Help please with DC motor control

On Thu, 31 Jul 2014 19:19:30 -0500, Tim Wescott
<tim@seemywebsite.really> wrote:

On Thu, 31 Jul 2014 17:13:33 -0700, etpm wrote:

On Thu, 31 Jul 2014 12:33:34 -0500, Tim Wescott
tim@seemywebsite.really> wrote:

On Wed, 30 Jul 2014 16:36:15 -0700, etpm wrote:

On Wed, 30 Jul 2014 16:07:28 -0500, Tim Wescott
tim@seemywebsite.really> wrote:

On Wed, 30 Jul 2014 08:56:25 -0700, etpm wrote:

SNIP

A DC motor can be regulated to deliver constant torque by using a
series resistor which measures the motor current,
then feeds the signal back to a servo amplifier with gain
adjustment.
But then you will not have constant (rotational) speed.

This is not the same problem as constant speed control.
For DC speed control you would use a tacho-generator or when
pulsing the motor you can use the motor itself as tacho-generator
during the pulse pause.


This is not a simple task, but there are some hand drill machines on
the market which have adjustable speed and adjustable torque.
To be precise: adjustable maximum torque.

One cannot have constant speed and constant torque in one when the
mechanical load changes permanently.
You cannot have constant torque if the load changes.
Only a maximum torque regulator can be achieved but at the cost of
changing speed.

w.
Thanks for the replies everyone. After reading the replies I am a
little confused. If I use a current and voltage limited power supply
and there is not quite enough power for the motor to pull against
the drag the motor will be stalled.

You mean current, not power. Torque in a permanent-magnet DC motor is
proportional to current, not power.

When I pull on the cord the motor will of course turn, but only when
I pull on the cord. If I pull the cord the motor will then just draw
less current, right?

That depends on where the voltage limit on the power supply kicks in.
If you have it set high enough then the current in the motor will stay
constant, and the torque will, too.

So the torque it develops plus the pulling force will be just enough
to overcome the drag, won't it?

Probably.

If not, then what am I missing?

Nothing that I can think of off the top of my head, but given the
other suggestions so far it does seem to be a wasteful way to go about
the job.
Greetings Tim,
I understand that torque is proportional to current. When I used the
word power I was thinking about watts, voltage times current. But I
see where I was wrong. Low current and high volts would be the same
wattage as high current and low volts. In this situation current is
what matters for torque. Though the other solutions have their merits
I think the convenience of having the same pulling force no matter
what speed I pull the cord, within reason, will be best for me. So
another question. I see lots of chips for driving low current BLDC
motors. If I was to use one of these and the motor was stalled would
the motor driver still try to make the motor move? Do any work like
that? They need no commutation but instead supply a constantly
rotating field and the motor tries to spin at the correct RPM? I need
to look on the web to see if what I want is available.
Eric

The "needs no commutation" part will kill you in this application.
AFAIK those chips _do_ need commutation -- they generate it by sensing
the back-
EMF of the spinning motor, which you can't do with a motor that's
stationary.

You'd need a BLDC motor with commutation outputs (usually three Hall
sensors), and a driver that knows what to do with those outputs. And if
you're moving it slowly, expect to feel a perceptible "bump" every time
a Hall detector switches.

For all that I'm usually in the business of providing fancy solutions to
simple problems -- I think you're over-complicating things here. If
you're going to do it with a DC motor, just get a surplus brushed motor
and a cheap constant-current supply, and don't mess with BLDC stuff
unless you're in it for the journey and not the destination.
Greetings Tim,
I am in it partly for the journey. To begin with I am just going to use
one of the DC motors I have sitting around waiting to be used. However,
after thinking about BLDC motors and commutation I realized that the
motor drivers must initially energize the motor coils to make a rotating
magnetic field so the motor will spin up to some minimum RPM so the the
driver can detect the back EMF or hall effect switches for commutation.
Which makes me wonder if I can find a driver which doesn't need
commutation but instead just energizes the coils at a certain frequency,
like the way an induction motor works. Then the motor can spin at any
RPM from stall up to the rotating field RPM. In fact, I bet someone has
written the code for an Arduino to do this. I should look for it.
Thanks,
Eric

The hall effect switches are static; they let the controller know what
quadrant the armature is in whenever they have power applied.

There are "sensorless" brushless motor controllers, but they are really
only useful for driving loads that present little or no back-torque at
low speeds, and that tend to run at 10% of top speed or higher - like
fans, or RC airplane propellers.

Induction motors are qualitatively different from BLDC motors. Induction
motors work because the armature gets dragged along with the rotating
magnetic field. As such they are "self timed". BLDC motors are
synchronous machines: the rotor must be in step with the rotating
magnetic field. You can run a BLDC motor by rotating the magnetic field
open-loop, but then its working like a stepper motor. You really can't
generate much torque reliably that way, and if the motor slips out of
synchronization the _average_ torque will go to zero, no matter how much
vibration is added by the coils.

For all of the above reasons, whenever a BLDC motor must generate
significant torque at low speeds, it's got hall-effect sensors.

I already knew about how induction motors work but I did not know that
BLDC motors would stall the way you describe. In my experience when a
stepper is being stepped fairly fast and enough load is applied to
stall it when the load is removed the stepper is likely stay nearly
stalled-it just vibrates-until the step frequency is lowered enough
and then it will start to step again. So BLDC motors act the same. I
should have figured that out. Thanks for pointing that out.
Eric

 

[Phil Hobbs]

On 7/31/2014 8:13 PM, etpm@whidbey.com wrote:

On Thu, 31 Jul 2014 12:33:34 -0500, Tim Wescott
tim@seemywebsite.really> wrote:

On Wed, 30 Jul 2014 16:36:15 -0700, etpm wrote:

On Wed, 30 Jul 2014 16:07:28 -0500, Tim Wescott
tim@seemywebsite.really> wrote:

On Wed, 30 Jul 2014 08:56:25 -0700, etpm wrote:

SNIP

A DC motor can be regulated to deliver constant torque by using a
series resistor which measures the motor current,
then feeds the signal back to a servo amplifier with gain adjustment.
But then you will not have constant (rotational) speed.

This is not the same problem as constant speed control.
For DC speed control you would use a tacho-generator or when pulsing
the motor you can use the motor itself as tacho-generator during the
pulse pause.


This is not a simple task, but there are some hand drill machines on
the market which have adjustable speed and adjustable torque.
To be precise: adjustable maximum torque.

One cannot have constant speed and constant torque in one when the
mechanical load changes permanently.
You cannot have constant torque if the load changes.
Only a maximum torque regulator can be achieved but at the cost of
changing speed.

w.
Thanks for the replies everyone. After reading the replies I am a
little confused. If I use a current and voltage limited power supply
and there is not quite enough power for the motor to pull against the
drag the motor will be stalled.

You mean current, not power. Torque in a permanent-magnet DC motor is
proportional to current, not power.

When I pull on the cord the motor will of course turn, but only when I
pull on the cord. If I pull the cord the motor will then just draw
less current, right?

That depends on where the voltage limit on the power supply kicks in.
If you have it set high enough then the current in the motor will stay
constant, and the torque will, too.

So the torque it develops plus the pulling force will be just enough
to overcome the drag, won't it?

Probably.

If not, then what am I missing?

Nothing that I can think of off the top of my head, but given the other
suggestions so far it does seem to be a wasteful way to go about the
job.
Greetings Tim,
I understand that torque is proportional to current. When I used the
word power I was thinking about watts, voltage times current. But I see
where I was wrong. Low current and high volts would be the same wattage
as high current and low volts. In this situation current is what matters
for torque. Though the other solutions have their merits I think the
convenience of having the same pulling force no matter what speed I
pull the cord, within reason, will be best for me. So another question.
I see lots of chips for driving low current BLDC motors. If I was to use
one of these and the motor was stalled would the motor driver still try
to make the motor move? Do any work like that? They need no commutation
but instead supply a constantly rotating field and the motor tries to
spin at the correct RPM? I need to look on the web to see if what I want
is available.
Eric

The "needs no commutation" part will kill you in this application. AFAIK
those chips _do_ need commutation -- they generate it by sensing the back-
EMF of the spinning motor, which you can't do with a motor that's
stationary.

You'd need a BLDC motor with commutation outputs (usually three Hall
sensors), and a driver that knows what to do with those outputs. And if
you're moving it slowly, expect to feel a perceptible "bump" every time a
Hall detector switches.

For all that I'm usually in the business of providing fancy solutions to
simple problems -- I think you're over-complicating things here. If
you're going to do it with a DC motor, just get a surplus brushed motor
and a cheap constant-current supply, and don't mess with BLDC stuff
unless you're in it for the journey and not the destination.
Greetings Tim,
I am in it partly for the journey. To begin with I am just going to
use one of the DC motors I have sitting around waiting to be used.
However, after thinking about BLDC motors and commutation I realized
that the motor drivers must initially energize the motor coils to make
a rotating magnetic field so the motor will spin up to some minimum
RPM so the the driver can detect the back EMF or hall effect switches
for commutation. Which makes me wonder if I can find a driver which
doesn't need commutation but instead just energizes the coils at a
certain frequency, like the way an induction motor works. Then the
motor can spin at any RPM from stall up to the rotating field RPM.

Doesn't work well, because it has zero ability to recover. Once the
actual rotor position is significantly different from what your circuit
thinks it is, the torque goes essentially to zero.

Cheers

Phil Hobbs


--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC
Optics, Electro-optics, Photonics, Analog Electronics

160 North State Road #203
Briarcliff Manor NY 10510

hobbs at electrooptical dot net
http://electrooptical.net

 

[whit3rd]

On Tuesday, July 29, 2014 9:34:01 AM UTC-7, et...@whidbey.com wrote:

I am working on a project in my shop that requires constant tension on
a nylon cord. The cord comes is pulled off of a spool a certain amount
and cut. I have drag on the spool because I don't want it keep
spinning when I'm not pulling. But pulling against this drag is
tiresome.

In large sewing machines, this is dealt with by unloading the thread upward
from a cone-winding, rather than circumferentially from a spool.

One might also adopt an old tape-drive trick, where the unspooled
line hangs in a U bend, and the brake on the spool engages
only when there's lots of droop in the outfeed. If one
ran the line up over a pulley, and sensed downward force on the pulley
to release the brake, the spool is stopped any time there's no
string tension. String tension causes switch causes brake engagement.

 

[Michael A. Terrell]

whit3rd wrote:

One might also adopt an old tape-drive trick, where the unspooled
line hangs in a U bend, and the brake on the spool engages
only when there's lots of droop in the outfeed.

Those tape machines used a vacuum to pull the loop to keep it from
knotting up. Some 16 mm projectors used a small variac and undersized AC
motor with a pulley mounted on a spring loaded arm to keep constant
tension on the film. Any variation caused the sound to stutter. I worked
with this in a film chain that used RCA TP66 projectors. No tension ran
the motor at full torque. Too much reduced it to the minimum. The came
in handy when loading a new reel. You could slip the leader between your
fingers when slowly letting the arm up to wind the leader onto the
reel. Run the nylon cord between some pulleys with a slight groove to
keep it centered. Pull on it, and that pulls against the control lever
to adjust the motor speed. Stop pulling, and it stops.


--
Anyone wanting to run for any political office in the US should have to
have a DD214, and a honorable discharge.