificate.
                    John Nagle
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Thanks for the tips, guys. Boston Gear looks like they may have some nice stuff, but geez, I've never seen a company make it so difficult even to find out exactly what products they have. Fortunately there's a local distributor nearby -- I'll go in today and look at a catalog.
Best, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Joe Strout wrote:

This is where a good local surplus outfit comes in handy. Most areas have at least one, but they don't always stand out. You gotta go looking for them.
Failing that, check out the surplus mail order like AmSci, H&R, C&H, etc. I like to buy assortments, if they're cheap enough, for the junk box. I've never had to buy a new gear from SDP, Boston, or their ilk.
Lastly, venture into ye good olde hobby store, and check out the replacement gears for electric R/C airplanes and cars (like the Traxxis E-Maxx). These are going to be molded nylon, but well made. I use a Traxxis replacement gear that retails for about $3 in stores in my 360 turret: product:
http://www.budgetrobotics.com/shop/?shop=1&cat7
They do have some with larger diameter, but the thickness is about what you're looking for.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Good tips, thanks.

That's a neat product (and a neat idea). I'm starting to cool on the idea of constructing my own gear train, though. I'm just not all that mechanically adept -- the theory is easy, but the practice is hard for me. If I can find a decent gearmotor (something strong with about 60 RPM of speed) that will fit into my case, I'll probably just buy a couple of those instead.
On the same page as your nifty turret, I see a box gearmotor that looks promising -- but it's hard to justify even a $6 purchase without any idea how much torque it'll have. Couldn't you rig up something where the motor pulls a spring until it stalls, and then you just read off the torque from the spring length (having cleverly calibrated this beforehand)?
Best, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Joe Strout wrote:

For surplus motors that I don't have many of it's not really worth it to build the rig. Static torque isn't all that worthwhile of a measure anyway. For gearmotors it really needs to be done with a dynamometer. Buy me one and I'll be happy to take the measurement for you!
You can guestimate the torque, BTW. One implicitly knows it uses a small PM motor (Mabuchi FA-130 or equivalent) that operates in the 7-9K RPM range; you know the final output speed; and you know the current draw at that speed. I can do the guestimate, but I can't prove it, so I don't publish it. I leave that as an exercise to the reader, as they say.
The other sources I listed typically have a good number of DC gearmotors to chose from. As they are in the business of selling these kinds of things they often have specs.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Heh. I had to go look up "dynamometer." Thanks for introducing me to something new today!

I think I'm still missing some critical formula. One that sounds very handy indeed! Let me see if I can infer the general principle: the current being drawn is a combination of speed, torque, and waste (noise and heat). For an estimate we probably assume that waste is either negligible, or factor it in as some reasonable constant. But I still don't see how to relate current, speed, and torque.
Just taking a wild stab, I'd expect current to be proportional to both, which suggests I = Torque*Speed. Unit check: (N m)*(rotations/sec) gives, er, Watts, not Amps. So the units don't work out.
I'm making an effort here, but I'm still a clueless newbie... can you give me a hint?

Yes, I'm finding that about half the places do, and half of them don't. All Electronics, in particular, has a nice selection, but no torque ratings on any of them. This makes it very hard for me to know what I'm getting for my money. Should I generally assume that for the same RPM, a higher current draw means a stronger motor? Or is that just going to get me a noisy, hot-running one?
I don't need an actual number, since I haven't any idea how much torque is required for my bot anyway (it's modular, and modules will come and go frequently, changing its weight from day to day). I just know that stronger is better, and I have only so-much money to spend, so I just need a way to compare two motors of similar cost and RPM to know which is the better value. Is there any trick you guys use when torque isn't listed?
Thanks, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Here's a case in point. Check out this Faulhaber 141:1 gearhead:
http://www.goldmine-elec-products.com/prodinfo.asp?number=G16279
This is described as "features high torque." But it's only drawing 20 mA at an RPM of 80. That's really low current -- compare it to Gordon's DCM-210 gem, with a no-load RPM of 100 and a current draw of 150 mA:
http://www.budgetrobotics.com/shop/?shop=1&cartB0202&cat6&
...or this Jameco motor, with a similar gear ratio and RPM, but a current draw of 630 mA:
<http://www.jameco.com/webapp/wcs/stores/servlet/ProductDisplay?langId=-1 &storeId001&catalogId001&productId@1007>
Both Gordon's and Jameco's models are half the price of the "high torque" one at Electronic Goldmine, and draw a lot (or in the Jameco case, a LOT) more current at similar voltage and RPM. What should a poor newbie make of this? Is the "high torque" description of the Faulhaber mere marketing puffery, and most likely these other two motors have substantially more? Or is it twice as expensive because it somehow outperforms the other motors while drawing a lot less current?
Gordon, I'm not trying to put you on the spot -- I love your stuff (have shopped there in the past and will do so again), and I wouldn't ask or expect you to badmouth the competition. But maybe you (or anyone else) can speak in general terms... how would YOU go about comparing these motors, which look similar to me in all respects except current draw?
Thanks, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Joe Strout wrote:

That's the no-load current. What you want are numbers like the stall current or a graph of current vs. torque.
That Faulhaber unit is a modestly-priced motor with a planetary gearhead, so it doesn't need much current to run with no load. Here's its data sheet:
http://www.faulhaber-group.com/uploadpk/e_1524SR_DFF.pdf
Rated no-load current is 21mA. Full load current is 630mA. Full load torque (at the motor shaft, not the gearhead) is 2.5 mNm, at 10,000 RPM. With a 141:1 gearbox, at the output shaft expect 352 mNm at 71 RPM, less frictional losses of maybe 35% in the gear train.    
The gearhead is a separate component bolted onto the motor. Plastic gears, incidentally; this is a light duty unit and rated only to 5,000 RPM (input) and 150 mNm torque. So the gearbox is weaker than the motor here.
So that's your answer. That drive is good for 35 RPM outputting 100 mNm torque continuous, about twice that intermittently.
100 mNm means 0.1 Newton-meters, or about 14 inch/oz. Which means that if you put a 1" radius wheel on the thing and wrapped a cord around it, it could lift 14 oz, a little over a pound.
                    John Nagle
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Joe Strout wrote:

Higher quality motors are more efficient, sometimes significantly so. Faulhaber motors are higher quality. If you can afford those, get them. (Of course, know they're still surplus. New Faulhabers cost $100+.) Another advantage is that the manufacturer of these typically provide torque ratings, sometimes right on the case of the motor.
The point is if you know a little about the basic motor used in a gearbox, you can at least approximate the output torque. Torque/current curves are predictable. Less predictable is friction loss in plastic gearings, but these can be approximated. I was being a bit faceteous about "knowing" the motors in the box gearmotors were Mabuchi FA-130s or equivalents; if you know your toy motors, you know this motor simply by appearance. In any case, you can take it as gospel that the motor is an FA-130 or reasonable equivalent.
Because the FA-130 is probably the second most commonly used small PM motor in the known universe for toy products you can often go by empirical comparison, and/or very basic calculations. If the no-load speed of the FA-130 is about a conservative 9000 rpm, and the no-load rpm of the gearbox is about 100 rpm (itself an approximation; I didn't use a tachometer), that's a 90:1 ratio. The FA-130 has a stall torque of 36 g-cm at 3 volts. Complicating things is that I tested the motor at 4.5 volts, which is a more common voltage when using a simple H-bridge like the L293; the toy these came from operated at 6 volts. You can apply some simple math as you calculate values at the intended voltage (and current) you will be using.
In the end the *hobby* of robotics means the adventure of experimentation, especially as it relates to surplus.
-- Gordon
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Gordon McComb wrote:

Those are low end Faulhaber motors. They have bronze sleeve bearings and plastic gears. They're intended for light load applications, even though they have largeish bevel gears stuck on the output end.
                John Nagle
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
Joe Strout wrote:

You might consider toothed belts or timing belts. They are more tolerant of inaccuracies on spacing (get a longer or shorter belt) they are quieter than gears and you don't need to pick a gear that supplies the mechanical spacing dictated by the rest of your design, just get a different belt.
For bearings, you might try Enco: http://www.use-enco.com They have ball and roller bearings at pretty reasonable prices. They also have a lot of tools.
I built a differential drive bot with one motor in front of the shafts and one behind the shafts. In addition to the gearmotors, I had encoders on the back of each motor so they were about 4cm longer. It worked out really well. I used toothed belts from SDP for the drive.
Good Luck, Bob
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Thanks, that's some good insight. I'll give it some thought.
Best, - Joe
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

why not look on ebay?
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload
id look on ebay ifin i were you
Add pictures here
<% if( /^image/.test(type) ){ %>
<% } %>
<%-name%>
Add image file
Upload

Polytechforum.com is a website by engineers for engineers. It is not affiliated with any of manufacturers or vendors discussed here. All logos and trade names are the property of their respective owners.