2GodBDGlory's Ideas/Mechanisms Topic

[2GodBDGlory]

There've been a bunch of times I've had just a little idea I wanted to post on here, without making a new thread for it. I've put them in different generic threads in the past, but I figured it might make sense just to make a personal thread to share stuff of this sort.

Anyways, the first thing I want to share is this ten-sided "Geneva Mechanism" I came up with today.

It's using a 10T splat gear with a pretty standard mechanism for rotating it intermittently, as in a stepper. However, rather than having a spring-loaded part pressing against the teeth of the gear as in a stepper, I've got a an axle that locks the teeth of the gear when it isn't being engaged. The same axle that runs the beam that steps the teeth is also hooked up to a wave selector, so that when it approaches the position in which it will have to move the teeth of the gear, the bump on the selector moves a linkage that briefly unlocks the gear, and then automatically locks it again after the motion is done.

It's working well in one direction right now, as seen in the video, but it's not working the other way. I think this is because of the sideways force the wave selector exerts on the black beam, but I'm pretty confident it could be fixed with a more thoroughly engineered solution than this proof of concept.

 

[2GodBDGlory]

And, I already came up with a design I like better... I should probably hold off on posting stuff like this until I refine it. Because of that principle, I'll wait on showing my new design until I decide if it's final or not.

[deehtha]

1 hour ago, 2GodBDGlory said:

It's working well in one direction right now, as seen in the video, but it's not working the other way. I think this is because of the sideways force the wave selector exerts on the black beam, but I'm pretty confident it could be fixed with a more thoroughly engineered solution than this proof of concept

With the rubber band on one side, your black beam experiences more force on the side. If you can get the force equally exerted on the other side, you should then be able to move in the other direction. Because of the looseness of the pieces, even the smallest forces can cause beams to be slightly angled.

[2GodBDGlory]

10 hours ago, deehtha said:

With the rubber band on one side, your black beam experiences more force on the side. If you can get the force equally exerted on the other side, you should then be able to move in the other direction. Because of the looseness of the pieces, even the smallest forces can cause beams to be slightly angled.

That probably is a factor, but I think the main thing is how the selector pushes the beam sideways. This sideways motion ends up pulling the axle back a bit before the beam has even popped up on the selector's bump, and I think it's this preselection that makes the one side work better than the other. Like I commented, though, I've got a totally different design that works quite a bit better, so I probably won't develop this design any further at present.

[SNIPE]

Cool use of the orange wave selector.

[Hrafn]

19 hours ago, 2GodBDGlory said:

And, I already came up with a design I like better... I should probably hold off on posting stuff like this until I refine it. Because of that principle, I'll wait on showing my new design until I decide if it's final or not.

😂 I hear you on that!  As soon as I post something I invariably realize the flaws with it.  

[Thierry-GearsManiac]

It reminds me this variant of a locker-based Geneva mechanism :

I noticed this principle long time ago in several door lock types, but I never thought about implementing it in LEGO for intermittent rotation.

[2GodBDGlory]

I was playing around with 16L links today, and came up with an unusual leaf spring design using them:

The link itself is fairly flexible and works well as a spring; the difficulty comes in attaching anything to it! I used this part:

https://www.bricklink.com/v2/catalog/catalogitem.page?P=32172&name=Technic, Pin Connector Block 3 x 3 x 1&category=[Technic, Connector]#T=S&O={"iconly":0}

which can clip onto an axle without sliding from one end. Unfortunately, it would slide forward and backward along the link, so I added a 6L link to place it longitudinally. I'm not sure how helpful this concept will be, but I thought it was a satisfying part usage!

[Jurss]

This is creative.

How that part really locks on that axle?

[2GodBDGlory]

Thanks!

The part locks on there surprisingly strongly, using the clips on the back of that part.

It's a very interesting part, because I don't know of anything else that allows axles to be attached that way!

[Jurss]

OK, I think I'll add this part to wishlist on briclink, need to try.

If axle will go trough (like on leaf spring axles), not that it just holds on to that one side, then it could really work.

Edited August 31, 2022 by Jurss

[2GodBDGlory]

7 hours ago, Jurss said:

OK, I think I'll add this part to wishlist on briclink, need to try.

If axle will go trough (like on leaf spring axles), not that it just holds on to that one side, then it could really work.

Yeah, it clips on and holds both sides. With enough force it can pop out of the side, but it holds fairly tight.

[2GodBDGlory]

Here's an interesting gearbox concept I've been working on lately:

I've been intrigued by gearboxes based on the planetary hub for a while, based on the potential to replicate real-life automatic-style gearboxes, and based on the potential for heavy-duty applications.

I'm not sure this is the ideal layout, and it's likely not the ideal build for this given layout, but it's an interesting idea, I think.

So, the shifting in this gearbox is handled by a sliding row of 20T gears, but it's not a traditional sliding-gear transmission. In one gear position, the gears are meshed with both a 28T turntable connected to the body of the planetary hub, and a 28T gear attached to the driveshaft coming from the motor to the CV shaft of the planetary hub. This locks the hub and the driveshaft to rotate together, making for an overall 1:1 rotation.

In the other position, the 20T gears are still meshed with the 28T turntable connected to the hub body, but they are also meshed with a fixed, non-rotatable 12T gear. In this position, the 28T turntable, and by extension the body of the hub, are locked in position, forcing the internal gears on the hub to come into play for a 1:5.5 reduction.

The design I currently have will stall a single XL motor in high gear, and will dramatically twist the output axle in low gear rather than skip any gears, so it's a reasonably robust design. I would recommend using it with less input torque, though, simply because the level of output torque in low gear is pretty unmanageable for any downstream drivetrain components when using an XL motor! Probably the main downside is the difficulty of sliding the gears for shifting, which looked annoying enough that I didn't bother trying for this proof-of-concept model. It's also worth mentioning that shifting while the motor is running isn't recommended, because the halfway position locks up everything and stalls the motor unless you shift fast enough.

Here's a video:

More images at: https://bricksafe.com/pages/2GodBDGlory/miscellaneous

[dustblue]

On 3/11/2022 at 9:33 AM, 2GodBDGlory said:

There've been a bunch of times I've had just a little idea I wanted to post on here, without making a new thread for it. I've put them in different generic threads in the past, but I figured it might make sense just to make a personal thread to share stuff of this sort.

Anyways, the first thing I want to share is this ten-sided "Geneva Mechanism" I came up with today.

It's using a 10T splat gear with a pretty standard mechanism for rotating it intermittently, as in a stepper. However, rather than having a spring-loaded part pressing against the teeth of the gear as in a stepper, I've got a an axle that locks the teeth of the gear when it isn't being engaged. The same axle that runs the beam that steps the teeth is also hooked up to a wave selector, so that when it approaches the position in which it will have to move the teeth of the gear, the bump on the selector moves a linkage that briefly unlocks the gear, and then automatically locks it again after the motion is done.

It's working well in one direction right now, as seen in the video, but it's not working the other way. I think this is because of the sideways force the wave selector exerts on the black beam, but I'm pretty confident it could be fixed with a more thoroughly engineered solution than this proof of concept.

 

wow this is very interesting! clever design!

4 hours ago, 2GodBDGlory said:

Here's an interesting gearbox concept I've been working on lately:

I've been intrigued by gearboxes based on the planetary hub for a while, based on the potential to replicate real-life automatic-style gearboxes, and based on the potential for heavy-duty applications.

I'm not sure this is the ideal layout, and it's likely not the ideal build for this given layout, but it's an interesting idea, I think.

So, the shifting in this gearbox is handled by a sliding row of 20T gears, but it's not a traditional sliding-gear transmission. In one gear position, the gears are meshed with both a 28T turntable connected to the body of the planetary hub, and a 28T gear attached to the driveshaft coming from the motor to the CV shaft of the planetary hub. This locks the hub and the driveshaft to rotate together, making for an overall 1:1 rotation.

In the other position, the 20T gears are still meshed with the 28T turntable connected to the hub body, but they are also meshed with a fixed, non-rotatable 12T gear. In this position, the 28T turntable, and by extension the body of the hub, are locked in position, forcing the internal gears on the hub to come into play for a 1:5.5 reduction.

The design I currently have will stall a single XL motor in high gear, and will dramatically twist the output axle in low gear rather than skip any gears, so it's a reasonably robust design. I would recommend using it with less input torque, though, simply because the level of output torque in low gear is pretty unmanageable for any downstream drivetrain components when using an XL motor! Probably the main downside is the difficulty of sliding the gears for shifting, which looked annoying enough that I didn't bother trying for this proof-of-concept model. It's also worth mentioning that shifting while the motor is running isn't recommended, because the halfway position locks up everything and stalls the motor unless you shift fast enough.

Here's a video:

More images at: https://bricksafe.com/pages/2GodBDGlory/miscellaneous

this one too! the concept is definately interesting. I myself use custom made planetary gears and turntables a lot, but never used the 28t turntable this way!

 

Edit: watched second time and saw you are using 46490, not custom made. still very interesting!

Edited September 25, 2023 by dustblue

[glowytheglowbug]

pretty cool gearbox design i wonder how much torque it can take though and is there more resistance to shifting when there is load?

[2GodBDGlory]

6 hours ago, glowytheglowbug said:

pretty cool gearbox design i wonder how much torque it can take though and is there more resistance to shifting when there is load?

Thanks!

In this configuration it can take at least full XL motor torque, which I think is honestly more than you need, because given how low low gear is, the level of output torque is already dangerous to axles!

There probably would be more resistance to shifting under load, but I'm really only recommending to shift while the motors are off, anyways.

[lmdesigner42]

That's a pretty neat gearbox concept, similar to real-life planetaries and it seems very robust. Would there be any advantage to using a traditional transmission driving ring to switch between a 1:1 ratio and going through the planetary hub?

[2GodBDGlory]

27 minutes ago, lmdesigner42 said:

That's a pretty neat gearbox concept, similar to real-life planetaries and it seems very robust. Would there be any advantage to using a traditional transmission driving ring to switch between a 1:1 ratio and going through the planetary hub?

Thanks!

That's actually a good thought--it hadn't crossed my mind! At first thought I wasn't sure the logic would work out for the different modes, but this mock-up here should have the correct logic to work:

As for whether it's advantageous or not, I'd probably have to build and test a full gearbox. It would definitely be easier to shift, which would be a massive plus. I think the gearbox itself would get larger--I'm thinking three studs longer with this driving ring, or maybe just two with the Yamaha one (This is including one stud that I think I'd be able to save between the turntable and hub) Having the shifting mechanism more compact would probably make it worth it, though.

The main question is whether or not the driving rings can actually cope with the levels of torque that a model would require. One helpful thing is that the driving rings are being driven at a 28:20 upgearing, so they have a bit less torque acting on them than the normal system torque. I generally don't trust them in truly heavy-duty situations, though, so it's hard to say.

Good suggestion!

[glowytheglowbug]

2 hours ago, 2GodBDGlory said:

Thanks!

That's actually a good thought--it hadn't crossed my mind! At first thought I wasn't sure the logic would work out for the different modes, but this mock-up here should have the correct logic to work:
 

As for whether it's advantageous or not, I'd probably have to build and test a full gearbox. It would definitely be easier to shift, which would be a massive plus. I think the gearbox itself would get larger--I'm thinking three studs longer with this driving ring, or maybe just two with the Yamaha one (This is including one stud that I think I'd be able to save between the turntable and hub) Having the shifting mechanism more compact would probably make it worth it, though.

The main question is whether or not the driving rings can actually cope with the levels of torque that a model would require. One helpful thing is that the driving rings are being driven at a 28:20 upgearing, so they have a bit less torque acting on them than the normal system torque. I generally don't trust them in truly heavy-duty situations, though, so it's hard to say.

Good suggestion!

ah thats a good solutution

[The_Cook]

Rather than use a shifter to lock the outer ring most real-life planetary gearboxes that I've encountered (https://www.rm1872.org.uk/GB32_Gearbox.html) use a brake band pulling tight against the outer ring to lock it, this overcomes the problem of having to slide the shifter into a moving gear.  How that could be replicated in ABS, not quite sure, possibly by using one of the gears with larger teeth, eg. splat gear, and engaging a pin to stop the rotation.

[2GodBDGlory]

4 hours ago, The_Cook said:

Rather than use a shifter to lock the outer ring most real-life planetary gearboxes that I've encountered (https://www.rm1872.org.uk/GB32_Gearbox.html) use a brake band pulling tight against the outer ring to lock it, this overcomes the problem of having to slide the shifter into a moving gear.  How that could be replicated in ABS, not quite sure, possibly by using one of the gears with larger teeth, eg. splat gear, and engaging a pin to stop the rotation.

Yeah, I think you're right about the bands being the typical way to do it. I'd agree that doing it with Lego would be pretty tough! I wonder if this track could be used for it: https://www.bricklink.com/v2/catalog/catalogitem.page?P=x939#T=S&O={"iconly":0}, but then you'd probably need some kind of rim that would have fairly large diameter for good grip, and a pinhole in the middle to allow drive through, and a way to attach it to the hub other than that center pinhole, which sounds pretty challenging.

The splat gear is an interesting idea, and one I hadn't thought of, but I think the working principle would be similar to the ones I've used in previous designs, like the automatic ones I put in this thread:

(I've also tried making manual versions on this working principle, but they didn't work very well)

If I'm not mistaken, though, you'd still need some kind of clutch to lock the drive axle to the body of the hub, since the belt/splat gear would only replace the locking side of the driving ring, and not the engaging side, and I haven't had any success at coming up with a reliable clutch going into the hub body.

Good thoughts!