Servo setup for H3-Walker(hexapod)robot

[ROBO-DON]

I am looking for the servo setup for the H3-Walker(hexapod)robot code on the NewMicros download page. Does anyone know where all 18 servo connections go on the ServoPod.

thanks,
ROBO-DON

[RMDumse]

The code is on the downloads page under "H3-Walker(hexapod)robot code" and in the code is the ordering of the servos.

( 2,1,0 9,10,11
( 5,4,3 6,7,8
( 17,18,19 14,15,16


*So the left front shoulder, forward and aft, is PWM0,
the left front shoulder, up and down, is PWM1,
the left knee, up and down, is PWM2,

The right front shoulder, forward and aft, is PWM9,
the right front shoulder, up and down, is PWM10,
the right knee, up and down, is PWM11,
etc.

The PWM run from 0 to 11. After that the timers are numbered
TMRA0 is 12
TMRA1 is 13
TMRA2 is 14
TMRA3 is 15

TMRB0 is 16
TMRB1 is 17
TMRB2 is 18
TMRB3 is 19

Not use, but the further timers would be

TMRC0 is 20
TMRC1 is 21

TMRD0 is 22
TMRD1 is 23
TMRD2 is 24
TMRD3 is 25

* Note: In the code left front shoulder, forward and aft, is called the shoulder, the left front shoulder, up and down, is called the knee, and the left knee, up and down, is called the leg tip.

[ROBO-DON]

Why does my servopod lock up when it gets to the halfspeed code. I can understand that the baud is going from 9600 to 4800, but shouldn't it do that after I load the code in NMITerm?
Here is what it does. . . . . .



IsoMax V0.6
HEX OK
YELLED OFF OK
: HALFSPEEDCPU F413 FA1 ! 82 F00 ! ; ( DIV BY 2
HALFSPEEDCPU NOT UNIQUE OK
OK
HALFSPEEDCPU ˜˜ž˜æ€˜€†




Originally posted by RMDumse
The code is on the downloads page under "H3-Walker(hexapod)robot code" and in the code is the ordering of the servos.

( 2,1,0 9,10,11
( 5,4,3 6,7,8
( 17,18,19 14,15,16


*So the left front shoulder, forward and aft, is PWM0,
the left front shoulder, up and down, is PWM1,
the left knee, up and down, is PWM2,

The right front shoulder, forward and aft, is PWM9,
the right front shoulder, up and down, is PWM10,
the right knee, up and down, is PWM11,
etc.

The PWM run from 0 to 11. After that the timers are numbered
TMRA0 is 12
TMRA1 is 13
TMRA2 is 14
TMRA3 is 15

TMRB0 is 16
TMRB1 is 17
TMRB2 is 18
TMRB3 is 19

Not use, but the further timers would be

TMRC0 is 20
TMRC1 is 21

TMRD0 is 22
TMRD1 is 23
TMRD2 is 24
TMRD3 is 25

* Note: In the code left front shoulder, forward and aft, is called the shoulder, the left front shoulder, up and down, is called the knee, and the left knee, up and down, is called the leg tip.

[nmitech]

The problem is the I/O base address register of the ServoPod and IsoPod are slightly different. I just revised the program on the download page so you can download it on ServoPod as well. Please read the comment section in the program

http://www.newmicros.com/isopod/appnotes/H3walker.txt

Randy, please let me know if I missed anything from the program

LC

[ROBO-DON]

Thanks so much for fixing that I/O problem.
I have some input to add . . . . .servoes 14 and 16 are switched and I cannot get any leg servoes to move. Are they suposed to?
Thanks,

[RMDumse]

Originally posted by ROBO-DON
I cannot get any leg servoes to move. Are they suposed to?

Yes, they should move. Well, after the program is downloaded, the last thing it is autostarted. So you have to reset or power cycle, or tell it to start by hand. But then every time it powers up, it will start into the preprogrammed tripod gate. To start it by hand, you'd just enter the activating word which is also autostarted, that is STARTUP in this program.

Do you have power supplied for the servos through the power rail? You should not rely on the onboard 5V for servo power. (You might be able to run one or two small ones, but there's only at most a about a hundred mills available there. 18 servos will require amps when moving).

[ROBO-DON]

Originally posted by RMDumse
Yes, they should move.
Do you have power supplied for the servos through the power rail? You should not rely on the onboard 5V for servo power. (You might be able to run one or two small ones, but there's only at most a about a hundred mills available there. 18 servos will require amps when moving).

I am using a 5 Cell sub C size NiCd battery (1700mA) and it works great. I have it running to the ServoPod voltage input and it runs flawlessly.
All shoulder and knee servos run great, but the legs do not move at all. I can tell that they are getting a signal because they are locked in about the middle position.
Thanks for all the info on getting started.

[Pasan]

Hello,

I am very interested in purchasing your servo DSP board.
My application will be using this board to control a biped robot.
I need to control 21 rc-servos and take in signal (analog) from 8 Force sensing resistors and also analog signal from a tilt sensor.

I used ASC16 and its resolution is quite good, i.e., it divides up the rotation angle into 4000 points. I ran into problem after it broke and they discontinued their product.

I wonder, what is the servo resolution of your board.

Thank you very much.
-Pasan

[RMDumse]

I'm not sure exactly what the question you are asking is, however I think you might be asking how carefully can we control the pulse width going to an RC Servo. The output of the RC Servo signals is much finer than the RC Servo itself can repeatedly assume, due to their internal construction and deadband. However, since the RC Servo range is over 1 millisecond, we can generate 2500 steps over that one millisecond range. This accounts for typically 90 degrees movement on the output horn, and many walking applications used almost 180 degrees of motion, so over the full movement of the servo, the steps would be about twice that.

In case I misunderstood, I will also mention the A/D's are 12-bit so they can resolve 4096 values.