Searching \ for 'driving servos' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: techref.massmind.org/techref/io/servo/index.htm?key=servo
Search entire site for: 'driving servos'.

Truncated match.
PICList Thread
'driving servos'
1998\05\04@044759 by Justin Grimm

flavicon
face
Hi all
When using PWM to drive servos (hs-300), I know you need a pulse width
of 1mS to 2mS, but do the pulses "have" to be 20mS apart or can the be
less,
say 1mS?
Thanks in advance.

Justin Grimm    spam_OUTreaperTakeThisOuTspamsouthwest.com.au

1998\05\04@080630 by Bill Cornutt

flavicon
face
----------
> Hi all
> When using PWM to drive servos (hs-300), I know you need a pulse width
> of 1mS to 2mS, but do the pulses "have" to be 20mS apart or can the be
> less,
> say 1mS?
> Thanks in advance.
>
> Justin Grimm    .....reaperKILLspamspam@spam@southwest.com.au
>

Justin,

Back in my 'early years' I had an expierence that shaped
my future.  While driving through the country with the
high school driving instructor we came upon a horse
laying down in a field.  The question arose as to if
the horse was sleeping or dead.  As we students
in the car could not come to a conclusion, the instructor
told me that "It will be your job Bill to find out if
horses sleep laying down."

I took this as an 'official' assignment, and next week
when our group again went driving, I was ready with my
report.  I was so proud of myself!

You have an inquiring mind and question established values.
This is a good trait.  It will take you far.  You will learn
much and advance the knowledge of computers. And for this I
thank you in advance.

But you may need a little 'push' on this track.  And here it
comes.

Justin, I challenge you to discover what happens when the
pulse rate is increased from 50 cycles and also what happens
when the pulse rate is decreased from 50 cycles.  You may
be on to something here.  You can become one of the PicList's
servo person's.  You may even prepare a little paper, with
graphs and such, for the benifit of the rest of us.  And create
a web page for the advancement of the art.  When you do this
and later as you accept the many awards that you earn, I
hope you will acknowledge that "Bill got me started on this
path."

I know that I'm extreamly proud of my knowledge of how
horses and cows sleep.  (I expanded the origional
research to include cows and can now tell a tired cow
from a sleeping horse by the way they are laying.)
And I always thank Mr. Thorton for stsrting me in
the right direction.

As to the question as to "why 50 cycles?", I can offer
one possibile answer.

Servos are like Nuns, they always travel in groups.
And as the pulse rate increases, the time to control
additional servos decreases.

I will be awaitting your results Justin, "if you
should accept this assignment".

Bill C.    billspamKILLspamcornutt.com

1998\05\04@082952 by Alvaro Deibe Diaz

flavicon
face
In an attempt to get more power from a standard Futaba servo, I made a
pulse train of 10ms period. The servo worked OK, and with a little more
torque (not double). I don`t know if you can do this with any kind of
servos... but I think that standard servos works on the same basis: wait a
starting pulse edge, internally generate a 1.5ms pulse, and finally do a
substraction between them. The result goes to the motor trough some power
electronics involving possibly a pulse-stretcher. I think that there is a
limit in the frequency of the incoming pulses due to this stretcher stage,
an the limit is in the point where the stretched pulses to the motor join
together.

In fact, I was trying to do a device to mix two standard RC channels (20ms
period) and get the two mixed outputs "turbo-charged" to 10ms... finally I
put the project apart due to some reasons, one of them being not to be sure
how to do this with a _say_ 16C73, with all his PWM generators, Caputre
modules, and running at 20MHz. ?Any help here?

Hope this helps.
-----Mensaje original-----
De:     Justin Grimm [SMTP:.....reaperKILLspamspam.....SOUTHWEST.COM.AU]
Enviado el:     lunes 4 de mayo de 1998 11:00
Para:   EraseMEPICLISTspam_OUTspamTakeThisOuTMITVMA.MIT.EDU
Asunto: driving servos

Hi all
When using PWM to drive servos (hs-300), I know you need a pulse width
of 1mS to 2mS, but do the pulses "have" to be 20mS apart or can the be
less,
say 1mS?
Thanks in advance.

Justin Grimm    reaperspamspam_OUTsouthwest.com.au

1998\05\04@093107 by Andy Kunz

flavicon
face
>When using PWM to drive servos (hs-300), I know you need a pulse width
>of 1mS to 2mS, but do the pulses "have" to be 20mS apart or can the be
>less,
>say 1mS?

Justin,

They only need to be about 6mS apart ON AVERAGE.  It really varies
depending on brand, model, etc.  The only way to know for sure is to try
and see.

If you go too fast, the servos get jittery.  It's a function of both the
internal RC circuit and the motor and gearing, as well as the pot.

Bottom line - play around.  6mS should be all you need, though.

Andy


==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

1998\05\04@095902 by Andy Kunz

flavicon
face
>how to do this with a _say_ 16C73, with all his PWM generators, Caputre
>modules, and running at 20MHz. ?Any help here?

I did this already.

I used a '73 with a GAL16V8 (I think, or 22V10) configured to allow me to
switch the PWM pins between input and output.  It worked nicely, gave 1uS
resolution using a 4MHz ceramic resonator.

Andy

==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

1998\05\04@124943 by wwl

picon face
On Mon, 4 May 1998 16:59:49 +0800, you wrote:

>Hi all
>When using PWM to drive servos (hs-300), I know you need a pulse width
>of 1mS to 2mS, but do the pulses "have" to be 20mS apart or can the be
>less,
>say 1mS?
They need to be around 20ms, certainly 1ms is too short. I have a
simple circuit to allow RC servos to be driven from the 16C74 PWM
port. let me know if you're interested.
    ____                                                           ____
  _/ L_/  Mike Harrison / White Wing Logic / @spam@wwlKILLspamspamnetcomuk.co.uk  _/ L_/
_/ W_/  Hardware & Software design / PCB Design / Consultancy  _/ W_/
/_W_/  Industrial / Computer Peripherals / Hazardous Area      /_W_/

1998\05\04@164848 by Jonathan

flavicon
face
I have code for driving two servos from a 16F84 (interrupt
driven).  It is on my web page under the BiMo robot,
http://www.ee.calpoly.edu/~jcline/labbook.htm
(I actually used hs-300 servos).



{Quote hidden}

1998\05\05@110927 by Justin Grimm

flavicon
face
Thanks for everyone's interest,
The reason I need to know if you can change a servo's pulsewidth is
because I dont want to run the code off interrupts. I am writing a program
that accepts a serial input (pic84 which will use the interrupts) then
drives
each servo (4) from that serial information.
I want to run the servo's at an approx 16mS pulsewidth
using a delay routine, but if the 84 receives a serial input while that's
happening the 16mS pulsewidth could be stretched a bit.
What I really need to know is-
is if the servo can handle a varying pulsewidth, say
from 16mS to 22mS, varying all the time but with the duty cycle always
remaining accurate.

BTW Bill, Servos can sleep lying down OR standing up. Clever hey!
Regards
Justin Grimm    RemoveMEreaperTakeThisOuTspamsouthwest.com.au


----------
{Quote hidden}

stage,
> an the limit is in the point where the stretched pulses to the motor join
> together.
>
> In fact, I was trying to do a device to mix two standard RC channels
(20ms
> period) and get the two mixed outputs "turbo-charged" to 10ms... finally
I
> put the project apart due to some reasons, one of them being not to be
sure
{Quote hidden}

1998\05\05@185927 by Andy Kunz

flavicon
face
>What I really need to know is-
>is if the servo can handle a varying pulsewidth, say
>from 16mS to 22mS, varying all the time but with the duty cycle always
>remaining accurate.

No problem.  The torque just will vary (and allow flutter in a dynamic
environment).

Andy

==================================================================
Andy Kunz - Statistical Research, Inc. - Westfield, New Jersey USA
==================================================================

1998\05\05@194608 by Eric Naus

picon face
Check out my homepage at http://webhome.idirect.com/~bine/  .

I have already done that to control my walker robot. Source code

is included and a test program is also included.

Have Fun

Eric

More... (looser matching)
- Last day of these posts
- In 1998 , 1999 only
- Today
- New search...