>
> I was trying to find a way to find out when to cut the power without
>having to resort to a optical encoder on the driving mechanism. My feeling
>is that it has to be some electrical indication somewhere when the motor
>reachs the new position.. I guess I am wrong...
You can always power it down say 1-2mS after you make the last step.
Beware though, you'll also need a delay on powerup, but I'm not sure how long.
You want the motor to re-center to the position you're in, before you start changing things.
Also, how do you know that it won't drift away, more than a step size, from position with the power off?
If you can sense the current, there is a current surge when it is
stepping until it reaches its final position. Once there it draws less
current since there is no real load.
If you look around the internet you should be able to find people using
this type of method to sense the speed of DC motors, since the current
drawn partly depends on where the coil is in its rotation. You should
be able to find some ideas there, the signals will be very similar since
a DC motor could be considered a self stepping stepper.
>Hi,
>
>
>
>>there are very nice stepper drive ic's that just take a step and
>>direction signal, and have an adjustable current limit. normally the
>>timing is done in software and is based on how fast the system can
>>accelerate and decelerate. <http://allegromicro.com/> makes some nice
>>stepper control chips and has some good technotes.
>>
>>
>
> They have some really nice parts but for my application they are
>overkill. It is a pick and place machine and I intend to let the motor
>without any power in between moving cycles. Usually the driver chips like to
>let power trough the motor all the time.
>
> I was trying to find a way to find out when to cut the power without
>having to resort to a optical encoder on the driving mechanism. My feeling
>is that it has to be some electrical indication somewhere when the motor
>reachs the new position.. I guess I am wrong...
>
>Best regards,
>Alexandre Guimaraes
>
>
>
>_______________________________________________
>http://www.piclist.com
>View/change your membership options at
>http://mailman.mit.edu/mailman/listinfo/piclist
>
>
>
>
>
> > It seems so awfully inneficient to drive the motor more time than
needed
> > !! Relying on time alone to cut the current in one phase and go to the
next
> > phase does not sound nice to me... I would prefer a more efficient way
of
> > doing it. Besides that I would love to take a little heat of the back of
my
> > poor lm317 !
>
> Look up chopped drivers.
Yeah... I know I am being cheap... But 10 watts of lost power is easier
for a one off project the getting the chopper driver to work properly.. I
have a bad memory on trying to make a chopper work with some big motors some
years ago and the delay ended up being some resistors with too much
inductance. It really took a while to sort out the problem at that time. For
this one with 1 amp motors I decided to stay away from trouble.
It seems to me now that the only way to control the stuff is really to
put a encoder or stay with the motor phase actuated a little longer than
needed to make sure it has had enough time to get to the new position. I was
hoping to find a better way.
> If you can sense the current, there is a current surge when it is
> stepping until it reaches its final position. Once there it draws less
> current since there is no real load.
>
> If you look around the internet you should be able to find people using
> this type of method to sense the speed of DC motors, since the current
> drawn partly depends on where the coil is in its rotation. You should
> be able to find some ideas there, the signals will be very similar since
> a DC motor could be considered a self stepping stepper.
That is good news.. That is the kind of effect I was looking for. The
only problem is that I am using a constant current supply that will just
"push" more voltage there when needed. I am not sure I will be able to
measure that current variations but it is surely worth trying..
> You can always power it down say 1-2mS after you make the last step.
> Beware though, you'll also need a delay on powerup, but I'm not sure how
long.
> You want the motor to re-center to the position you're in, before you
start changing things.
>
> Also, how do you know that it won't drift away, more than a step size,
from position with the power off?
The mass that is driven by the stepper is much smaller than the holding
torque of the magnets in the motor, that should help. To make sure I could
detect the arrival at the new position and keep the motor driven a while
longer to make sure it will not keep moving because of the inertia of the
system.
I guess I will just put a small fan in the power supply and blow away
the 10 watts of heat that it will be generating.. It will be easier than
testing all the stuff that would need to be done...
> Hi,
>
> >there are very nice stepper drive ic's that just take a step and
> >direction signal, and have an adjustable current limit. normally the
> >timing is done in software and is based on how fast the system can
> >accelerate and decelerate. <http://allegromicro.com/> makes some nice
> >stepper control chips and has some good technotes.
I agree with this suggestion; take a look at the popular L297/L298 pair.
> They have some really nice parts but for my application they are
> overkill. It is a pick and place machine and I intend to let the motor
> without any power in between moving cycles. Usually the driver chips like
> to let power trough the motor all the time.
I don't know how much your P'N'P header is mechanically loaded, but to
operate at high speeds probably you'll need to keep the motors powered at
the stop points, to avoid the inertial effects.
> I was trying to find a way to find out when to cut the power without
> having to resort to a optical encoder on the driving mechanism. My
> feeling is that it has to be some electrical indication somewhere when
> the motor
> reachs the new position.. I guess I am wrong...
Yep (if i understand what you want), you'll need to sense phase's current.
Another possibility is to use another stepper (one by axis) mechanically
coupled, as a pulse generator, but i think the encoder solution is better.
Just my 2 cents.
Your project seems very interesting; ¿What will be the file input format to
the machine?
Best Regards,
S.-
--
Sebastian E. Garcia
sgarcia at fi.uba.ar
Embedded Systems Development 54.11.4833.4123
Buenos Aires, Argentina.
You can read the output voltage,
when the load current drops the current regulator's output voltage will rise
drastically giving an indication that the motor has reached it final
position.
KF4HAZ - Lonnie
----- From: "Alexandre Guimaraes" <listas@
<snip>
> That is good news.. That is the kind of effect I was looking for. The
> only problem is that I am using a constant current supply that will just
> "push" more voltage there when needed. I am not sure I will be able to
> measure that current variations but it is surely worth trying..
>
> Thanks a lot.
>
> Best regards,
> Alexandre Guimaraes
> You can read the output voltage,
> when the load current drops the current regulator's output voltage will
rise
> drastically giving an indication that the motor has reached it final
> position.
That should be quite easy to measure if the variation is big enough.. I
will try that over the weekend and let you know how it worked out.
Alexandre, try here , http://www.romanblack.com/
look at Roman's dc motor encoder.
I t may be what you are looking for.
victor
{Original Message removed}
I have done quite much work with step motor.
Firstly, I found its step angle is totally not 7.5deg/step,
or 1.8deg/step. Then I found, the solution micro-step
even more like a joke, no sin relation and no predictable
position.
Then I found, if go to close loop control field, stepper
motor can provide very fine adjusting in angle. And even
under varing torque. I used them as niddle valve-flow rate
controling, successfully.
Step motor, is not good choice for controlling absolute position,
such as robot arm.
> I have done quite much work with step motor.
> Firstly, I found its step angle is totally not 7.5deg/step,
> or 1.8deg/step. Then I found, the solution micro-step
> even more like a joke, no sin relation and no predictable
> position.
That is not critical in my application. For pick and place of 0805
components I have plenty of resolution even using full step and belt drive.
> Then I found, if go to close loop control field, stepper
> motor can provide very fine adjusting in angle. And even
> under varing torque. I used them as niddle valve-flow rate
> controling, successfully.
That is quite interesting. Can you elaborate more on how this fine
adjustment is done ? You are using a optical encoder as the feedback loop
control ? How are you driving the steppers ?
----- Original Message -----
From: "Alexandre Guimaraes" <.....listasKILLspam@spam@logikos.com.br>
To: "Microcontroller discussion list - Public." <piclistKILLspammit.edu>
Sent: Thursday, September 09, 2004 7:05 PM
Subject: Re: [EE]: Stepper motor position.
> Hi,
>
> > I have done quite much work with step motor.
> > Firstly, I found its step angle is totally not 7.5deg/step,
> > or 1.8deg/step. Then I found, the solution micro-step
> > even more like a joke, no sin relation and no predictable
> > position.
>
> That is not critical in my application. For pick and place of 0805
> components I have plenty of resolution even using full step and belt
drive. {Quote hidden}
>
> > Then I found, if go to close loop control field, stepper
> > motor can provide very fine adjusting in angle. And even
> > under varing torque. I used them as niddle valve-flow rate
> > controling, successfully.
>
> That is quite interesting. Can you elaborate more on how this fine
> adjustment is done ? You are using a optical encoder as the feedback loop
> control ? How are you driving the steppers ?
>
> Best regards,
> Alexandre Guimaraes
>
>
> _______________________________________________
> http://www.piclist.com
> View/change your membership options at
> http://mailman.mit.edu/mailman/listinfo/piclist
>
Steppers are usually driven in the following fasion:
Coils are driven with constant current, the voltage at standstill is quite
low.
At ramp up the driver starts at the start/stop frequency usually above 100hz
and ramps up from there till the max frequency
The max frequency and ramp up slope depend on the type of motor, current,
peak voltage and required torque.
Deceleration is done with the same slope ending at the start/stop frequency.
In order to reduce heating of driver and motor is generally done by reducing
the coil current a to a percentage of the run current depending on the
holding torque required for the application.
If a ecoder is used in the design this is not done to controll motor current
but to detect the steppers actual position because of it's inherrant problem
of missing steps when the load or speed are too great or the
acceleration/deceleration is too fast. often no feedback is included in the
loop and the encoder merely generates an error condition signal.
Hope this may give you some ideas on how to proceed
logikos.com.br> wrote:
