RoboSavvy Forum

Parallel Leg Mechanism Design Discussion

2012-05-09T03:59:17+01:00

Hi Tyberius,
You said Plm design give more torque.
Can i use 3 Ax12+ in each joint to get more torque ?
Like 3 Ax12 in knee joint and 6 in ankle joint each leg

Statistics: Posted by MOHIT JINDAL — Wed May 09, 2012 3:59 am

Parallel Leg Mechanism Design Discussion

2011-10-25T17:09:23+01:00

I'm using bushes/mufs call them what you want. I've seen that most Japanese builders use them too. Other than that I've seen bearings(3x7x3) with a(1mm) flange that do the trick too

Statistics: Posted by mog123 — Tue Oct 25, 2011 5:09 pm

Parallel Leg Mechanism Design Discussion

2011-10-25T04:34:12+01:00

Apologies for digging up an oldish thread... but I'm just now getting (back) into the hobby, and have stumbled across this very helpful info.

I discovered parallel linkage legs in Hitec, which did very well in this year's Robo-One contest with only 11 servos. That rocks. I'd like to do the same thing using RoboBuilder servos (mainly because I already have a lot of them — though I'm tempted to try the new UpTech servos too, because they're slightly stronger and cheaper).

So I'm starting the slow process of learning to design and fabricate my own brackets. Most of this is straightforward, if not easy for a software geek like me. But one thing I still don't know:

What are you guys using for the bearings in these things?

Seems to me that one leg has about 16 of these, and they need to be fairly small... and they're load-bearing, too. Is this just some combination of standard screws and washers, or are you using something more sophisticated?

Thanks,
- Joe

Statistics: Posted by Joe — Tue Oct 25, 2011 4:34 am

Parallel Leg Mechanism Design Discussion

2011-05-12T17:04:21+01:00

clockwork wrote:
This thread in the forum was a key reason I wanted to try a PLM design.
I've finished most of the fabrication for the lower body - but still have a ways to go on the rest of the robot.

Just have to say, this is absolutely gorgeous. Can't wait to see more!

Statistics: Posted by Tyberius — Thu May 12, 2011 5:04 pm

Parallel Leg Mechanism Design Discussion

2011-04-30T01:33:41+01:00

This thread in the forum was a key reason I wanted to try a PLM design.
I've finished most of the fabrication for the lower body - but still have a ways to go on the rest of the robot.

Statistics: Posted by clockwork — Sat Apr 30, 2011 1:33 am

Parallel Leg Mechanism Design Discussion

2011-02-06T00:13:59+01:00

I've actually noticed something about IGNIS's design. It doesn't use 2 servos for bending the calf/hip. I don't see the sense in it's PLM design then. Isn't the idea to use 2 servos on a Y-cable to remove backlash and increase torque?

Statistics: Posted by mog123 — Sun Feb 06, 2011 12:13 am

Parallel Leg Mechanism Design Discussion

2011-01-11T03:17:22+01:00

Ausrobo wrote:
Tyberius,

Work on my parallel leg design is moving very slowly. I still have a number of parts to make. I am very impressed with your latest design.

While doing a little testing it occurred to me that as the leg part is driven from both ends there is no real need for the parallel structure. The servos, by being in master/slave configuration, effectively produce the same movement as the introduction of the parallel element. Have you considered this in your design?

Peter

You don't have the option of offsetting the two paired servos slightly to reduce/eliminate backlash however, nor the physical structure to support that.

Statistics: Posted by Tyberius — Tue Jan 11, 2011 3:17 am

Parallel Leg Mechanism Design Discussion

2010-11-18T06:50:42+01:00

Tyberius,

Work on my parallel leg design is moving very slowly. I still have a number of parts to make. I am very impressed with your latest design.

While doing a little testing it occurred to me that as the leg part is driven from both ends there is no real need for the parallel structure. The servos, by being in master/slave configuration, effectively produce the same movement as the introduction of the parallel element. Have you considered this in your design?

Peter

Statistics: Posted by Ausrobo — Thu Nov 18, 2010 6:50 am

Parallel Leg Mechanism Design Discussion

2010-10-12T08:13:37+01:00

Heres another PLM humanoid I'm working on for the IRC2010 competition later this month. Uses 21x RX-24F and a new, unreleased Vanadium Labs ArbotiX controller. Not named as of yet:

Leg closeup:

Statistics: Posted by Tyberius — Tue Oct 12, 2010 8:13 am

Parallel Leg Mechanism Design Discussion

2010-10-09T17:03:10+01:00

Funny. I tought about tube yesterday. Tubes to join the servos together actually.

Statistics: Posted by UncleBob — Sat Oct 09, 2010 5:03 pm

Servo Hubs

2010-10-09T08:13:16+01:00

Andrew,

I have been working on building the hubs that link the carbon fibre tubes to the servos. I am pleased with the first version, however it took way to long to make. I will need 32 of these, so I need a more streamlined way of producing them. The second version was faster but I am concerned if it will be robust enough. Off to design something in between.

Cheers,

Peter

Statistics: Posted by Ausrobo — Sat Oct 09, 2010 8:13 am

Parallel Leg Mechanism Design Discussion

2010-10-01T18:00:13+01:00

Ausrobo wrote:
Tyberius,

I have been following this forum for some time and your design for parallel legs using dual servos has inspired me to come up with my own. I have been working with a Kondo KHR-3HV for some time and would like to continue to use that platform, however need to translate everything from Japanese is just a little tiresome. The Robotis platform looks pretty good.

My goal is to build a biped at around 80cm. I am thinking of legs of about 60cm. The design I have come up with is based on using carbon fibre tubes for the vertical elements in the legs and the servos, top and bottom for the mounting. Unfortunately my cad skill are not up your level and the plans are on paper.

I did have on question. I have been looking at the ankle and hip joints and am concerned that like the legs, they will need two servos each. Have you considered this on your design.
Also how is the design going, have you moved on to construction?
Peter

Hi Peter,

Would love to see your designs when you get them on a digital medium!

I have actually had the same thoughts about the ankle/hip joints. The way I have it designed currently, I could forgo using the current hip/ankle brackets and replace them with the brackets in the knees. Those brackets could then attach to a double EX-106 assembly quite easily. I had even set aside 4 additional knee brackets in my BOM specifically for this purpose.

My current plan was to try it out with a single EX-106+ in the ankle/hip roll positions, and then move on to the double ex-106+ configuration as needed. I'm fairly certain I will need that configuration to get the desired results however. I'm also looking at getting the size down a bit on my own, under 60cm total height so it's still viable for the kidsize Robocup league.

Statistics: Posted by Tyberius — Fri Oct 01, 2010 6:00 pm

Parallel Leg Mechanism Design Discussion

2010-10-01T09:47:45+01:00

Tyberius,

I have been following this forum for some time and your design for parallel legs using dual servos has inspired me to come up with my own. I have been working with a Kondo KHR-3HV for some time and would like to continue to use that platform, however need to translate everything from Japanese is just a little tiresome. The Robotis platform looks pretty good.

My goal is to build a biped at around 80cm. I am thinking of legs of about 60cm. The design I have come up with is based on using carbon fibre tubes for the vertical elements in the legs and the servos, top and bottom for the mounting. Unfortunately my cad skill are not up your level and the plans are on paper.

I did have on question. I have been looking at the ankle and hip joints and am concerned that like the legs, they will need two servos each. Have you considered this on your design.
Also how is the design going, have you moved on to construction?
Peter

Statistics: Posted by Ausrobo — Fri Oct 01, 2010 9:47 am

Parallel Leg Mechanism Design Discussion

2010-07-08T22:36:13+01:00

Stephane wrote:
I never feel very comfortable when I don't have both end of a servo axis attached to the other section (such as servos for neck, waist, hips, shoulders and elbow twisting in Tyberius design).
On the other hand, changing these joints is not something so easy.

With a big robot like yours, don't you worry that the load on such attached servo will wear them out faster?

(I am probably worrying for nothing, but I can't help thinking that I must find another way to do when I have this kind of joints, not very successfully so far.)

These particular joints are reinforced using a different type of horn and a thrust bearing beneath. The current build of Giger was at one point, 7.2kg in weight and about 2/3 the proposed size. It did not stress these joints in any considerable way. The Kung Fu variant that I competed with in Korea at KRGF was lightened up a good amount to just over 5kg, but was under quite a bit more 'wear' given the type of shuffle gaits and impacts I was taking in competition. This again didn't prove to be an issue for these joints. I think as long as they're placed in a manner that they don't have a vertical side loading force on them, it's fine.

Statistics: Posted by Tyberius — Thu Jul 08, 2010 10:36 pm

Parallel Leg Mechanism Design Discussion

2010-07-08T22:26:57+01:00

Tyberius wrote:
14 EX-106+ and 10 RX-64 + all the frame components and onboard controller, it certainly adds up quick. I do work for Trossen Robotics though, so I get a bit of a discount.

Finished up head design, 80.5cm tall including the head.

I envy you so much. I'd like to work in a company like that. I'm currently designing a humanoid under 400$ (http://xmog123x.blogspot.com/2010/06/im ... alive.html)

Statistics: Posted by mog123 — Thu Jul 08, 2010 10:26 pm