Re: Engaging Gears Perpendicular to Shaft Direction?
 

eeek... I gotta say I didn't do that! :ahh: (Though admittedly, small things like how much those shafts would deflect when they had a pneumatic on them were definitely calculated (.00045 inches))

To do so real quick:

Maybe 250 in-lbs on that center gear, 1 in diameter, moment arm of a half inch off center... the pneumatic is about 3.4 inches away from the pivot point, at around a 15 degree angle, providing a max moment arm in the opposite direction of around 95 lbs * 3.4 inches with a 1.5 inch bore... There's also the force on the gears through the shaft centers, sin(20)*250 is around 85, times it's moment arm of 1.5.... So we've got around 70 in lbs to spare with the half inch. whew! :cool: Thanks for the reminder...

I appreciate you noticing, and while this is true... sometimes outsiders see something obvious that the "careful thought" folks have missed which will make a device infinitely better... which is why it's out here. :)

I've reread this about 10 times now.. I just can't visualize what you're saying. Are you suggesting having some loose chaing that is engaged and disengaged with sprockets? If you keep the gears engaged constantly.. how would you shift?

I've already thought of some nice ideas about changing the sizes of the blue connecting sprockets to let them affect the gear ratio, and keep the gears a (larger to reduce load) constant size... but I'm certain this is not what you're proposing. If you could clarify, I'd truly appreciate it!

Thanks,

Matt

Re: Engaging Gears Perpendicular to Shaft Direction?
 

I think he means to move gear to chain, rather then gear to gear.
Have both final gears spinning, but rotate them to move the proper one into contact with the output chain. Puts all the meshing forces onto the chain.

At least, that is what I understood him to be saying.


Wetzel
~~~~~~~~~~~~~~~
Making and breaking chain with needlenose pliers is fun.
I swear.

How's this kluggie cut up?
 

The idea is to move the gears closer together so that they are always turning with the input gear.

Then put an idler sprocket on the same shaft as the input shaft (now the shaft marked with the green cross).

Then pivot the both output shafts on a lever that also pivots about the input shaft (it is outlined roughly in a T shape in the image attached).

Then when you drive this link, you engage one sprocket or the other with the chain. (The sprockets are those round black circle things)

You can have the air cylinder either be strong enough to take the load you are pulling on the chain -- in which case you sort of get a chain tensioner built in for free. Or you can add a drive link between the aircylinder and the sprocket pivoting gig - in this case, you can make the whole thing non-backdriveable and then tension in the chain cannot push the mechanism out of engagement (I like this solution better... ...but time is limited so I will have to discuss it another time if there is interest).

The chain routing is shown for the mechanism in high gear (the thickish blue line).

I hope this helps.

Joe J.

Re: How's this kluggie cut up?
 

That's BRILLIANT Dr. J !! :D

Re: How's this kluggie cut up?
 

Can the chain &/or sprocket(s) take this type of abuse (shift on-the-fly ...) ?

Re: How's this kluggie cut up?
 

If I had my choice of chain or gear teeth that had to take the abuse I know which one I would pick.

Shift on the fly may be a bit of a stretch, but I suppose that it wouldn't be too bad if you could figure a way to use #35 chain.

Joe J.

Careful is an adjective not to be glossed over...
 

You forgot the input torque of 250 in-lbs (or else you have to include the torque from the internal short chain -- which magically will work out to be the same 250 in-lbs).

I see 4 torques to deal with
In high gear with driving the input CCW,
Tinput = 250 in-lbs
Taircylinder = 95 lbs*3.4 in = 320 in-lbs
TtoothNormal= -(250 in-lbs/.5 in)*.5 in = -250 in-lbs
TtoothSep= ((250 in-lbs /.5) * sin(20deg) ) *1.5 in = 250 in-lbs
This is okay because they sum to 470 in-lbs (and CCW is toward engagement).

If you do the same thing only have the input torque switch to CW you will get
Tinput = -250
Taircyliner= 320
TtoothNormal = +250 (the normal tooth load is trying to ENGAGE the mechanism in this case)
TtoothSep = -250
Sum = 70 in-lbs

Having done the summing carefully, I get the same answer as your "quick calculation", so what do I know?

I suppose I will shut up now ;-)

Joe J.

Re: Engaging Gears Perpendicular to Shaft Direction?
 

Jeez!
Doing that would put waaaay more torque through this box than I would think is healthy. (I haven't run numbers on this). In order to function correctly, this thing would have to be as "early" in the gearbox as possible, to minimize torque on the shifter.

Matt--
I'm very impressed with this design.
This may be a step closer to the "rookie solution" to shifting the FIRST community could use so badly. (Sorry Andy ;))

I'm very curious to see how it will work out for you guys. :D

John

Re: Engaging Gears Perpendicular to Shaft Direction?
 

Not necessarily, if you’re doing a multiple motor gearbox you could just combine in the gearbox, go straight to the shifter, then adjust in the chain. Doing it this way could actually be really helpful as only the smallest sprockets come in nice sized bores and you need a very large sprocket if you only want to drive the wheel(instead of the whole shaft) in your chain.

National champion team 469 used the exact same(or at least very similar) shifting design in 2003 with two motors per side without any trouble, even though the shifter was very late in the transmission.

Re: Engaging Gears Perpendicular to Shaft Direction?
 

One change that can be easily made to the shifter to increase the speed change without using different gears could be making the output shaft always be the shaft that currently has the 1:1 ratio, and changing the blue sprocket to a 20 tooth if the other one is a 10 tooth, or something similar. This would give you a 3:1 reduction and then a 2:1 reduction giving you 6:1 and a 1:1. you could do less or more teeth to make the increase more or less dramatic. That way newer can be very competitive with just an atwood drive having maybe 6ft/s and then 1ft/s or whatever they would like.

Re: Engaging Gears Perpendicular to Shaft Direction?
 

The torque applied depends greatly on your motor combining, and what speed they are matched at. Especially if this combiner gearbox is designed to go straight to the wheel after (as Rob mentioned). The final sprocket reduction is limited by several factors. (Such as Chain size... note: I believe "national champion team 469" blew 1/4" chain like nobodies business towards the end)

Also... I'm almost 100% positive, "national champion team 469" used a shifter NOT like Matt's, but similar to 60-2002, or 116-2002 -- where the gears slide laterally. This system is able to handle much more torque than Matt's (in which applied torque would actually try to push the shifter out of gear).


John

Re: Engaging Gears Perpendicular to Shaft Direction?
 

Yes it does, however that torque is the same if you use a dog shifter or a shifter such as this. Plus he said he was possibly using 16 pitch gears, and steel 16 pitch gears are a lot stronger than your average shifter.

Keep in mind, his design doesn't even use #25 chain. While the reductions possible using sprockets is somewhat limited, its still pretty wide. Using #35 chain, you can get as much as a 3:20 reduction, which is pretty good. If you needed any more than that you could just change the ratios inside the shifter or add another box with another reduction after.

Some Info
 

Actually, this really depends on what you mean by "average shifter" I would wholeheartedly rather bet me life savings (which is admittedly negative since I'm a college student) on a dog style shifter than gears engaging face to face, if you're speaking about going on the fly. WHICH THIS DESIGN IS NOT INTENDED TO HANDLE. :)

I've plugged some numbers into a Lewis bending equation spreadsheet I've made, and let me tell you, with the 1" gears, failure is VERY possible, depending on where you place it in the drive train. Using the initial configuration I had, the factor of safety was around 0.7... :ahh:

I most definitely do use #25 chain in this design, since it's loads are planned to be somewhat light. If I were placing this on a drive train, I'd probably slow the Chips down by maybe by 3 or 4, throw that shaft into this magic box, and probably use sprockets on the wheels to slim it down by another 4. By this, I don't expect the gear box to see torques exceeding around 50 or 75 in-lbs. I think that #25 is safe in these situations, even with the small diameter of the sprockets being used. #35 chain might add another half pound per box, for no real good reason.

Just some thoughts,

Matt

Re: Engaging Gears Perpendicular to Shaft Direction?
 

I think you fail to understand what I'm saying, and the main principle behind my comment.

Any torque applied through THIS design, will cause the gears to try to disengage from each other. The only thing preventing this is the force of the pneumatic, holding the shafts together.

In a Dog-style, Pin-Style, Mesh-Style... Shifter, the kind of shifters we see every day in this competition: The shafts are held rigidly. Therefore the applied torque CANNOT seperate the gears from meshing (unless it deflects the shaft, which is a discussion for another time).

See the major differences?
See why Matt's design would have MAJOR torque limitations that wouldn't apply to a "normal" shifter?

I can guaruntee you haven't seen any design like this one in the FRC, so it is unfair of you to draw parallels between this box and another and make broad statements concerning it's capabilities.

I love 469, they are a great team, their last 2 bots have been works of art, and their 2002 bot kicked my butt pretty hard. Great designs, in every part of their bots, including the shifter.
But... that doesn't really apply here, now does it?

Unless you understand all the principles involved, and the design itself... which you've shown some confusion about... perhaps you should limit the broad statements you make about it?


John

Re: Engaging Gears Perpendicular to Shaft Direction?
 

Joe Johnson had said earlier that the design should be able to hold in gear just with the piston. If that was ever a problem you could always increase the bore size. With over 120 pounds pushing against it you can put some considerable torque into the shifter and under normal circumstances not have to worry about it disengaging. Matt had even said that he has a spring system in the workings to hold it in tighter, so if you were afraid of it comming out in low gear you could always reinforce it with springs.


I believe the statement about the chain sizings was reffering to the final chain coming from the output shaft and powering the wheels.


John, lets try to keep negative comments out of general discussion. The point of the thread is to express ideas and possible improvements to Matt's design and to TEACH people if they don't understand a concept. You could possibly explain something you feel is being misunderstood instead of suggesting the person not post ideas they may have. Undeserved comments should be kept in private, not on the forum discussions.

With the possibility of using multiple motors, you can still put this shifter as early in the transmission as you would with any other shifter. From my understandings Matt was trying to make a shifter for newer teams who do not have the machining capability that some of us have. No where has anyone said that the design was perfect, but you can always work in some additional support to the shifter to accomodate a more torquey drive to an extent.