Searching \ for '[PIC] multi dc motor control' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page: www.piclist.com/techref/microchip/ios.htm?key=motor
Search entire site for: 'multi dc motor control'.

Exact match. Not showing close matches.
PICList Thread
'[PIC] multi dc motor control'
2008\07\23@175426 by threewheeler7

picon face

ok i have a top priority design that needs to control 4 motors at a time.
some of you may remember my post about using servos, i had to go with dc
brushed motors. looking at microchips AN893, i may use 4 PIC16F684 circuits
and leave what order they do what they do up to a master PIC. this is to
save development time (which is very critical here), and use it to develop
the master PIC firmware. Now a am left to decide how to talk to these pics(i
was thinking rs232 since thats what the sample firmware already dose) and
come up with a good scheme for ramping up and down the speed of the motors
in software. so i would like suggestions on this stuff for now if possible.
--
View this message in context: www.nabble.com/multi-dc-motor-control-tp18619878p18619878.html
Sent from the PIC - [PIC] mailing list archive at Nabble.com.

2008\07\23@194044 by Martin K

face
flavicon
face
threewheeler7 wrote:
> ok i have a top priority design that needs to control 4 motors at a time.
> some of you may remember my post about using servos, i had to go with dc
> brushed motors. looking at microchips AN893, i may use 4 PIC16F684 circuits
> and leave what order they do what they do up to a master PIC. this is to
> save development time (which is very critical here), and use it to develop
> the master PIC firmware. Now a am left to decide how to talk to these pics(i
> was thinking rs232 since thats what the sample firmware already dose) and
> come up with a good scheme for ramping up and down the speed of the motors
> in software. so i would like suggestions on this stuff for now if possible.
>  
Use SPI including chip-select lines. It's more reliable than RS232.
-
Martin

2008\07\23@194250 by Apptech

face
flavicon
face
> ok i have a top priority design that needs to control 4
> motors at a time.
> some of you may remember my post about using servos, i had
> to go with dc
> brushed motors.

If "had to" is related to other issues than control, such as
power or precision of control etc then DC motors may be a
better choice than servos. If servos can do the job then
controlling them is vastly easier than controlling DC motors
to do the same job. (eg multiple channels of eg  1 to 2 mS
pulse approximately every 20 mS is a far easier task than
multiple channels of continuous PWM). As always, knowing
more about the task would help get better answers. Very
importantly, are you aiming at position or speed control?
And, if position, static or dynamic. eg dynamic position
where you want to track a rotating target position would be
'lots of fun'.

> looking at microchips AN893, i may use 4 PIC16F684
> circuits
> and leave what order they do what they do up to a master
> PIC.

AN893 is fine if you need that level of complexity and don't
mind multiple processors and interprocessor digital
communications. If high power and high precision is involved
this may be an OK approach but it's overkill for what is
needed if you are aiming at something around what an RC
servo could achieve. Overkill can be good :-).

{Quote hidden}

RS232 is fine enough for command sending. If you want
responses as well its not formally specified for multidrop
on a single circuit but logical ORing of slave TXDs and
responses only to master polls will cure handle that as long
as you don't allow multiple simultaneous slave transmits.
Hardware wise RS422/485 is better over any distance. SPI may
be your friend with multiple slaves.

If I were doing this and time was of the essence I'd be
inclined to try and do it with a single processor as far as
the overall control was concerned. eg you could produce
software PWM outputs, one per motor. Then either use
discrete hardware to do actual motor control or if desirable
use one processor per motor driven by the PWM. These
processors are just "hardware replacement" items with no
significant feedback, no need to process digital commands
etc. PWM could even be turned into an analog signal with
filtering and the discrete hardware or slave processors then
simply match the motor result to that. Development is much
easier than using digital coms to multiple processors and
trying to handle real-time interaction bugs across a
multi-processor system. PWM speed should be easily fast
enough with soft PWM to filter to DC well.

If you filter PWM to DC and need slower PWM than a single
poll RC filter will handle without either ripple (small RC)
or slow response (big RC) then using eg a 3 or 4 pole formal
low pass filter using either cheap op amp sections or even
just emitter followers can produce far better results than
just a simple RC. I've found that an eg 4 pole Bessel
response can produce noticeably better results in practice
for filtered PWM tracking than a faster initial rolloff
filter (eg Butterworth or Chebychev). Lots of info on this
on web or ask in due course if you take that route.

Again, a more detailed spec on the target requirements will
help get better answers.



       Russell.. .

2008\07\23@205420 by threewheeler7

picon face

well firs off servos are unfortunately out of the question. second precision
is not much of an issue at all, just speed control, third certain specific
details unfortunately cannot be disclosed as requested by my client.
also why in AN893 does it run a full bridge? maybe i don't fully understand
the differences between full and half but isn't half bridge significant for
most apps? i plan on using hbrige driver chips rather than separate fets and
drivers. is there any problem with that?
--
View this message in context: www.nabble.com/multi-dc-motor-control-tp18619878p18623579.html
Sent from the PIC - [PIC] mailing list archive at Nabble.com.

2008\07\23@215847 by Apptech

face
flavicon
face
> well firs off servos are unfortunately out of the
> question. second precision
> is not much of an issue at all, just speed control, third
> certain specific
> details unfortunately cannot be disclosed as requested by
> my client.
> also why in AN893 does it run a full bridge? maybe i don't
> fully understand
> the differences between full and half but isn't half
> bridge significant for
> most apps? i plan on using hbrige driver chips rather than
> separate fets and
> drivers. is there any problem with that?

It's increasingly obvious that what you want is too vaguely
defined to expect useful answers. It should be easy enough
without violating an NDA (formal or informal) to much more
tightly specify what you are trying to do.

How many motors.
What precision of speed control.
What resolution.
Open loop or closed lop considered OK (impacted by above two
answers)

What power level.
What speed range (X to Y  or 0 to Y).
What response time to changes.
Any intermotor accuracy issues.
eg if this was for a wheelchair controller then the ability
for left and right wheels to track very accurately would
probably be more important than the absolute speed issue.
Acceptable cost.

...

More information = better answers.

Not worth starting to talk about drive control hardware etc
without more specifics.


       Russell


2008\07\23@232151 by Vasile Surducan

face picon face
On 7/23/08, threewheeler7 <spam_OUTthreewheeler7TakeThisOuTspamhotmail.com> wrote:
>
> ok i have a top priority design that needs to control 4 motors at a time.


Define "4 motors at a time". That's the key of your success or failure.


{Quote hidden}

> -

2008\07\23@235241 by Jinx

face picon face
> threewheeler7 wrote:
> come up with a good scheme for ramping up and down the speed of
> the motors in software

My 6-bit PWM project may be of help

http://www.piclist.com/techref/microchip/HV_PWM_Proto.htm

In a cycle of 64 TMR0 interrupts -

Each IRQ, check duty cycle stores to see if output(s) will be on

IRQ1 - check whatever sets duty cycle #1
IRQ2 - check whatever sets duty cycle #2
IRQ3 - check whatever sets duty cycle #3
IRQ4 - check whatever sets duty cycle #4

IRQ5-63 - free to do other tasks

The output frequency range is 100Hz - 279Hz, which would be
useable for motor drives. The project as presented has only one
PWM output. If you don't need the LCD that's another eleven
potential outputs available that could be serviced very easily in
the existing ISR

2008\07\24@001903 by Apptech

face
flavicon
face
>> ok i have a top priority design that needs to control 4
>> motors at a time.



YAPseudocode:
Multiple software PWMs.
IRQ timer increment.
Preferably PWM level check is run once per timer step.
Slower than this gives edge noise.
Much slower gives mayhem.
Faster than per IRQ is OK but gives edge noise.

____________________

' Per IRQ:

   timer++
   if timer > ktimernax then timer = 0

____________________

' Per IRQ (preferred) or main:
' imax PWM channels

For  0 <=i <= imax
     PWMOut(i) = (PWM(i) > timer)


2008\07\24@082752 by David Novak

picon face
I suspect that a half bridge solution will only allow the motors to turn in
one direction.

The hbridge chips will work fine as long as you don't require a large amount
of drive current.

David

http://www.dajac.com


> {Original Message removed}

2008\07\24@085311 by Apptech

face
flavicon
face
>I suspect that a half bridge solution will only allow the
>motors to turn in
> one direction.
>
> The hbridge chips will work fine as long as you don't
> require a large amount
> of drive current.

Too many uncertainties here to be able to be sure of what's
needed, alas. His motors could need from Watts to kilowatts
(unlikely). Directional control need unknown. Precision
unknown. More input please ... .


       Russell

2008\07\24@094623 by threewheeler7

picon face



Apptech wrote:
>
>
>
> It's increasingly obvious that what you want is too vaguely
> defined to expect useful answers.
>

Although it is not exactly true, think small scale animatronics with no
precise movements.


Apptech wrote:
>
How many motors.

3


What precision of speed control.



What resolution.

of what?


Open loop or closed lop considered OK (impacted by above two
answers)

closed loop with feedback from reflective type optical encoders, one each
set up just to count pulses for position. (however, because of the
characteristics of this design, once i get my motors in, i may decide to go
open loop to simplify design)


What power level.

2@up to 15w
1@up to 3w



What response time to changes.

whatever it takes to give a clean movement without a jerky look to it (ie.
ramping up and down the speed of a motor should look as clean as of you
where adjusting its speed with a pot)


Any inter motor accuracy issues.

no


Acceptable cost.

parts total under $30





--
View this message in context: www.nabble.com/multi-dc-motor-control-tp18619878p18632261.html
Sent from the PIC - [PIC] mailing list archive at Nabble.com.

2008\07\24@094924 by threewheeler7

picon face


Apptech wrote:
>
>
>
> It's increasingly obvious that what you want is too vaguely
> defined to expect useful answers.
>

Although it is not exactly true, think small scale animatronics with no
precise movements.


Apptech wrote:
>
> How many motors.
>
3



Apptech wrote:
>
> What resolution.
>

of what?


Apptech wrote:
>
> Open loop or closed lop considered OK (impacted by above two
> answers)
>
closed loop with feedback from reflective type optical encoders, one each
set up just to count pulses for position. (however, because of the
characteristics of this design, once i get my motors in, i may decide to go
open loop to simplify design)


Apptech wrote:
>
> What power level.
>
2@up to 15w
1@up to 3w



Apptech wrote:
>
> What response time to changes.
>
whatever it takes to give a clean movement without a jerky look to it (ie.
ramping up and down the speed of a motor should look as clean as of you
where adjusting its speed with a pot)


Apptech wrote:
>
> Any inter motor accuracy issues.
>
no


Apptech wrote:
>
> Acceptable cost.
>
parts total under $30





--
View this message in context: www.nabble.com/multi-dc-motor-control-tp18619878p18632261.html
Sent from the PIC - [PIC] mailing list archive at Nabble.com.

2008\07\24@135537 by gardenyu

picon face
part 1 1147 bytes content-type:text/plain; charset="gb2312" (decoded quoted-printable)

 A company called Allegro has some good full bridge and three phase controller chips. I guess you may control 2 or 3 motors in that way.

> Date: Fri, 25 Jul 2008 00:52:07 +1200> From: .....apptechKILLspamspam@spam@paradise.net.nz> Subject: Re: [PIC] multi dc motor control> To: piclistspamKILLspammit.edu> > >I suspect that a half bridge solution will only allow the > >motors to turn in> > one direction.> >> > The hbridge chips will work fine as long as you don't > > require a large amount> > of drive current.> > Too many uncertainties here to be able to be sure of what's > needed, alas. His motors could need from Watts to kilowatts > (unlikely). Directional control need unknown. Precision > unknown. More input please ... .> > > Russell> > -- > http://www.piclist.com PIC/SX FAQ & list archive> View/change your membership options at> mailman.mit.edu/mailman/listinfo/piclist
_________________________________________________________________
用手机MSN聊天写邮件看空间,无限沟通,分享精彩!
http://mobile.msn.com.cn/

part 2 35 bytes content-type:text/plain; charset="us-ascii"
(decoded 7bit)

2008\07\25@132251 by Bob Axtell

face picon face
I think using just one PIC would be better, because the communication task
might overwhelm you.

--Bob A

On Wed, Jul 23, 2008 at 2:54 PM, threewheeler7 <.....threewheeler7KILLspamspam.....hotmail.com>
wrote:

{Quote hidden}

> -

2008\07\25@211550 by threewheeler7

picon face

while driving a motor with PWM, many say to set the freq above the audible
hearing range, but in many cases isnt the motor loud enough that it covers
the harmonics created by the PWM drive?
--
View this message in context: www.nabble.com/multi-dc-motor-control-tp18619878p18662047.html
Sent from the PIC - [PIC] mailing list archive at Nabble.com.

2008\07\25@225403 by Apptech

face
flavicon
face
> while driving a motor with PWM, many say to set the freq
> above the audible
> hearing range, but in many cases isnt the motor loud
> enough that it covers
> the harmonics created by the PWM drive?

For some values of "many".
You can definitely have motors "sing" extremely audibly with
PWM.
This may not be a problem, depending on your application. In
a few applications it may even be an advantage if it adds
techo-appeal to your product. Fairly niche target audience*
though :-).  * In this case, unlike many where the word is
misused, AUDience is entirely correct.



           Russell



'[PIC] multi dc motor control'
2008\08\14@174717 by William \Chops\ Westfield
face picon face

On Jul 25, 2008, at 7:17 PM, Apptech wrote:

> This may not be a problem, depending on your application. In
> a few applications it may even be an advantage if it adds
> techo-appeal to your product.

There's some micro-RC gear (ie blimps) that use the motor windings
as a sort of speaker for audio feedback of configuration status.
Very clever...

BillW

2008\08\14@212916 by Jake Anderson

flavicon
face
William Chops Westfield wrote:
> On Jul 25, 2008, at 7:17 PM, Apptech wrote:
>
>  
>> This may not be a problem, depending on your application. In
>> a few applications it may even be an advantage if it adds
>> techo-appeal to your product.
>>    
>
> There's some micro-RC gear (ie blimps) that use the motor windings
> as a sort of speaker for audio feedback of configuration status.
> Very clever...
>
> BillW
>
>  
Actually most hobby brushless motor controllers do this,
Some even play "music" when they are turned on.
This is up to 3Kw in size, they also typically PWM at ~3Khz

More... (looser matching)
- Last day of these posts
- In 2008 , 2009 only
- Today
- New search...