go away spammer

Author Topic: Hexapod Leg design - is there something wrong ?  (Read 7997 times)

0 Members and 1 Guest are viewing this topic.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Hexapod Leg design - is there something wrong ?
« on: February 27, 2010, 03:16:59 PM »
Hello, i just finished designing a leg for a hexapod with 3 degrees of freedom.

This is the first image, a 3d view of the leg:


This is the front view:


This is the top view: (i think the problem is here)


should square 2 (surrounding the leg with servo 1) be aligned along line 1?

I mean the servo that will be attached to the body of the hexapod being in line with leg shown in square2.

So does it have to be aligned or not?

Offline little-c

  • Robot Overlord
  • ****
  • Posts: 186
  • Helpful? 1
Re: Hexapod Leg design - is there something wrong ?
« Reply #1 on: February 27, 2010, 04:35:31 PM »
you should be able to have the leg offset from its pivot, just depends on if its going to hit anything. as long as the brackets can take the slight extra load. which should be fine unless your hexapod is heavy enough to be pushing the limits of its structure.

I can see no real reason why the leg can't be offest. maths is slightly more complicated. involves a few extra terms in the equation, but other than that, it should be fine.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #2 on: February 27, 2010, 11:09:37 PM »
Thanks little-c, i feel relieved, i thought i had to redesign or add any part.

Offline HyperNerd

  • Robot Overlord
  • ****
  • Posts: 139
  • Helpful? 5
  • N3rd1n80r
Re: Hexapod Leg design - is there something wrong ?
« Reply #3 on: February 28, 2010, 05:04:34 AM »
One thing I wish to add:

What material is the middle beam (thigh) made from?
If it's lexan / plexiglass, you could run into stability issues - that beam will wobble like a bitch.
I suggest adding an identical beam on the other side to brace it.

 -HyperNerd
There are 10 types of people in the world - those who understand binary, and those who don't.

Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
    • dunk's robot
Re: Hexapod Leg design - is there something wrong ?
« Reply #4 on: February 28, 2010, 05:49:24 AM »
i'm a little concerned about the strength of the servo where the servo horn (the round bit) joins the top servo gear.
with this design there would at best a lot of wear on the top bearing and at worst the joint between the servo horn ad top gear would snap.

if you go with HyperNerd's idea of a 2nd identical thigh but connect it to pivots in-line with the servo horns on the rear of the servo it will make the whole structure more stable.

a similar 2nd pivot pont on the "hip" servo will provide the same added strength there.


then again,
you have the servos. give one a wobble. how strong do they seem?


dunk.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #5 on: February 28, 2010, 07:31:08 AM »
Here are the specs of the servos:
Torque at 4.8V: 25.0kg.cm
Speed: 0.19sec/60 degrees
Dimension: 35 x 21 x 40mm
Weight: 69g

i'm a little concerned about the strength of the servo where the servo horn (the round bit) joins the top servo gear.
with this design there would at best a lot of wear on the top bearing and at worst the joint between the servo horn ad top gear would snap.
the servo horn( the round bit) is actually fixed to the beam, the body of the servo will rotate thus rotating the leg.

What material is the middle beam (thigh) made from?
i think i will use aluminum as its strong and light. Is this a good choice, given the specs of the servos ?


Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
    • dunk's robot
Re: Hexapod Leg design - is there something wrong ?
« Reply #6 on: February 28, 2010, 09:47:25 AM »
i'm a little concerned about the strength of the servo where the servo horn (the round bit) joins the top servo gear.
with this design there would at best a lot of wear on the top bearing and at worst the joint between the servo horn ad top gear would snap.
the servo horn( the round bit) is actually fixed to the beam, the body of the servo will rotate thus rotating the leg.
yes, but i'm concerned how strong the servo is.


dunk.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #7 on: February 28, 2010, 10:36:23 AM »
yes, but i'm concerned how strong the servo is.

Doesn't this show how strong the servo is (Torque: 25.0kg.cm) ? I am not really sure.

Offline waltr

  • Supreme Robot
  • *****
  • Posts: 1,944
  • Helpful? 99
Re: Hexapod Leg design - is there something wrong ?
« Reply #8 on: February 28, 2010, 11:54:54 AM »
I think Dunk means the physical strength of the servo output wheel, bearings and gears. Although one would think that the servo mechanicals can handle whatever torque the servo can produce.
As a gut feeling I too don't like the looks of the entire leg being supported by a single servo wheel but I have seen many examples on the web of hexapods built this way so it must work.

Great threads on your hex bilas.

Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
    • dunk's robot
Re: Hexapod Leg design - is there something wrong ?
« Reply #9 on: February 28, 2010, 12:14:44 PM »
I think Dunk means the physical strength of the servo output wheel, bearings and gears.
nearly.
i'm not concerned about the force round the servo axis.
the servo is designed to drive forces in this direction.
i am concerned about forces on the other axis's at right angles to the plane of normal rotation. (what's the plural of axis?)

for example,
the "hip" servo (the one on the left hand side of all your pictures),
when the servo horn is attached to the base of the bot will have a lot of force on the weakest part of the servo.
look at the 2nd picture and imagine the servo horn being which is to be attached to the body of the bot held in place while the "foot" is pushed upwards.

in a cheap servo i'd be concerned of warping or snapping of the output shaft of the servo.
on better quality servos i'd still wonder about excess pressure on the output shaft bearings decreasing life expectancy.


if my explanation still is not making any sense,
just grab your servo by the body and twist the horn from side to side (*not* round and round) and see if there's much bending or movement in directions other than the usual rotation.


dunk.

Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
    • dunk's robot
Re: Hexapod Leg design - is there something wrong ?
« Reply #10 on: February 28, 2010, 12:28:18 PM »
actually,
here's some pictures of solutions to the problem:
http://www.lynxmotion.com/images/jpg/asb033.jpg
http://www.lynxmotion.com/images/jpg/asb044.jpg

see how the servo is supported by a second pivot in line with the axis of the servo output shaft?

obviously it depends on the strength of your servo and the weight of your robot whether this will be an issue or not.


dunk.

Offline Razor Concepts

  • Supreme Robot
  • *****
  • Posts: 1,856
  • Helpful? 53
    • RazorConcepts
Re: Hexapod Leg design - is there something wrong ?
« Reply #11 on: February 28, 2010, 12:36:09 PM »
Either that, or use linkages (ball-links and threaded rod) to take the load off of the servo.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #12 on: February 28, 2010, 12:56:13 PM »
here's some pictures of solutions to the problem:
http://www.lynxmotion.com/images/jpg/asb033.jpg

Thanks for the hint, but do you mean implement this solution only to the servo to be attached to the body of the hex (servo on the left), or at least to that servo ?
« Last Edit: February 28, 2010, 01:06:00 PM by bilals »

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #13 on: February 28, 2010, 12:57:36 PM »
Either that, or use linkages (ball-links and threaded rod) to take the load off of the servo.

I am really new to this stuff, i have no idea what your talking about  :)

Offline waltr

  • Supreme Robot
  • *****
  • Posts: 1,944
  • Helpful? 99
Re: Hexapod Leg design - is there something wrong ?
« Reply #14 on: February 28, 2010, 01:28:45 PM »
Yep, that's what I was trying to say Dunk. The Lynx stuff is a good solution for standard servos.

The 'force' we're concerned about the the bending moment of the servo shaft.  Adding the second pivot point eliminates most of this bending force on the servo's shaft.


Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #15 on: February 28, 2010, 01:49:13 PM »
Yep, that's what I was trying to say Dunk. The Lynx stuff is a good solution for standard servos.

The 'force' we're concerned about the the bending moment of the servo shaft.  Adding the second pivot point eliminates most of this bending force on the servo's shaft.

So do you mean guys something like this (see pic):


where (1) is just a part of the body, and (2) is the second pivot point, aligned with the servo horn > goes into the body and is allowed to rotate. How to reduce the friction due to the rotation of (2) ?
« Last Edit: February 28, 2010, 01:55:35 PM by bilals »

Offline HyperNerd

  • Robot Overlord
  • ****
  • Posts: 139
  • Helpful? 5
  • N3rd1n80r
Re: Hexapod Leg design - is there something wrong ?
« Reply #16 on: February 28, 2010, 01:57:17 PM »
That's exactly what we mean :)

Looks pretty good now, but I'm still gonna rant about the beefiness of that middle beam ;D

 -HyperNerd
There are 10 types of people in the world - those who understand binary, and those who don't.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #17 on: February 28, 2010, 02:01:00 PM »
Thanks HyperNerd  ;D ;D ;D

Offline HyperNerd

  • Robot Overlord
  • ****
  • Posts: 139
  • Helpful? 5
  • N3rd1n80r
Re: Hexapod Leg design - is there something wrong ?
« Reply #18 on: February 28, 2010, 02:51:54 PM »
Just saw your edit..

To reduce the friction, use the bearings that the Lynxmotion C brackets come with. They fit inside the mounting holes on the C brackets, then you bolt them to the servo brackets (I think). Have a look on the website ;D

 -HyperNerd
There are 10 types of people in the world - those who understand binary, and those who don't.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #19 on: February 28, 2010, 03:26:25 PM »
To reduce the friction, use the bearings that the Lynxmotion C brackets come with.  -HyperNerd

I just saw the bearings, they look pretty nice + small. Any idea what would the bearing outer and inner diameters be ?
« Last Edit: February 28, 2010, 03:30:12 PM by bilals »

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #20 on: February 28, 2010, 03:29:14 PM »
Take a look at this:


Does the legs look proportional to the body? Does it look Good ?

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: Hexapod Leg design - is there something wrong ?
« Reply #21 on: February 28, 2010, 03:30:04 PM »
Hi,

Why is this in the electronics forum?  It ought to go in "Mechanics and Construction".
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline HyperNerd

  • Robot Overlord
  • ****
  • Posts: 139
  • Helpful? 5
  • N3rd1n80r
Re: Hexapod Leg design - is there something wrong ?
« Reply #22 on: February 28, 2010, 03:31:16 PM »
I'm sorry I have no idea - I have never used these brackets before, I've only looked at them on the website.

 -HyperNerd
There are 10 types of people in the world - those who understand binary, and those who don't.

Offline bilalsTopic starter

  • Full Member
  • ***
  • Posts: 81
  • Helpful? 0
Re: Hexapod Leg design - is there something wrong ?
« Reply #23 on: February 28, 2010, 03:32:12 PM »
I'm sorry I have no idea - I have never used these brackets before, I've only looked at them on the website.

Anyway, thanks for your help HyperNerd  ;)

Offline dunk

  • Expert Roboticist
  • Supreme Robot
  • *****
  • Posts: 1,086
  • Helpful? 21
    • dunk's robot
Re: Hexapod Leg design - is there something wrong ?
« Reply #24 on: February 28, 2010, 04:33:57 PM »
ok, that's the top servo on the leg fixed.

i'd still be concerned about the other 2.
it all depends how strong your servos are.


dunk.

Offline waltr

  • Supreme Robot
  • *****
  • Posts: 1,944
  • Helpful? 99
Re: Hexapod Leg design - is there something wrong ?
« Reply #25 on: February 28, 2010, 06:47:33 PM »
ok, that's the top servo on the leg fixed.

i'd still be concerned about the other 2.
it all depends how strong your servos are.


dunk.

And what the hexapod weights and how fast it is expected to move. F = ma

 


Get Your Ad Here

data_list