ID3h-r (Indirect-Drive-3DOF-Hexapod-round)

Lynxmotion Legged Robots
1

*Ok Well after a very kind arrangement with Jim i would like to create a follow up thread to the “Indirect-drive leg thread” on the progress of this
project.

First, before i start showing off “and after i become the envy of others” I’d like to take this opportunity to say a big Thanks to Jim and
Everyone at Lynxmotion for the support and sponsor of this project. And also the Influence and support from the Lynxmotion community.*

OK well its been a very interesting build and fun to do. its great to see my design come to life.
Ill start by Introducing the New …

[size=150]ID3h-r [/size]Hexapod!
(Indirect-Drive-3DOF-Hexapod round)

After unpacking a mountain of parts I soon got to work sorting out into workable groups.

Here are some of the leg designs appointed by for this project: (HS645MG’s)

**3DOF-A:**3" Phoenix femurs Aluminium Tubing - 3.0"
http://i531.photobucket.com/albums/dd355/innerbreed/jimsdesign.jpg

**3DOF-B:**4" Custom femurs Aluminium Tubing - 3.0"
http://i531.photobucket.com/albums/dd355/innerbreed/100_2299.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2295.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2294.jpg

**3DOF-C:**3" Phoenix femurs Aluminium Tubing - 4.50"
Image to be supplied soon. plus it would never be able to stand up properly due to the large sized tibia to femur. :frowning:

**3DOF-D:**4" Custom femurs Aluminium Tubing - 4.50"
http://i531.photobucket.com/albums/dd355/innerbreed/100_2298.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2296.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2297.jpg

For the picture demonstration i have chosen the 3DOF-B: mainly as i have found it to give the best overall performance regarding Step length, body IK functions and overall mobility. ref to converted phoenix/CH3r code - ch3r2ps2 (Thanks Xan/Kurte)
no video yet as i have ran into a few teething problems.

First here are some pictures of the ID3h-r
http://i531.photobucket.com/albums/dd355/innerbreed/100_2300.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2302.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2303.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2304.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2306.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2307.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2305.jpg

Ok i understand there are a few problems with converting the Phoenix code for the H3-r range. the problems im finding is that its Right Rear leg is responding differently then the others, plus im still working out the init start pos for the legs. i think some of the rotations need tweaking. im currently using the ch3r2ps2!

im also going to strengthen the rod how jim did with his arm design. those rods to bend if limits are exceeded!

2

Looks great!

As Zenta mentioned in the other thread, there have been issues with with variable and table overflows when converting the phoenix code to run on the CHR-3.

Things like: in phoenix_v20.bas

I needed to update the body offset tables (cOffsetX and cOffsetZ) from sbytetable to swordtable

When I did a quick file difference of the one on my machine and a fresh copy up on the web that is the main difference I see, other than a bunch of debug code to figure that out.

Good Luck
Kurt

3

That is a very cool hexapod with excellent ground clearance. I can’t wait to see video of it walking and performing. :smiley:

8-Dale

4

Hi,

That’s awesome! The slim and long legs looks very good, great to see it all together. The CH3 body was just perfect for this design, since the coxa part require some space while moving. Another part I really liked is that the wiring looks pretty easy to do.

Can’t wait to see it walking! I would recommend using Xan’s V20 phoenix code, its alot faster. :smiley:

btw: Congratz with all the new parts! :wink:

5

thanks for your comments. im currently using V2.0 here on this page:
viewtopic.php?f=8&t=5961&p=63997#p63997

yes the h3-r chassis does work very well. this says to me that it will also look quite cool in the In-line h3 chassis too.
jim did a great job on those 4" custom femurs. plenty of clearance. :wink:

im going to get time to work on it today so i should have something later today.

6

regarding the InitStartPos of the feet:
im having problems inputting the right numbers. for some reason when i measure the feet pos relative to the coxa and put these into the (InitPos) section, they never seem to be correct. the legs will either sit under the body or spread out?! :confused:

i have added here my dimensions for you to see. not sure if you can help me from this but…

[code];[BODY DIMENSIONS]
cCoxaLength con 36
cFemurLength con 102
cTibiaLength con 112

cRRCoxaAngle1 con -600 ;Default Coxa setup angle, decimals = 1
cRMCoxaAngle1 con 0 ;Default Coxa setup angle, decimals = 1
cRFCoxaAngle1 con 600 ;Default Coxa setup angle, decimals = 1
cLRCoxaAngle1 con -600 ;Default Coxa setup angle, decimals = 1
cLMCoxaAngle1 con 0 ;Default Coxa setup angle, decimals = 1
cLFCoxaAngle1 con 600 ;Default Coxa setup angle, decimals = 1

cRROffsetX con -69 ;Distance X from center of the body to the Right Rear coxa
cRROffsetZ con 119 ;Distance Z from center of the body to the Right Rear coxa
cRMOffsetX con -138 ;Distance X from center of the body to the Right Middle coxa
cRMOffsetZ con 0 ;Distance Z from center of the body to the Right Middle coxa
cRFOffsetX con -69 ;Distance X from center of the body to the Right Front coxa
cRFOffsetZ con -119 ;Distance Z from center of the body to the Right Front coxa

cLROffsetX con 69 ;Distance X from center of the body to the Left Rear coxa
cLROffsetZ con 119 ;Distance Z from center of the body to the Left Rear coxa
cLMOffsetX con 138 ;Distance X from center of the body to the Left Middle coxa
cLMOffsetZ con 0 ;Distance Z from center of the body to the Left Middle coxa
cLFOffsetX con 69 ;Distance X from center of the body to the Left Front coxa
cLFOffsetZ con -119 ;Distance Z from center of the body to the Left Front coxa[/code]

[code];[START POSITIONS FEET]
cRRInitPosX con ? ;Start positions of the Right Rear leg
cRRInitPosY con ?
cRRInitPosZ con ?

cRMInitPosX con ? ;Start positions of the Right Middle leg
cRMInitPosY con ?
cRMInitPosZ con 0

cRFInitPosX con ? ;Start positions of the Right Front leg
cRFInitPosY con ?
cRFInitPosZ con -?

cLRInitPosX con ? ;Start positions of the Left Rear leg
cLRInitPosY con ?
cLRInitPosZ con ?

cLMInitPosX con ? ;Start positions of the Left Middle leg
cLMInitPosY con ?
cLMInitPosZ con 0

cLFInitPosX con ? ;Start positions of the Left Front leg
cLFInitPosY con ?
cLFInitPosZ con -?[/code]

7

Hi Jonny,

Try these values. I just picked them up from PEP.

[code];[START POSITIONS FEET]
cRRInitPosX con 69 ;Start positions of the Right Rear leg
cRRInitPosY con 25
cRRInitPosZ con 120

cRMInitPosX con 138 ;Start positions of the Right Middle leg
cRMInitPosY con 25
cRMInitPosZ con 0

cRFInitPosX con 69 ;Start positions of the Right Front leg
cRFInitPosY con 25
cRFInitPosZ con -120

cLRInitPosX con 69 ;Start positions of the Left Rear leg
cLRInitPosY con 25
cLRInitPosZ con 120

cLMInitPosX con 138 ;Start positions of the Left Middle leg
cLMInitPosY con 25
cLMInitPosZ con 0

cLFInitPosX con 69 ;Start positions of the Left Front leg
cLFInitPosY con 25
cLFInitPosZ con -120[/code]

8

PEP… doh! i should have know better! :blush:

i will try them shortly. thanks for that Kåre. :wink:

i have checked my connections on that rear leg and all is ok. i then loaded up the old powerpod program to see if i was having the same problem with that rear leg. works ok in powerpod!

But it just doesn’t like phoenix code. :confused:

Thanks again. :wink:

9

i couldnt help myself, i had to pick it up again. ok, those values work perfectly.
im still yet to sort out the rear right tibia! and slightly adjust the pitch of the legs.

After a few tests it shows that some of the rods do bend, and even pop out if the threaded end of the socket! also some minor things like the ball joints unscrew them selves. a bit of lock tight i think, nothing bad!

to stop the rods from faulting in the areas mentioned above, i will tighten up the Min/Max values in the code, used lock tight on the threads and buy some thin aluminum tubing to feed over the rods for a more ridged assembly!

apart from that we are moving in the right direction! :wink:

10

What exactly is this leg doing different? Are you sure it is only this leg? i.e. if you move in different directions it is always this leg. Also from your previous comments I believe you have the version with the SBYTETABLE converted to SWORDTABLE.

If this is the case, I would check your CHR-3_cfg.bas file and check to make sure the mins and maxs for this servo look OK. We had a case earlier that on one servo the values were reversed, which caused a leg to not work right… On mine it looks like all of the tibias are defined like:

cRRTibiaMin1 con -400 cRRTibiaMax1 con 750

Not sure what else to suggest yet.

Kurt

11

hi kurt,

im using the converted swordTable version.
well i start up the hex to initialize it, and the rear right tibia swings up (the other move down to the intended pos). when i reverse the Min/Max values its worse but in the opposite direction and seems to ignore the M/M values and over shoots!

yes your Min/Max values are the same as mine. i have tried changing the values and also reversing them but its still not correct.

while trying to walk it also seems that it has more travel (over shoots) and is much faster. almost like it has further to travel, therefore its going quicker to get to each point. if you know what i mean.

i might try to plug in another servo in the pins and see if its doing the same.
iv put these in quotes so i can highlight them. you will see they look fine. :confused:

thanks for helping. :wink:

12

I will take a look and see if I see anything obvious.

So far I see we are using a different SSC configuration. My pins look like:

[code];[SSC PIN NUMBERS]
cRRCoxaPin con P0 ;Rear Right leg Hip Horizontal
cRRFemurPin con P1 ;Rear Right leg Hip Vertical
cRRTibiaPin con P2 ;Rear Right leg Knee

cRMCoxaPin con P4 ;Middle Right leg Hip Horizontal
cRMFemurPin con P5 ;Middle Right leg Hip Vertical
cRMTibiaPin con P6 ;Middle Right leg Knee

cRFCoxaPin con P8 ;Front Right leg Hip Horizontal
cRFFemurPin con P9 ;Front Right leg Hip Vertical
cRFTibiaPin con P10 ;Front Right leg Knee

cLRCoxaPin con P16 ;Rear Left leg Hip Horizontal
cLRFemurPin con P17 ;Rear Left leg Hip Vertical
cLRTibiaPin con P18 ;Rear Left leg Knee

cLMCoxaPin con P20 ;Middle Left leg Hip Horizontal
cLMFemurPin con P21 ;Middle Left leg Hip Vertical
cLMTibiaPin con P22 ;Middle Left leg Knee

cLFCoxaPin con P24 ;Front Left leg Hip Horizontal
cLFFemurPin con P25 ;Front Left leg Hip Vertical
cLFTibiaPin con P26 ;Front Left leg Knee[/code]

Will try to plug your other values in and see if anything looks obvious.

Kurt

13

ok thats great. i need to confirm its not just me being dumb!

iv also change the pin - still the same 'so
i change it the vals from
cRRTibiaMin1 con -400
cRRTibiaMax1 con 750

to
cRRTibiaMin1 con -750
cRRTibiaMax1 con 400

to
cRRTibiaMin1 con 400
cRRTibiaMax1 con -750

to
cRRTibiaMin1 con 750
cRRTibiaMax1 con -400

and no success! :unamused:
iv also tried another servo!
S T R A N G E?

14

its completely code related, i have uploaded v1.3 and everything is aligned fin.

its walking at least!
Unfortunately no video as of yet!

im going to order some new thicker/stronger threaded rod’s and ball joints. they seem to be very low in tolerance and cannot hold the strain. something to think about if anyone wished to build this.

15

Yep - sounds strange…

I hacked up some debug outputs that only output on legs 0 and 3 (which I think are the two rear legs. I attached the file. If you look through a bunch of stuff is under #ifdef DEBUG. Also some are #if 0 (which when previously debugging I output other stuff until I localized issue…)

But for the fun of it, you might try it and see if we can start to see where the outputs diverge… Currently on my CHR-3 config, once I hit the start button, I start to see outputs that look like:

I will try to change to your configuration info and see what I get, but I may be gone the rest of the afternoon, so you might give it a shot. You will need to look at the code to see what all of these values are, but at the end SWE shows the status as Solved, Warning, Error. If not 100 than start looking back. You will notice most of the values for the two legs are the same!

…
Kurt
phoenix_v20.bas (42.1 KB)

16

Quick update, I tried your config, so far at just going to idle dont see anything obvious:

L: 0 TotXZ: 210 -121 G4: 0 10000 B4: 0 10000 A4: 0 10000 BodyIK(0 0 0) = POS(52 80 91) IKA1: 6960 IKSW: 10371 IKA2 11580 => CA: 7 FA: -162 TA: -322 SWE: 100 L: 3 TotXZ: 210 121 G4: 0 10000 B4: 0 10000 A4: 0 10000 BodyIK(0 0 0) = POS(52 80 91) IKA1: 6960 IKSW: 10371 IKA2 11580 => CA: 7 FA: -162 TA: -322 SWE: 100

Kurt

17

That’s awesome!

:smiley:

18

thanks tom.
we are having a few mechanical problems as well as the known phoenix/h3r code conversions problems
As a prototype concept i like to make sure im getting as many problems with it as i can. so the improvements can be made for future builds.

At the moment we know that the legs are not as strong and the rods need reinforcing or changing. im popping out at some point today to pick up some new parts that i think will help.

Another very surprising thing is, is that its heavier than previous hexapods.
Weight (w/out batteries)
AH3 = 3lb 13oz.
Phoenix = 3lb 10oz.
ID3 = 4lb 10oz. :open_mouth:

the weight is making the rods pop out from their threads and also bending due to force when walking. (mainly during rotation)

just some more pictures:
http://i531.photobucket.com/albums/dd355/innerbreed/100_2309.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2310.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2314.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2317.jpg
http://i531.photobucket.com/albums/dd355/innerbreed/100_2318.jpg

19

Well it really looks cool anyway. lol :smiley:

Looks like we totally underestimated the loads on that joint.

Have you tried the aluminum tubing to strengthen the linkages?

Do they make a through hole dog bone in 2-56 like they do for 4-40? :blush:

20

unfortunately iv not had a chance to get any tubing or new links yet. a few problems at home. :frowning:
my thoughts are to look at getting some ball likes used in rc racing. they are very strong. not sure i will find any in 2-56 but it will need something strong. i will try to get some of the 4-40 thread and rods. will i still be able to use these with the baring hubs and metal servo horns or will i need to make a new part?