Engaging Gears Perpendicular to Shaft Direction?

Many teams, most with reasonable success, have made shifting transmissions by engaging gearsets by sliding gears along an axle with the proper distance between the gears for the correct mesh. This is shown as a top view on my fast sketch below.

What I’m wondering is if there have been any teams that have attempted, with or without success, to engage gear sets in a direction perpendicular to that, as shown as a side view in the sketch below.


  1. I am aware this is more difficult
  2. I am aware that you need to be careful about backlash and proper spacing
  3. I am aware that you run the risk of stripping teeth due to the possibillity of teeth prematuring engaging on the outter-most edges instead of at the pitch diameter.

All that aside, has it been done?

Thanks in advance.


Team 469, last years 2nd national champion had a shifter that worked like what you are describing. I got to drive it around at IRI and from what I experianced it worked out very well for them. You could ask them about it because it probably required a lot of work and math.

I dont see why this would be any more beneficial than the “traditional” method. You would have to have your shaft and bearing in some sort of slot so that the gear could be engaged. It seems quite a bit more complex for no obvious benefit. Maybe there is one that I just dont see, but as cool as this seems, It doesnt really look too practical to me.


Well, it all depends on application.

I agree a sloted shaft and gear wouldn’t be pretty difficult and wouldn’t save any complexity. That’s why I’ve got a gear moving on a pivot. Just looking for feedback on this. Thanks!


i’m sure that some where in the CD gallery (2001, i thought), there is a picture of a robot that uses a ‘swing’ shifter that works like the one you described. I would that this sort of shifter might be better than a ‘sliding’ shifter, because the distance that the shifting mechanism would have to move a very short distance to engage the gear’s teeth. of course, this could also work against you, for example if you are a little off on your calculations, you might end up engaging the gears for both high and low speed at the same time.

one drawback for this idea: when you have gears engaging you want the distance between the centers to be fairly precise and consistant

the gears tend to want to push each other apart

when you have a traditional gearbox, the shafts are held in a plate on bearings - they aint going anywhere

but with your idea, the mechanism that moves into place will have to be precise, repeatable, and in effect it will need to lock-in, so the gears dont push apart under heavy load.

There is a common debug problem with FIRST robots - we often get our bots working great on the table, with the wheels spinning in the air

then when we put in on the floor, and try to push something heavy, gears strip, shafts snap, motors cook… Its hard to anticipate the secondary forces and effects that take place under a heavy load

other than that, I think if you only shift at a standstill, or when the motors are not loaded up, it should work.

This is where I’m going to need a mechanical engineer with a little more experience than me…

  1. With how much force will they want to push each other apart? I assume that this actually varies… so I guess I’m looking for the maximum force. I assume this is some ratio of the torque.

and the other obvious question is:

  1. In which direction is this force applied? As for direction, I assume a significant amount of the force is pushing them directly tangent to where the teeth mesh, but I think this is this too much of a simplification. Is the direction actually related to the gears’ pressure angle? What about with mutiple teeth engaging? This seems like it could develop into a rather complex problem…

I’ll guess it’s something like (T / (D/2))*SIN¶ is the force pushing them apart in the plane of the shaft centers… but I’d really appreciate it if someone had something better than a guess. :slight_smile:

Thanks in advance,


To my knowledge, many teams (including #469, #60, and #67) have used this type shifting mechanism with great success.

If my memory is correct, teams #469 and #60 both used very wide gears that they slid an idler gear and its shaft axially in order to engage high and low gears.

In 2001, Team #67 used a hex drive shaft and slid the gears on this hex with a motorcycle gearbox style shifting fork.

Anyway, I don’t see why folks would argue that this method would be more complex than using so “shifting dogs” to engage gears.

It can be made to work either way.

For those who don’t remember, this is the machine that I first recall seeing this type shifter used on (Team 60’s 2001 robot):

Joe J.

I have often wondered this, although never mentioned it for the reasons you listed above. I have a toy, which uses some demented form of miter gears in a shifting transmission( i think they call them crown gears. Legos have them too.) Anyway it looks like this:

Please excuse the paint art. This design is probaby impractical for a number of reasons, but none the less i think its interesting. It inspired this:
Its probably still impractical but whatever.

Errr, I feel really stupid now. I looked at the drawings and for some reson assumed that you would be pushing the teeth of the gears together, face to face, rather than the diamter, if that makes any sense. This is basically the same system as Team 116 used too.


Actually… I think this is what I want to do?

A repost of that picture with some text.

I know you can slide the gear along a shaft into a position.


I want to move the gear and its shaft in the plane of gear rotation, as shown in the bottom sketch.

Feel free to make comments, apparently I wasn’t too clear the first time around…



The force between the gears acts along the line of pressure. This line is determined by the pressure angle.

Therefore, you are correct with your formula.



Depending upon the specific circumstances when shifting, and if you are planning to be able to shift “on the fly” then I believe you will have a couple of significant concerns with this type of design. <disclaimer>I am NOT a gearbox expert - engineers with more gearbox/transmission experience are welcomed/encouraged to coment/correct my statements</disclaimer>

  1. Stripping the gear teeth could be a huge problem. When the gears first begin to engage (while moving) there will be a significant load concentrated across a very small shear area as just the edges of the teeth of both gears are in contact. At this instant there is also a much larger than usual (about 2X) moment arm which will cause increased bending stresses at the root of the tooth.

  2. It will be very difficult to maximize your efficiency while allowing gear shifting in the plane you suggest. A few thousandths of an inch change in center distance can make the difference between gears binding if too tight, or having too much slop/backlash if too far apart.

OK experts … fire away and educate me …

Yes, you could do this, it would take time and in my opinion pretty spiffy enginering. But there would be some issues like grinding. My team had the CIM motors geared down on a the bot we used 2 years agot. Even though the gears were in constant contact, there was a very small bit of play in the distance between two of the gears, this caused the gear to become flat over time. So you would have a big issue with grinding gears in this idea. There is the possibility of binding, too much pressure pulling that gear into the powered gear could cause problems. You also must think about the movement you are creating. You could close out that entire gearbox, so that nothing could get into it, but you probably won’t. Having this movement in competition could be a serious problem. If your opponent has an arm or some object protruding out of the bot, that extremety could get between the gears, and you could not shift. You are also creating a nightmare in the repair department. This design, in my opinion, could cause you to spend more money than needed, and cause you to have to repair parts more often.

A tip I use when working on practicle gear box designs is “when designing, draw something would find easy to build and repair, draw something that will accomplish your goal quickly, draw something that will that doesn’t have more parts than needed, and draw something that will acomplish your goal in a simple method.”

I summerized some of you concerns in bullets. I am aware of all of those, I think I made a pretty clear in my first post:

My response is going to sound grouchy, and I’m definitely not:
Before you start stating problems with a “design”, make sure you’re actually looking at a design! :slight_smile:

(All of those bulleted concerns, especially the last three extrapolations, were based on a 5 second MS Paint sketch with 4 circles and an arrow. It might have been jumping the gun a bit…)

However, I do appreciate the concern, Mr. Ivey. I know your intentions were in the right place. I’ve just seen this happen in other posts too, by many other people. It’s just something I wanted to make everyone aware of to make the CD forums an even better place!

To stick on topic…

Has sliding gears in their plane of rotation in a gear box been successfully done?


Yes, I think that it has been done. Team 393 (yet another Indiana team… wooo!), did it in 2002. They did it with sprockets. I don’t know why, but maybe it was for easier engagement. I do recall that this system was noisy and the sprockets wore down. Paul Osborne or Caleb Osborne (father and son) are the guys to ask on that team. Libby Ritchie is the team leader.

Here is a link to their site:
Full Metal Jackets

I did not find any pictures of that 'bot, but I am sure that they are “out there” (no time now to look).

Good luck,

you mean meshing two sprockets together like gears, or moving a chain?
Also has anybody ever considered/tried something like a bicycle shifter?

being a mountain bike enthusiast, i’ve put a little bit of thought into the subject, and have come away with a the main impression that while it can be done, it will have a few limitations, most of them do to the fact that a bike’s shifting system is made to run in only one direction: forward. because of this, if you pedal backwards the chain won’t shift, and if the derailer is in bad shape (or poorly made) the chain will come off the sprockets. in addition, the shifter would be a strictly on-the-fly shifter. it would be impossible for it to shift if the robot is stationary, or moving too slowly. another thing that may give a FIRST team trouble is the fact that most high-end mountain bikes use specially made sprockets that help the chain climb from one sprocket to the next. these sprockets would probably have to be made from scratch.

http://travel.howstuffworks.com/mountain-bike.htm <–some pretty good stuff on mountain bike design

As a matter of fact, we on 188 tried this very thing last year. It certainly looked good on paper, but we ran into serious issues with the gears grinding, and the centre distance of the gears being extremely susceptible to changing. One moment the gears would be in mesh, the next, the forces acting on the gear (due to rotation) would push the gears apart, causing a loss of contact. To combat this, we tried pushing harder on the gears, so that the gears would be “happier” while running. The problem with that was that the gearbox was now too tight to accelerate properly–the gears would just jam. (Then all of a sudden, as we tried to carefully adjust the spacing, they’d move out of alignment, putting us into neutral once more. At least the motors weren’t getting damaged in neutral :rolleyes:.)

With regard to the jamming issues–don’t even think of overpowering it, and then settling down to a happy equilibrium once the robot gets up to speed. First of all, it’s a waste of power when you need it the most (i.e. acceleration from a standing start, or in a pushing match), and second of all, our robot last year had 6-motor drive (Bosch, F-P and CIM), and still couldn’t overcome the jamming issues with much finesse. It comes down to the fact that your centre distance has to be within a few thousandths of an inch of the optimum spacing (quoted centre distance plus 0.003" to 0.008" is a good bet for the optimum value).

If you do want to try it, keep your tolerances very, very tight (there’s no room for error here). Also, we used 20-pitch steel and cast-iron gears–you might want to try these, since despite the unusual forces exerted on the gears, they held up admirably.

Lastly, design the gearbox so that it can be locked into one gear (high or low) if all hell breaks loose. We made some significant last-minute changes to remove the shifter stage and set it into high gear, when we decided that it was too much trouble. It was good enough to come second at the Cdn. Regional and win W. Mich., even without a low gear…

I agree with that one. While we did not use a shifting mechanism last year (we had some cool ideas, but just ran out of time), we have successfully stripped a set of bevel gears during the Canadian Regional so badly, you could barely tell these were gears once :smiley: There was too much torque on the shafts, and we did not support them well enough. This could definitely be a problem with the mechanism you are describing. Remember, no amount of testing you do will prepare you for the actual competition, since no one can force themselves to put a robot they built with their own hands through some of the stuff that happens to it during the actual competition :smiley: