reverse dc motor

[glinjik]

hi im getting the idea for a project what i need is to get a motor to turn one direction for a set distance then after a set time turn the the other direction for a set distance and repeat

so a timed circuit to switch on a relay and then after a delay time turn on another relay to make the motor turn the other way and repeat
a dc motor say about 30v

does this make sense to anyone not even sure im in the right forum section

[glinjik]

so i think i have me answer

[beanflying]

A few basics of DC motors. Generally the brushes are offset to perform best in one direction over another so reverse them and the performance won't be the same. So timed'' will not be particularly accurate even if you ignore other electro/mechanical considerations like load, acceleration and the rest of it.

So typically if you require any sort of precision you count the rotation with some sort of encoder. This needn't be a commercial encoder depending on the RPM you are doing it can simply be a photo diode and a slit on a plate an a micro to count the pulses.

So let us know some more detail and yes this is as good a section as any to ask 

[glinjik]

thank you for your reply in fact it will only require to rotate a little over half a turn i think a simple micro switch at the end of its travel would cut the power to relay and stop the motor until the timer starts again in the opposite direction, i guess quality capacitors would be needed to get the timing right but its doesn't need to be super accurate  about 15 mins +/-

[Benta]

Generally the brushes are offset to perform best in one direction over another so reverse them and the performance won't be the same.

I can't really recognize this. Yes, universal motors are extremely timed, but the "can" type PMDC industrial/automotive motors I've worked with are not. We're talking 20...500 W here, typically 12...36 V.
As an example: my "Jolly Motor" model catamaran runs with two oppositely rotating dual propellers (direct drive), and there's no difference in power or speed between the left and the right prop.

[jpanhalt]

!) If you are stuck on a DC motor with brushes, be sure to get a shunt wound motor.  Series wound, like older automobile starters, are not easily reversed.
2) For half a turn, I would use a stepper motor or a geared motor, like is used in a servo without the electronics. 
3) A little more difficult would be a servo motor.  That might be my first choice.
4) If you use a DC brushless motor, those are like 3-phase motors.  All you need to do is reverse two of the three wires, which is easily done with a DPDT mechanical switch.

[Benta]

I think we're talking PMDC motors here. Not 100% certain.

[beanflying]

Generally the brushes are offset to perform best in one direction over another so reverse them and the performance won't be the same.

I can't really recognize this. Yes, universal motors are extremely timed, but the "can" type PMDC industrial/automotive motors I've worked with are not. We're talking 20...500 W here, typically 12...36 V.
As an example: my "Jolly Motor" model catamaran runs with two oppositely rotating dual propellers (direct drive), and there's no difference in power or speed between the left and the right prop.

You generally advance the brushes to reduce arcing and you get an efficiency improvement. Talking in the order of 5-10 degrees and advanced. If you then reverse that same motor you will be running with brushes retarded and it will arc badly and run less efficiently.

Way back in the dim dark past of R/C electric flight a few of the contest classes only allowed stock Mabuchi motors but we were allowed to tweak them so we bench ran them and cranked on the backplates with a pin spanner to set the brushes for performance. Also way back in time I was playing with some early Electric human vehicles and we were running brushed motors typically one for each wheel so we were ordering from the manufacturer motors timed to suit. I don't miss brushed motors 

https://www.mabuchi-motor.com/product/knowledge/performance/phase.html

[Infraviolet]

I'd tend to suggest a h-bridge chip is often more practical than relays, solid state is longer lasting than moving relays. You could have either physical (low-current) switches as end stops to change the direction signals to the h-bridge when the motor reaches an end of its motion, or reduce wear over time by having some sort of non-contact proximity switch as the endstop.

[james_s]

Most general purpose PM DC motors are made to be reversible. The timing is set at a compromise rather than being optimized for a specific direction of rotation.

That circuit looks like it should work. You could also use a transistor or mosfet H-bridge, or there are H-bridge ICs that have everything built in, there's just a logic input to select direction. It is very, very common to have devices that run small motors in both directions. Think power windows on cars, hobby servos, and power zoom lenses on camcorders.

[Simon]

thank you for your reply in fact it will only require to rotate a little over half a turn i think a simple micro switch at the end of its travel would cut the power to relay and stop the motor until the timer starts again in the opposite direction, i guess quality capacitors would be needed to get the timing right but its doesn't need to be super accurate  about 15 mins +/-

I don't think you will be getting a brushed motor to do half a turn with any precision, they are not meant to work like that. It sounds like you want a stepper motor. Perhaps you should explain more about you application.

[coppercone2]

the best solution would be some kind of gear box thing IMO, the brushes are often very bad in many motors. And it like wears in to some contour. A relay that switches in a reverse gear mechanism via a solenoid.

[james_s]

Motors with integral reduction gearboxes are readily available. There are loads of applications that use these, I've seen them a lot on zoom and focus mechanisms on camcorders, CD players, robotics, etc.

[DavidAlfa]

There're cheap servos for such small and precise movements.
You would then drive It with two 555, one generating the pwm signal that tells the servo where to move,.the other would be a slow timer (1..2seconds, whatever) which would modify the first 555 pwm duty, thus modifying the servo position.