Extending a cantilevered drive shaft.

Here’s a sticky situation that may or may not have an elegant solution.

I’m working to development a prototype drivetrain that my team will be building over the next two months with the hope that we will run into and solve any problems that crop up long before the short 2006 season starts. Among the requirements I imposed on myself, it must be easy to manufacture, cheap, easy to assemble and disassemble, and robust.

As such, I’ve elected to use a turn-key transmission. This iteration is based on the AM Gearbox from AndyMark. As things turn out, the output shaft is a bit shorter, I think, than that of the kit-provided transmissions.

http://img404.imageshack.us/img404/5631/px6bdrivesprockets7uo.jpg](http://imageshack.us)

As you can see, I have a sprocket hanging in space right now that probably won’t stay put – no matter how hard we all concentrate. I need to somehow extend the shaft to support that sprocket without making a new shaft.

What I anticipate is creating a cluster of two sprockets by bolting through each of them and a spacer so as to achieve and maintain the 2" separation I require. The end of the gearbox output sprocket appears to have a threaded hole, though Andy’s drawings don’t call it out specifically. I was considering then bolting through an extension and into that threaded hole, effectively extending the shaft length by 5/8" or so.

http://img444.imageshack.us/img444/4928/px6boutputextension6dl.jpg](http://imageshack.us)
http://img346.imageshack.us/img346/5820/px6bexplodedoutput8pz.jpg](http://imageshack.us)

Is this a ‘really bad idea’? I’m hoping that bolting both sprockets together will help to minimize the shear the currently unsupported sprocket exerts onto the extension it would sit upon.

Any ideas?

I know this is a really bizarre, obtuse question. I just wanted to see if I could get some additional opinions about best practices in cases like these so I am not later forced to abandon all of the neat design features this chassis has because of something as inconsequential as shear :wink:

From the drawing you have, I can’t really see much. Imean, I can sort of understand, but it’s hard to see. It looks like both sprockets are supported in the drawing. Could you show us another angle of the gearbox, say an end view or isometric? That way we can all see more clearly and provide better help.

If you look at the topmost image, look closely at the shaft between the two sprockets. It ends before entering the leftmost sprocket. It’s close, but close isn’t good enough :slight_smile:

The second image shows a really quick rendition of my proposed solution. The red piece represents the extension that would be bolted into the end of the existing output shaft. The leftmost sprocket from the top view would rest wholly on the red extension piece.

A picture paints a thousand words.

I hate to tell you this, but I just don’t see any way the screw that holds the extension in place being able to handle the torque that will be applied to it. In effect, that screw will be all that is driving the outboard sprocket and it’s applied load.
Now, if the extension was to have a section that fit over the shaft protruding past the first sprocket with a key way slot to drive it, no shoulder on the outboard side so the outer sprocket could then be mounted to it, and finally the center drilled through so that a screw passed through a washer , through the extension and into the trans output shaft to hold it all together, then that “might” work. Clear as mud??

I understand that the screw alone would shear under the load exerted by the chain onto the outermost sprocket. I was hoping that, by bolting the sprockets together, those bolts would transmit some of that load onto the main shaft. I understand the solution you propose as well, though I don’t believe we’re capable of broaching a keyway and thus, it’s a bit harder to implement. :slight_smile:

I may just opt for a different chain routing so as to completely eliminate the outermost sprocket.

Since you mentioned simplicity, I wont go into the welding options. If you have access to a mill and either a skilled operator or DRO then what you could do is drill four very precise holes in the sprocket that rides on the shaft and tap them. It is Very important that the bolt circle that these holes lie on is concentric with the bore. Then, make a spacer with a similar hole pattern but make them through holes just barely large enough to clear the screw. Then on your outboard sprocket, do the same thing. Bolt it together (and tight) and now you have a double sprocket to the spacing of your liking and you need not do anything to the shaft. The outboard sprocket is drven by the screws going through it. Remember to make sure you select a bolt hole circle such that the heads of the bolts do not interfere with the chain. At least that’s what I would do, given the design considerations.

EDIT: A way to do this without machine tools would be to clamp both sprockets and the spacer together with a bolt through the bores (and a nut on the other side). Get everything lined up and then tighten up that bolt real good. Make white paint line down all three pieces so you can line them up easily later. Then set up this assembly on a drill press and drill through all three at once with the bit that the tap requires. Once you have the four holes, pull the assembly apart and drill one sprocket and the spacer to just clear the bolts and tap the other sprocket. Assemble the three pieces with your four bolts making sure your white index marks line up. Having a shaft in the bore while tightening is a good idea btw to ensure concentricity

OOOpps, I didn’t catch this part. My bad. :frowning:
Your suggestion is another option that should work. :slight_smile:

In Sanddrag’s post, he mentions a couple ways to implement this fairly easily. The edited version really would be a fairly simple and effective way to do it with very little tooling required.

I tell you, that Sanddrag, what a genius!! :smiley:

Maddie, this is a neat design. Thanks for using the AM Gearbox into this mechanism.

I agree that bolting the sprockets together will work, but that second sprocket is a looong way from the gearbox.

Here is another solution: Make a longer output shaft. That shaft is simply 1/2" steel hex stock (we use 4140, but you don’t have to). You would simply need to turn down the hex to fit into the 1/2" id gears and add a keyway in the right spots.

Not only may this be easier than some other solutions, but you can support the far end of the output shaft this way.

Good luck,
Andy

Do the little posts shown in this picture serve any purpose? It’s hard to tell from the plans available on the site. They make the part look harder to manufacture, at the very least :slight_smile:

If it’s just a turned hexagonal shaft, we should probably be able to make that. I’ll have to see about the keyways, though.

Can you turn down the hubs enough to atleast get and E-clip on the shaft?

Yes, that’s the one thing I wasn’t able to add before the battery on my pocketpc died earlier today. I do believe it will work, but you shouldn’t put too much tension on the chain. I was also going to add that maybe you can still make something to screw into the end of the existing shaft to fill the empty void in the outside sprocket which will help things a little.

And I too am curious as to what that little peg is in the hex portion of the shaft in that AM photo.

If you don’t have a broach set it is possible to mill a keyway into the shaft. Not sure of the resources available to your team but I wanted to bring up that option.

Not sure why you recommend an E-clip? An E-clip will hold a sprocket on but won’t drive it. You need a keyway or what Sanddrag mentioned to do that.

Also milling down the sprockets will eliminate the 2" separation Maddie requires and planned for.

Actually, a keyway is usually milled/cut in a shaft and broached in a bore.

Now, for making a keyway in a bore without an arbor press. There is the possibility you can use something different for your press, like a mill or a drill press. Although I wouldn’t recommend it. It could damage the machine. Another possibility would be to use a vise (such as a mill vise) that can open it’s jaws wide enough to fit the gear and broach. Close the vise, and it presses the broach through. A precision vise can make an excellent press in a pinch.

For making a keyway in a bore without a broach, there is a way you can sort of “fudge” it. We’ve done this before and haven’t had any real problems. What you can do, is set up the sprocket/gear on a mill so that it’s face plane is parallel to the plane of the mill table. Using an edgefinder (and DRO) or a drill blank, locate the center of the gear and lock one of your mill axes in place. If you are using an 1/8" key, put a 1/8" endmill in a collet in the machine. Move the other mill axis such that the center of the endmill is directly above the point on the circumference of the bore. Take a plunge cut nice and easy straight down. (when I say easy I mean it. Touch the downfeed handle with something like the weight of a couple feathers. Those 1/8" tools are quite fragile) Now, move the axis of the mill table that is unlocked another 1/16" into the gear and take another plunge cut. You now have a 1/16" by 1/8" rectangle of material gone (just like a regular keyway) with the added “bonus” of a half circle on top. The key will fit just fine and be able to transmit torque reliably. However, over time it will want to move into that half circle area. So, fill the half cirle with something, anything, a toothpick, piece of paperclip, epoxy, whatever.

Alternatively, you could reduce the “rectangle” portion of this keyway and grind the key rounded somewhat. Or maybe get a 1/8" by 3/16" key, and grind it round one one side so it fits perfectly into your rectangle with half round on top keyway.

EDIT: If you use this method to make a keyway, make sure you are using a center cutting endmill.

Something I’ve been known to try is to give the mill table a little horizontal motion as the tip of the end mill is being very lightly plunged into contact with the workpiece (it helps to mill a small flat on the top of the piece first, to avoid trying to do this on the rounded surface). Where very fine precision isn’t necessary, this can produce acceptable results. I’ve done it with 1/8" and 9/64" keyways, with three-fluted non-centre-cutting endmills, and it works well enough. Workpiece materials were 416 stainless, if I remember correctly.

(Edit: For a non-centre-cutting endmill like the one that I described above.) Work slowly, and remember to never try to plunge the end mill straight down, even after the cut starts. Let it cut at an angle.

That is simply a 3/32" roll pin. It serves the purpose of holding a gear in place on the shaft. It is much cheaper than machining a step.

Andy B.

I’ve plunged (center cutting) endmills straight down on numerous occassions as if they were no different from a drill bit and have had great results. A professional machinist has even advised me to do so (to accomplish what I needed to).

I certainly wouldn’t go replace my drill bit set with a set of endmills and pop them in my cordless drill, but I think in the right application there is nothing wrong with plunging an endmill straight down. On those tiny 1/8 and 3/32 inchers, I’d be more inclined to plunge it straight down rather than take a heavy cut from the side in the case of making a makeshift keyway in a bore. I think the endmill would have less chance of breaking that way.

If the space allows, might welding the hubs of the two sprockets together work? it seem that you have the space available to secure one sprocket, but by welding it you could potentially have the teeth of two sprockets on the shaft of one. No idea for sure, but that may work… wouldn’t be too much welding to have done.

I was referring to “three-fluted non-centre-cutting endmills” in that post (because that’s what I had available in that size). In general, of course you can use a centre-cutting endmill as if it were a drill. Cutting diagonally like that does demand some care (and a hand on each of the vertical and horizontal feeds), but it yields good results. I wasn’t advocating a heavy cut, either; quite the opposite, in fact.

You’d have to put a spacer in between to keep the desired distance apart. But yes, welding up the assembly would be a good option, but you’d have to have a way to keep it all lined up (shaft through bores would be the best) and you’d have to make sure it doesn’t get distorted (don’t heat it for too long at a time and don’t cool it rapidly). Once you have those two things worked out, heck, you could even arc/stick weld it.

I sort of assumed you didn’t want to weld anything, which is why I posted about drilling and bolting together. But I didn’t really base that assumption on anything.

But if you have any sort of welder (you will probably need around 130 amps) then that would definitelty be the way to go. Also, if you have access to a welder, I think it is definitely something a HS student can do. Just take all proper safety precausions and practice for about a whole day first.

I think it better to get that outboard sprocket driven by the first one rather than have another shaft or something. Instead of making the shaft longer, make the two sprockets one. Who cares if the second one is out in the air. As long as it is rigidly attached to the first one, I’d say you’re good to go.