Auto reverse mechanism

[Brian Rupnow]

I am getting a yen to machine something, but not an engine. I am somewhat intrigued by mechanisms that have an "automatic reverse" function to them. Not just a simple function like reversing gears, but more like something that moves along a short track, stops at the end of travel, then reverses movement and moves back to the original starting point and then repeats. The power source for this would be one of my many small engines. I have googled "reversing mechanisms" and there is a great array of articles posted about this. Of course, I want it to be something with a considerable "WOW" factor to it. I have the capacity to make spur gears here, but not internal gears nor funky gears like you see on a mangle.(Which is super neat.) The biggest problem I see to a mechanism like this is that when it gets to the "turnover" point, if it has to shift a lever to go into reverse it will probably just "hang up". The actual mechanism which moves something along can be a threaded rod with a threaded follower that reverses rotation, or a rack/pinion drive, even possibly a flat belt drive that shifts from a clockwise rotating pulley to a counter-clockwise rotating pulley. The big deal is that it has to do this without human intervention. I don't have the capacity to make a double reverse thread like you see on "level-wind" fishing reels. My tools are strictly limited to a manual lathe and a manual mill and a full set of 24 pitch gear cutters. I will have no problem designing whatever it is that I do, it's more the fact that I have to figure out exactly what I want to do. This is a wide open game, so if you have a good suggestion or a link to this type of mechanism working, by all means go ahead and tell me about it.---Brian

[Gas_mantle]

Brian,

I used to be a collector of Meccano and some guys in that field of interest created allsorts of peculiar mechanisms, if you type 'Meccano mechanisms' in youtube you may find something of interest.

You'd be surprised what kind of interesting ideas you can get from trolling through some of the more ambitious Meccano models :-)

Peter

[Brian Rupnow]

Yes Peter, I have had a look at Meccano mechanisms, and although they are very clever, I haven't seen exactly what I'm after.---Brian

[John Hill]

Brian, I am confident you could make the double reverse thread with your mill provided you have a collection of gear wheels such as a lathe change wheel set and a means to mount your workpiece between centres horizontally on your mill table.  The job requires you to gear the work piece to rotate very slowly driven from your mill table feed.


Here is what I did..http://madmodder.net/index.php/topic,11137.msg129456.html#msg129456

As you can see I used a rotary table and chuck to hold the workpiece but I am sure there are other ways including two centres with a gear on the workpiece.

John

[Brian Rupnow]

This is a rough idea of how a spring loaded toggle would work. The tension spring will always try and make the toggle rest either as shown, or rotated to be against the other stop, with a small dead spot when the toggle is perfectly vertical between the two positions. Any force applied in direction A would make the toggle swing up to the dead spot (vertical) and then a bit more force will make it snap over to the second position. This "swing" of the toggle would have to be used to move a set of tumbler gears into position to reverse the mechanism. It's not a great drawing, but I think it shows what I'm thinking of.---Brian
http://imageshack.com/a/img923/3755/hu0oJO.jpg

[BillTodd]

These sorts of things?

[Brian Rupnow]

Yes Bill, that kind of thing. However, as I said in my initial post, I don't have cnc equipment, nor the ability to cut anything other than external spur gears.

[sbwhart]

Made something like that when I was an apprentice form memory it was a square form thread left and right hand so you ended up with a diamond shaped thread.

They also used a similar thing on the UK 30mm Rarden cannon to load the ammo the gunner fed a clip of ammo to the feed tray and wound a handle clock wise it fed the ammo across the tray picked the first round up fed it into the chamber and then the loader returned to the start position, again the screw had a left and right hand thread the left hand had courser pitch so that it returned quicker.

https://en.wikipedia.org/wiki/RARDEN

Stew

[BillTodd]

Yes Bill, that kind of thing. However, as I said in my initial post, I don't have cnc equipment, nor the ability to cut anything other than external spur gears.

that resersing gear could be fabricated from easily cut geared parts i.e. the semi circular ends coild bemade from half a spur gear the circular rack made with a rotary table (angled appropriately)  .if the tooth clearance is generous enough , it should mesh well enough for. demo.

the real one looks makeable
http://douglas-self.com/MUSEUM/POWER/gear/gear.htm

[Brian Rupnow]

I like this mechanism
<a href="https://www.youtube.com/watch?v=Bxtthb_6bO8" target="_blank">http://www.youtube.com/watch?v=Bxtthb_6bO8</a>

[ShopShoe]

Brian,

I think that one has the visually intersting appeal that you desire, particularly if the gears are easily visible to the observer.

ShopShoe

[Brian Rupnow]

Yes, I would have the gears visible. I can make the gears. I can buy a 20" long rack from McMaster Carr. I think I like this one, and reverse should be simplified by running the toggle into a fixed stop to shift it.

[Stuart]

Brian
You have been standing with that in front of you for quite a while ,
Apart from the rack its the banjo reverse on your lathe to reverse the lead screw

[Jasonb]

Brian
You have been standing with that in front of you for quite a while ,
Apart from the rack its the banjo reverse on your lathe to reverse the lead screw

not on our far eastern machines

[Stuart]

opps though Brian was from the USA and only used American iron 

but I did have a typo I meant to say tumbler reverse

but you live and learn as I use British iron its the norm for me 

[Brian Rupnow]

Stuart--I am aware of the tumbler gears in a lathe, but I want my mechanism to be self reversing.

[Brian Rupnow]

http://imageshack.com/a/img924/2116/rElEUl.jpg

[Brian Rupnow]

This is what my gear cluster would look like, using gears that I can cut. 1.042" p.d. and 1.5" p.d. the rack can be purchased from McMaster Carr.
http://imageshack.com/a/img923/3594/in0cAs.jpg

[Brian Rupnow]

Okay!!--I followed George's instructions, and it worked. The design isn't finished yet, but it's getting close. I would prefer to have the rack setting still and the pulley and gears moving back and forth, but that means I would require some kind of tensioning device on the drive to the large pulley.

[steamboatmodel]

As a child a chum of mines dad had a small steam locomotive that ran on a straight length of track. there were three stations one at each end and one in the middle. the engine would slowdown then stop at the end stations, then reverse.  After every five trips down it would stop at the middle station, where he had a water tower that you could fill the tender. It was all steam powered and tripped by pegs set between the rails. We moved before I got old enough to really understand how it all worked.
Regards,
Gerald.

[Jasonb]

Brian, if you want the tumbler to move then you could use a belt wrapped round the tumbler shaft on a small pulley, a pulley at each end and one of those driven by your large pulley. There are several ways to wrap the belt and keep it in tension but below is the simple principal.

[Brian Rupnow]

Thank you Jason.

[zeeprogrammer]

Just popping in to say I'm watching. Looks to be interesting and I can see a lot of applications. 

[Brian Rupnow]

This will be my travelling carriage. Three of the gears are the same, so they can all be cut at once, then parted off to length. The blue ball bearings attached to the carriage plate will run in grooves in the stationary rack support plate to give guidance and stability.

[Brian Rupnow]

Hi Zee---Glad you're watching. I'm not sure if this is going to work or not, but it's interesting. The animation I posted looks good, but we all know that just because you can do something on a computer don't necessarily make it so.---Brian

[Brian Rupnow]

Here we have the full set up. The jackshaft at the bottom is centered between the two endplates, and the drive belt to the top pulley from the bottom is a 0.100" diameter rubber o-ring.  The jackshaft pulley is actually a double pulley, with the larger diameter hooked up to one of my engines via a second rubber o-ring which is not shown. As the moving gear cluster goes from right to left or vice-versa, the rubber drive belt stretches to accommodate it. The grey colored horizontal bar has a groove milled in each side to accommodate the blue roller bearings on each side of the "traveller" to keep everything aligned.

[10KPete]

Interesting!! A very nice do nothing machine... I love it!

Oh, I think you'll find that the o-ring isn't stretchy enough for that application... ????

Pete

[Brian Rupnow]

So, here we have the beginning of the old "cut three gears at once" trick. Outer diameter of gears and outer diameter of hubs have been turned, and reamed to 0.352" diameter center hole, all in the same set-up. (It was going to be a 5/16" center hole, but  discovered that I didn't have a 5/16" reamer at the very last minute, so I went up one size to 11/32" which I do have.) This method will let me cut the teeth on all 3 gears at once, then part them off in the lathe. Of course it also gives me the opportunity to screw up all three gears at the same time, so I'll try not to do that.

[Brian Rupnow]

Once again the machining Gods seem to have smiled on me. There are 25 spaces between the teeth on these gears. When I finally crank the rotary table around to that 26th position and don't hear any metal being cut, it gives me a great feeling of relief.

[Brian Rupnow]

I've reached my limit of fun for today. Three brass gears are finished, and the one larger gear I am making from 1018 steel. The blank is all finished, and tomorrow I will cut the teeth on it. And yes, I have decided that it needs a toggle to keep the teeth on the gears engaged with the rack. I will put some cad models of the toggle up tomorrow.

[Brian Rupnow]

Interesting!! A very nice do nothing machine... I love it!

Oh, I think you'll find that the o-ring isn't stretchy enough for that application... ????

Pete

Rubber O-rings are surprisingly stretchy. The travel on the reversing mechanism is fairly short, so I think the o-ring drive will stretch enough to accommodate it. if not I will put a spring loaded tension arm in there instead.

[10KPete]

Well, in any case I know you'll figger out something, Brian. That's the neat thing about your threads; you reveal your thinking process as you go along. Very educational..

Pete

[Brian Rupnow]

And here we are, the entire spur gear family all posing for a family photo.

[Ye-Ole Steam Dude]

Beautiful work Brian and a very interesting thread.

Thomas

[Brian Rupnow]

We have a finished and fully operational gear cluster. This didn't go quite as smoothly as I would have hoped. The chuck on my rotary table is attached with bolts in T slots, but no centering spigot. Sometimes, (I expect for sins I have committed in the distant past) the damned chuck moves a bit off center. It's not visible, but you can end up with slightly off-center gears. The pitch circle ends up being shifted very slightly from the bore. This of course results in gears that mesh very good on one side of their rotation, and are very very tight on the other side of their rotation. However, with a bit of fettling and cursing, everything now meshes very well, and I've done enough work for today.

[Brian Rupnow]

Since I don't have room for a spring tension toggle device directly below the actuator arm because of the shaft running thru, then I will put a compression spring acting on it from above. This will serve as my toggle device.

[Art K]

Brian,
Just wanted to let you know that I am following along. You always come up with interesting gadgets for your engines to run.
Art

[Brian Rupnow]

Thanks Art.  It gets harder and harder to come up with something different. I like something different like this, and I'm never absolutely sure whether it's going to work or not until I have finished it. I did make sure to select gear sizes that will work on model engines, so if the design is a total wash, at least I can reclaim them.---Brian

[Ye-Ole Steam Dude]

Brian how are you holding the 3 brass gears on the shafts?

Thomas

[Brian Rupnow]

After I got home from my "away office" this evening. I whittled out the "Pivot anchor plate". Got all of the holes drilled and reamed, and the profile roughed out. I will pick up the four ball bearings tomorrow.

[Brian Rupnow]

Brian how are you holding the 3 brass gears on the shafts?

Thomas

The bore in the gears and backing plate is 0.344" diameter. The shafts are made from 3/8" material, turned down to 0.344" all except for a 1/16" long section at the end. The shaft is pushed thru the gear, thru the backing plate, and silver soldered. The 3/8" diameter head holds the gears on.

[Brian Rupnow]

Today we have the four bearing mount shaft made and pressed into the aluminum backplate, with the bearings Loctited to the shafts. Why does one bearing have a blue grease seal while the other three are black? Beats me--Must be something to do with "diversity".

[Brian Rupnow]

In the final phase of completing the gear cluster and "toggle" device, I decided it would be wise to make the compression of the toggle spring adjustable.

[Brian Rupnow]

I've been a busy little beaver. The 6" diameter pulley is finished, and the toggle is finished. The toggle seems to work very well, and definitely will "kick" the gear mechanism down into position as soon as it is "over-centered" in either direction. Hopefully, tomorrow I will get the slotted track machined and attach the rack to it.---Brian

[Brian Rupnow]

Today I have the first trial assembly of all the important pieces. Everything fits and clears properly, but I'm not getting enough "kick" from the toggle to shift things cleanly. I may mill some material away below the rack to let the rack set a little farther away from the gear cluster. This will allow for more movement when the gear cluster shifts position, and the toggle should have more movement so should be more effective.

[crueby]

Should be some way to make a clock pedulum out of that, 'tick' on the movement to one side, 'tock' on the movement to the other...

 

[Gas_mantle]

Looks like it's taking shape nicely :-)

I came across this reversing oddity by accident and wondered if it may interest you if you decided to have another go at something :-

<a href="https://www.youtube.com/watch?v=3aobPgGzB-U" target="_blank">http://www.youtube.com/watch?v=3aobPgGzB-U</a>

[Brian Rupnow]

Well, I'm finished enough to begin having doubts about whether it will work or not. It works when I turn the pulley by hand and use my other hand at the end of stroke to help flip the toggle over. However, as I expected, the problem is going to be getting that toggle to flip over nicely and consistently. I've spent enough of this weekend working to get it done to this stage. I will play with it in the coming week to see if I can get the toggle to work better.

[BillTodd]

it might need springs on the ends of the buffers to ensure a clean switch

Bill

[Florian Eberhard]

Hey Brian

I like those mechanisms you are building, they (also if they are only on a sketch or so) always attracht my attention

I just wanted to write
maybe you need to gear up the flywheel so it turns with more rpm so you cet enough energy out of it at the end of the track
But after thinking again it doesnt seem to be the problem since you are driving the mechanism via the flywheel. I guess there is a state where both gears are not engaged and this is exactly where the whole thing comes to a stop?

I think then either Bill's Idea with the springs at each end could help or some kind of "lever" between the toggle spring and the toggle itself so the "machine" pushes over that kind of switch and only after that, the spring will force the toggle to change gears.

Acutally its the same as with the single cylinder steam feed pumps. They only work because there is the shuttle valve which switches over before the steam for the main cylinder is reversed to the other side and therefore the movement of the main cylinder stopped (which drives the first valve piston or c-valve).

If you have something that goes over the dead center before the driving force is cut off, then this should work better than it does right now.


Cheers Florian

[zeeprogrammer]

You'll get it Brian. No doubt. 

@Gas_mantle - cool video.

[paul gough]

Maybe having a look at, 'Ingenious Mechanisms, for designers and inventors' by Frankln D. Jones might help or extend your design as there are chapters on reversing mechanisms. It is a four volume work, might be in a library, but easily obtainable second hand, try Abebooks, should be pretty cheap in Nth America as it is a U.S. publication from Industrial Press. I have always enjoyed mechanical contrivances they seem to be as fascinating as watching monkeys. Regards Paul Gough.

[Brian Rupnow]

After playing with this thing for an hour after supper, it has loosened up considerably. I have come to the conclusion that I need a stronger spring and more hysteresis in my toggle. I'm sure that a lot of you old guys could use a bit more "hysteresis in your toggle", but that's not what I'm talking about. I can buy a stronger spring sometime this week, and I think that if I lower the rack by milling away about 0.100" of material below the rack, that will increase the "hysteresis". (Let the toggle move more between full left and full right position.)

[Noitoen]

Instead of a coil spring you could use a leaf spring. Look at the snap-action functioning of a microswitch to get the idea.

[BillTodd]

part of your problems could be the direct connection between the switch and the drive. i.e. as the gear drives toward the stop it is lifted out of the rack by the toggle .

if you look at a toggle switch (light switch) the operating lever can move with out the contacts moving until the spring is moved over centre.

if you adjusted your mechanism so that the stops moved the upper pivot (top.of the spring) until it flipped the rocker , it would work more like a toggle switch

[Brian Rupnow]

 I'm kind of hung up on this mechanism at the moment, and I've been busy with other stuff. I just spent today driving 390 kilometers  to see my mother and take her a birthday cake and celebrate her 97th birthday. I will probably get back to this sometime in the next week or two.---Brian