VERY simple dc motor drive circuit?

[Tim Wescott]

The fundamentals of constant speed motor control are shown at...

http://analog-innovations.com/SED/MotorDriver-Simple.pdf

The only caveat is that the external RM may not track the internal RM
over temperature.

Some day, when I'm in the mood, I will ponder that problem ;-)

...Jim Thompson

And the magnets will vary in strength over temperature.

I understand that there are also PWM driver circuits that read the motor
voltage when the driver amp is high impedance, basically using the motor
as a tachometer. This doesn't solve the magnet's variance with
temperature, but that's not a big issue normally.

I haven't tried it, but someday...

--

Tim Wescott
Wescott Design Services
http://www.wescottdesign.com

Posting from Google? See http://cfaj.freeshell.org/google/

"Applied Control Theory for Embedded Systems" came out in April.
See details at http://www.wescottdesign.com/actfes/actfes.html

 

[Jim Thompson]

And the magnets will vary in strength over temperature.

I understand that there are also PWM driver circuits that read the motor
voltage when the driver amp is high impedance, basically using the motor
as a tachometer. This doesn't solve the magnet's variance with
temperature, but that's not a big issue normally.

I haven't tried it, but someday...

Now that you mention it, I remember the trick of reading the back EMF.

However, where does the current go ?

Then there's EC motor (electronically commutated) controllers like I
designed for Bosch. The speed is a function of the clock rate.

I still have one of their motors here that is as big as an alternator
;-)

...Jim Thompson

 

[John E.]

Jim Thompson sez:

The fundamentals of constant speed motor control are shown at...

http://analog-innovations.com/SED/MotorDriver-Simple.pdf

A straightforward solution.

What is U1? Any op-amp?

Is Rext chosen to be the same R as the motor?

To make speed variable, should R1 be replaced with variable?

The only caveat is that the external RM may not track the internal RM
over temperature.

That's OK. I think that speed regulation is not so critical. The goal is to
be able to adjust full range. If it's noticed that speed has changed due to
thermal reasons, can easily be manually adjusted. It should settle down once
motor has warmed...

Some day, when I'm in the mood, I will ponder that problem ;-)

If you would be so kind... (c:

Thanks,

 

[Ken Smith]

That's all well and fine how ever, a must smaller and more
ergonomic approach would be to use a 555 driving a Hex Fet which
can all mount on a back of a panel mount pot.
More than likely the Fet will not need a heat sink.

Don't forget the diode on the motor to eat the spike as the MOSFET turns
off.

If you run the pin 5 of the 555 to a zener, you can compensate for the
variations in the input voltage.

 

[JerryT]

5vdc/670mA max miniature gear motor (load is such that current will be far
below 670). I want to set speed with pot control and leave it (may replace
pot with fixed R when best speed is discovered). Forward only, no reverse.
Just need single transistor, if possible, to do the work. Will use wall wart
p.s. for power.

Speed accuracy, etc. not necessary. This is for friend's art exhibit.

Google turns up an extraordinary number of patents, data sheets for quite
bright ideas using controller ICs, H-bridges, digital controls with feedback,
etc. Much more than I'm looking for.

Your ideas or pointers to designs greatly appreciated.

http://www.uoguelph.ca/~antoon/circ/pwm555.html

JT

Thanks,

 

[Jim Thompson]

Jim Thompson sez:


A straightforward solution.

What is U1? Any op-amp?

Either an OpAmp with enough power capability to run the motor
directly, or an OpAmp with a boost transistor added at the output.

Is Rext chosen to be the same R as the motor?
Yes.


To make speed variable, should R1 be replaced with variable?

No, VC is the "command"

That's OK. I think that speed regulation is not so critical. The goal is to
be able to adjust full range. If it's noticed that speed has changed due to
thermal reasons, can easily be manually adjusted. It should settle down once
motor has warmed...


If you would be so kind... (c:

Thanks,

...Jim Thompson

 

[Wim Lewis]

There's a new convention around here. Henceforth, any person posting "Use
a PIC" without providing your code, circuit, and development kit is
to be automagically deemed a "Troll".

In that case, from now on I will be suggesting the use of AVRs! I wouldn't
want to be a "troll".

More seriously ... while I agree that using a uC to generate an unvarying
square wave is overkill, I think you're diluting the term "troll"
into meaninglessness ... it doesn't just mean "someone who posted something
I don't agree with", you know.

Actually, the uC approach might not be so silly. DigiKey shows about the
same prices for PIC10F200s as for 555s, and the PIC would require fewer
external components (after programming, of course). Ditto the ATtiny11 and
some of the Zilog Z8s. Sheesh!

 

[Lionel]

5vdc/670mA max miniature gear motor (load is such that current will be far
below 670). I want to set speed with pot control and leave it (may replace
pot with fixed R when best speed is discovered). Forward only, no reverse.
Just need single transistor, if possible, to do the work. Will use wall wart
p.s. for power.

Speed accuracy, etc. not necessary. This is for friend's art exhibit.

LM317, a pot, & a couple of resistors should do the trick easily &
cheaply. Suitable circuits can be found on most LM317 data sheets.

 

[Tomi Holger Engdahl]

5vdc/670mA max miniature gear motor (load is such that current will be far
below 670). I want to set speed with pot control and leave it (may replace
pot with fixed R when best speed is discovered). Forward only, no reverse.
Just need single transistor, if possible, to do the work. Will use wall wart
p.s. for power.

Speed accuracy, etc. not necessary. This is for friend's art exhibit.

Google turns up an extraordinary number of patents, data sheets for quite
bright ideas using controller ICs, H-bridges, digital controls with feedback,
etc. Much more than I'm looking for.

Your ideas or pointers to designs greatly appreciated.

Take a look at the links here
http://www.epanorama.net/links/motorcontrol.html#dc

 

[Ken Smith]

LM317, a pot, & a couple of resistors should do the trick easily &
cheaply. Suitable circuits can be found on most LM317 data sheets.

That's not good. Every problem must be solved by a circuit using either
an LM555 or a PIC, perferably both :>

Earlier, I suggested the LT1083 as a better way to go than the LM317
because of the heat issue. I have also thought of a very ugly circuit
like this:



Input +V
------+------------------+----------------+----- To motor
! ! !
! \ !
! / ---
! -- \ --------- ^
! ! ! ! !
! --- +---/\/\---++----+----- To motor
! --- ! !
\ ! !/e !
/ <----/\/\--+---! PNP !
\ !\ !/
! +---------! NPN
! ! !\e
\ \ !
/ / GND
\ \
! !
GND GND

This requires much less current in the POT than the single transistor one.
The NPN's heat is a lot lower too.

 

[Lionel]

That's not good. Every problem must be solved by a circuit using either
an LM555 or a PIC, perferably both :>

Hey, LM317s are /nearly/ as traditional an answer to any kind of
electronics problem as the classic 555, & much more traditional than
those new-fangled PICs. (And my science teacher in highschool used
LM3909 as his solution for most things.)