Didumos69

[POC] Reliable two-way 90° stepper w/ instructions (ideal for Servomotor)

Recommended Posts

In pursuit of a reliable two-way 90° stepper that is capable of handling high torque without slipping, I came up with this solution. It has been specifically designed for use with a PF Servomotor, that is for 90° input rotations, clockwise and counterclockwise.

How does it work?

While idle, the differential housing can't rotate, because it's locked by two ratchets (which lock in one direction and slip in the other direction). The ratchets are tied together with a single white silicon band, only one stud distant from their rotation points (not visible in video). When the input engages, one ratchet opens, but the other ratchet prevents the differential from rotating along with the input axle. The result is that the input flows through the bevel gears inside the differential housing, and outputs with a 1:1 ratio. When the input returns, the 90° limiter prevents the output from returning, forcing the differential to slip over the one ratchet that is still engaged. After returning completely, the open ratchet re-engages and the differential housing is completely locked again.

When you listen carefully, the return gives three clicks. The differential housing returns with half the speed of the input axle. This means it will rotate backwards 45°, which covers exactly 3 teeth of the 24-teeth side of the differential.

Background

It all comes very precise. Here are a few requirements that needed to be met:

  • After a return, when the differential is idle and locked by both ratchets, the exact position where the ratchets hook on to the differential should be such that one tooth of the 24t side of the differential is pointing straight up. This is needed to get equal starting points for shifts in both directions. The previous shift should not put one direction in favor of the other by leaving a slighlty off-center differential. The 12t gears hook on to the differential at exactly the right height to obtain a properly centered differential after each shift.
  • When the input returns, the open ratchet needs to re-engage with the 24t side of the differential only after the 3rd tooth passes the other ratchet. If the open ratchet re-engages too early, the differential will not return the required 45 degree (3 teeth). To prevent the open ratchet from closing too early, the orange (0X)-piece which controls the ratchets needs to sit tight against the ratchets, even in its centered (idle) position. That way the ratchet will only be fully re-engaged once the (0X)-piece points straight up again. I used the reddish brown 3L axles with stop to bring the ratchet close enough to the (0X)-piece.
  • When the output is blocked, for instance in case of a 4th-to-1st gear block in a sequential gearbox, the ratchets should not slip, not even when a Servo-motor is used as input. This means an engaged ratchet should firmly hook on to the differential to keep it put. The 12t gears do a very good job in that respect.

I also tried with a 28t differential and an 8t gear as ratchet, but a 45 degree return would translate into 3.5 teeth, which would practically result in a 3 or 4 teeth return and thus a biased starting point for the next shift.

Instruction on Rebrickable.

Edited by Didumos69

Share this post


Link to post
Share on other sites

Very nice. A while ago i was wondering about using 2 ratchets and releasing one somehow for each directional input - i never got any further with the idea! This is a great solution, and one i am likely to borrow in my next moc - i will try to remember to credit this design!!

Share this post


Link to post
Share on other sites

As far as I understand you can somehow use these pieces instead of 12z gears. The goal is to limit the differential to have one-way rotation only, right?

4106469

Share this post


Link to post
Share on other sites
3 hours ago, ibessonov said:

As far as I understand you can somehow use these pieces instead of 12z gears. The goal is to limit the differential to have one-way rotation only, right?

No, you cannot. I tried using that part and it locks the differential in two directions; it hooks on to the differantial too low to make enough difference between locking and slipping. There must be other configurations possible, but it all comes very precise. Here are a few requirements that need to be met:

  • After a return, when the differential is idle and locked by both ratchets, the exact position where the ratchets hook on to the differential should be such that one tooth of the 24t side of the differential is pointing straight up. This is needed to get equal starting points for shifts in both directions. The previous shift should not put one direction in favor of the other by leaving a slighlty off-center differential. The 12t gears hook on to the differential at exactly the right height to obtain a properly centered differential after each shift.
  • When the input returns, the open ratchet needs to re-engage with the 24t side of the differential only after the 3rd tooth passes the other ratchet. If the open ratchet re-engages too early, the differential will not return the required 45 degree (3 teeth). To prevent the open ratchet from closing too early, the orange (0X)-piece which controls the ratchets needs to sit tight against the ratchets, even in its centered (idle) position. That way the ratchet will only be fully re-engaged once the (0X)-piece points straight up again. I used the reddish brown 3L axles with stop to bring the ratchet close enough to the (0X)-piece.
  • When the output is blocked, for instance in case of a 4th-to-1st gear lock in a sequential gearbox, the ratchets should not slip, not even when a Servo-motor is used as input. This means an engaged ratchet should really hook on to the differential. The 12t gears do a very good job in that respect.

I also tried with a 28t differential and an 8t gear as ratchet, but a 45 degree return would translate into 3.5 teeth, which would practically result in a 3 or 4 teeth return and thus a biased starting point for the next shift.

Edited by Didumos69

Share this post


Link to post
Share on other sites

I'd enjoy building this as a proof of concept to appreciate the mechanism but noob question alert...

This needs to be the older differential with 16/24 teeth right?

Share this post


Link to post
Share on other sites
Just now, ukbajadave said:

This needs to be the older differential with 16/24 teeth right?

Yes, it needs to be the differential with 16/24 teeth. A full parts list is available on Rebrickable.

Edited by Didumos69
typo

Share this post


Link to post
Share on other sites

Kudos, great idea. :thumbup:  Can/will you use it in the diagonal drive project? Size matters I suppose. 

 

Edit: I've built it and tested. For me it tends to fail as it meets some resistance from the output side.

Edit 2.Hooked it up on a gearbox with the rotary catch and having a hard time around the 4th gear on a loaded drivetrain. Could be just this particular setup though. Did you run any test under load?

Edited by Attika

Share this post


Link to post
Share on other sites
50 minutes ago, Attika said:

Kudos, great idea. :thumbup:  Can/will you use it in the diagonal drive project? Size matters I suppose. 

 

Edit: I've built it and tested. For me it tends to fail as it meets some resistance from the output side.

Edit 2.Hooked it up on a gearbox with the rotary catch and having a hard time around the 4th gear on a loaded drivetrain. Could be just this particular setup though. Did you run any test under load?

It is because of my diagonal drive project that I am in pursuit of a reliable stepper, but I didn't test whether it generates enough torque to shift back from 4th to 3rd gear.

In what sense does it fail? In my setup the output is about as strong as the servo itself. Only little power is needed to push a ratchet aside. In my manual video I show what happens when I block the output: the input is blocked too.

What if you test your setup with a servo directly attached to your gearbox?

Share this post


Link to post
Share on other sites
Just now, Didumos69 said:

It is because of my diagonal drive project that I am in pursuit of a reliable stepper, but I didn't test whether it generates enough torque to shift back from 4th to 3rd gear.

In what sense does it fail? In my setup the output is about as strong as the servo itself. Only little power is needed to push a ratchet aside. In my manual video I show what happens when I block the output: the input is blocked too.

What if you test your setup with a servo directly attached to your gearbox?

As I figured, the reason is the number of connections. On the slack of the bevels, and lego being plastic, it looses some degrees on the output. The loss is proportionate to the resistace applied. You can try yours without gearbox. Just apply some resistance with your thumb on the output saft and see if it can complete the gearchange. 

Nevertheless it is a remarkable concept. I should say: Not bad for a prototype. :sweet: I've got something to play with this evening.

Share this post


Link to post
Share on other sites
1 hour ago, Attika said:

As I figured, the reason is the number of connections. On the slack of the bevels, and lego being plastic, it looses some degrees on the output. The loss is proportionate to the resistace applied.

I don't really recognize this, but there certainly is a limit to the amount of resistance it can handle. I will have to see whether this is going to help me further with my diagonal drive. The video below gives some confidence. If it won't do the job, then I'm afraid a Servo-motor will never give enough torque to make that shift from 4th gear back to 3rd gear.

Edited by Didumos69

Share this post


Link to post
Share on other sites
19 minutes ago, Didumos69 said:

I don't really recognize this, but there certainly is a limit to the amount of resistance it can handle. I will have to see whether this is going to help me further with my diagonal drive. The video below gives some confidence. If it won't do the job, then I'm afraid a Servo-motor will never give enough torque to make that shift from 4th gear back to 3rd gear.

Convincing experiment, very out of the box kind as well. :grin: I spent some time to find the source of the anomaly. I had the gearbox driven by an L motor. I held my finger lightly on the central diff to imitate the resistance that coming back from the drivetrain. All I can think as the source of the problem after your latest test is the lenght of the axle between the stepper and the rotary catch. In my case it was 8 stud. That might be unrealistic in a proper build however. 

Looking forward to see what comes out of this idea and sorry if my doubts were holding you back. In the other hand the world would have been poorer by a video where a supercar is pushed and pulled with a freshly designed seq. stepper. :wink: It must be one of a kind...

Share this post


Link to post
Share on other sites

I like this design, it seems very reliable. Do you have any concerns about the shifting time? Does it vary a lot with the different voltages available from the BuWizz unit? 

Share this post


Link to post
Share on other sites
9 hours ago, Attika said:

Convincing experiment, very out of the box kind as well. :grin: I spent some time to find the source of the anomaly. I had the gearbox driven by an L motor. I held my finger lightly on the central diff to imitate the resistance that coming back from the drivetrain. All I can think as the source of the problem after your latest test is the lenght of the axle between the stepper and the rotary catch. In my case it was 8 stud. That might be unrealistic in a proper build however. 

Looking forward to see what comes out of this idea and sorry if my doubts were holding you back. In the other hand the world would have been poorer by a video where a supercar is pushed and pulled with a freshly designed seq. stepper. :wink: It must be one of a kind...

You're thoughts a hands-on experiences are always welcome!

Next step in my dual diagonal project will be to skip the stepper and control the rotary catches with the servo directly, making sure I cover 2nd, 3rd and 4th gear. I can do that without many changes and it will test whether the servo is strong enough at all. If it works, I will try to build this new stepper in.

3 hours ago, BusterHaus said:

I like this design, it seems very reliable. Do you have any concerns about the shifting time? Does it vary a lot with the different voltages available from the BuWizz unit? 

No it doesn't vary a lot, only in extreme cases like the video above.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

  • Recently Browsing   0 members

    No registered users viewing this page.