Searching \ for '[PIC]: DC Motor open-loop control stategy' in subject line. ()
Make payments with PayPal - it's fast, free and secure! Help us get a faster server
FAQ page:
Search entire site for: 'DC Motor open-loop control stategy'.

Exact match. Not showing close matches.
PICList Thread
'[PIC]: DC Motor open-loop control stategy'
2001\04\19@164514 by Karl Skinner

picon face
Hi Piccers, can anyone help please with getting a better control strategy
for open-loop control of a DC motor driving a fairly large inertia. I'm
thinking along lines of "rate-lag" to improve response, but have been in
Piclist and Google to no avail.

Looking forward to your replies
Karl Skinner

-- hint: The PICList is archived three different
ways.  See for details.

2001\04\19@172303 by Eisermann, Phil [Ridg/CO]

And that probably means a good deal of friction too, right? BTW what do you
mean by 'fairly large' inertia? Got a number?

The better control strategy is of course closed loop. But since you're
asking open loop, i guess that's not an option. Disclaimer: I'm not a
control systems expert. But there is a limit as to how accurate you're going
to be able to get with an open-loop system. And if you have external
disturbances of any sort (one of the main reason people use control
systems), open loop isn't going to get you there.

Do you have a model of the system? Have you done any experimental
determination of the system's behavior, such as time-domain or frequency
domain response? Do you need to move the load over a wide range of speeds
(as in a profile), or just a constant speed? How often do you have to move
the load? All of these things will play a role, especially with open-loop
control. Motor characteristics tend to change as the motor warms up.
Sometimes it makes a difference, sometimes not.

{Original Message removed}

2001\04\20@042016 by Peter L. Peres

picon face
I suspect that you use speed control. Inertia is usually compensated by a
larger D term in the regulator. With large inertia and large D terms you
will have limiting in the amplifier when near the 'ends' of the domain.
This may be an excellent self-oscillation starter if not accounted for,
especially if there is also an I term. It also helps if the driver can
sink current for braking, not just source it. A push-pull driver is a
must imho.


-- Going offline? Don't AutoReply us!
email with SET PICList DIGEST in the body

2001\04\20@131712 by Andrew Errington

Since you are using open loop you have no choice but to drive the motor and
cross your fingers.  You have no feedback, and therefore no way to determine
whether the motor is moving (in the right direction) or if it is stalled.

I assume you wish to vary the motor's speed (if not then you are just
turning the motor on, so the issue is moot), so you are probably using PWM.
I think you need to know the range of loads you expect the motor to
encounter, and the minimum PWM duty cycle you need you get the motor going
with the largest load.  Any duty cycle above that will guarantee to move the
motor, but of course if the load is light then the speed will be greater
than if the load is heavy.

If you had a simple circuit (like a retriggerable multivibrator) that could
tell you whether the motor is stalled or not then you could use that to
'kick start' the motor, i.e. run the motor at 100% duty cycle until it is
moving, then drop down to the desired PWM (that you had previously
determined would be sufficient to keep it moving).  However, if you are
going to build such a circuit you might as well wire the sensor into the

Without such a circuit you need to try different PWM rates and measure how
long the motor takes to reach the desired speed.  You can then do something
like "run 100% for n seconds" then drop to desired PWM, where n is related
to the desired final speed.  This will work for the motor starting from a
standstill.  You then ought to be able to characterise how the motor
responds to changes in PWM, and work out how long to cut the PWM to 0% or
boost to 100% to get closer to the new desired speed whenever it changes.


> {Original Message removed}

More... (looser matching)
- Last day of these posts
- In 2001 , 2002 only
- Today
- New search...