For better walking

[dangerousthing]

I've been viewing the videos of this wonderful beast, and I've noticed some solvable problems with the hip joint.

In the human body, this joint has a lot to do and it has powerful muscles. It looks like strength is even more needed in this bot. The hip joints have to take almost the full weight of the bot. This is needed for walking, getting up, getting down. and balancing.

My guess would be a motor with the same speed and twice the torque.

[danic85]

Hi, thanks for posting!

Yes torque is definitely a concern if we were working towards making this thing walk, as the entire weight of the robot would need to be on one leg during some of the motion.

At the moment, with both legs firmly on the ground, the stall torque of these servos isn't even close to being reached from what I can see during testing. But if we eventually do hit that issue we can upgrade for a different standard servo without (hopefully) any modification to the 3D prints.

Great to see you in the community, would you like to introduce yourself? I don't recognise the username so apologies if we have already spoken on another platform.

[dangerousthing]

I don't know if there is a better place for this, so I'll do it here.

I'm D. Jay Newman (I use Jay). I'm 65 years old. I've worked on building hobby robotics from about 2010 through 2015. I'm starting to pick them up again now. Professionally, I've worked in academia until around 2012, when I resigned to be the full-time caregiver for my late wife.

I've written one book, Linux Robotics. Frankly, it is mostly because halfway through, my wife became sick. I've also written articles for robot magazines, like Servo and one other.

I suffer from Ulnar-tunnel syndrome. I should get the Charachorder One (a chording keyboard which should be better for my hands.

I am glad to be here.

[SentryCoderDev]

welcome to discussion @dangerousthing i am Emir 17 years old developer i am from turkey and i am a contributor in this repository. I had done some research on this subject and the idea of ​​an electromagnet came to my mind. Before I took a step, the magnet on his foot activated and prevented it from falling, just like astronaut suits.
If your opinion doesn't make sense, this idea may be considered.

[dangerousthing]

That seems interesting, but what if the bot isn't walking on a surface that doesn't attract magnets? Most surfaces that I use to run a robot are non-magnetic, like floors and wooden tables.

Unless I am misunderstanding your comment, Emir.

[SentryCoderDev]

yes, you are right about that, it was just an alternative solution. A metal surface is needed for the idea, which means walking in a very limited area, 3d model change or as you said, different torque motors and the weight balance needs to be rearranged in the hip part.

[danic85]

Good to meet you! Yes we should probably have an 'introductions' area. I'll add something.

[dangerousthing]

Hi, thanks for posting!

Yes torque is definitely a concern if we were working towards making this thing walk, as the entire weight of the robot would need to be on one leg during some of the motion.

At the moment, with both legs firmly on the ground, the stall torque of these servos isn't even close to being reached from what I can see during testing. But if we eventually do hit that issue we can upgrade for a different standard servo without (hopefully) any modification to the 3D prints.

Great to see you in the community, would you like to introduce yourself? I don't recognise the username so apologies if we have already spoken on another platform.

[dangerousthing]

About hip joint. When it is standing upright isn't the maximum stress on the hip. I believe that it's maximum is probably when the robot is standing up from a squat or even worse, wants to stay a little bit squatted for some reason.

Perhaps stepper motors might be better, as might BLDC motors. Both of these motors have more torque at low speed.

[Chompalot]

If you supported the weight of the robot at the point where it normally rests, with for example springs would that reduce the torque required by the servo?

[SentryCoderDev]

I think yes the closer it is to 90 degrees the less strain the servos will have

[danic85]

Yeah you could definitely add a mechanism like that to take some of the load... You can actually calculate how much weight your servos can handle before reaching stall torque so this doesn't need to be an unknown before building a design. I think I covered this in one of the videos, if that's helpful.

[SentryCoderDev]

@Chompalot a system like spring in my mind i need some time but i was thinking of doing it too very good idea keep it up

[jsgroobotdream]

what about a motorized wheel in the body with enough mass(1) and a foully controlled dc motor(2) that by inertia and change of speed of the wheel can create enough time on one legg to move the other (see image)

[danic85]

Definitely an option. I'd like to see this tested 👏