2
\$\begingroup\$

My control system is a simple microcontroller between 3.3V and 5V, but the stepper motors I require are 12V. I have ordered a 14.8V battery that I intend on regulating down to 12V to drive the motors, but should I also regulate down to 5V for the control system or use a separate power source?

In short, are the control systems and motor drives typically powered by the same source or are they independent? I know that high series resistance batteries can show significant output voltage drop when high current is drawn, but my question is really regarding common practice in power design.

Assuming the voltage drop from high current is low (not sufficient to affect the 5V regulated bus), would it be reasonable to use the same power source for both systems?

Using something like this, for example: http://www.hobbyking.com/hobbyking/store/__24579__Turnigy_5V_6V_5A_Heli_UBEC_for_Lipoly_6_16V_.html

\$\endgroup\$
3
2
\$\begingroup\$

You'll usually have one main power source to the project. Having multiple sets of batteries (A batteries for the filaments, B batteries for the plates) went out of style decades ago.

You want LOTS of decoupling between the microcontroller supply lines and the stepper supply lines. Ditto the microcontroller and the stepper driver board supply lines. Stepping motor coils are good inductors (DUH!), and they throw BIG spikes back at the controller when they are switched. You do NOT want those big spikes to get back into your microcontroller supply. Bypass capacitors and filter capacitors, conservatively rated, are your FRIENDS.

Stepping motor driver design is a Black Art. It is not unusual to see a 12V stepping motor being driven from a 50V supply, as this allows creative design to minimize settling time. The old Airpax stepping motor catalog had some REALLY good design information on this.

In this day and age, decent quality cheap switching power converters are easy to come by. For a battery-powered system, these days, I'd look for one of these, in preference to the old 7800-series linear regulators, for the microcontroller supply.

\$\endgroup\$
3
  • \$\begingroup\$ Could you elaborate why you would go against the linear option? With the current draw of the microcontroller being so insignificant compared to driving motors it seems like the linear would be my choice. \$\endgroup\$ – tman Sep 22 '14 at 19:23
  • \$\begingroup\$ @tman, mostly habit at this point. If the robot is going to spend significant amounts of time sitting, powered-up but quiescent, the microcontroller draw can start to add up. It all depends on the controller. Also, you have to take heat dissipation into account. For a linear regulator, current out = current in + a little slop, so 12V -> 5V linear means well under 50% power efficiency. If you're drawing 100 mA, then your microcontroller is chewing 500 mW and your regulator heat sink is dissipating 700 mW to the surrounding air. \$\endgroup\$ – John R. Strohm Sep 22 '14 at 20:54
  • \$\begingroup\$ @tman: That should have been current in = current out + a little slop. I'm having one of those weeks. \$\endgroup\$ – John R. Strohm Sep 25 '14 at 15:10
1
\$\begingroup\$

I would say absolutely. You may need to ensure your circuit can handle the fluctuations in current(smoothing cap) and protect the circuit from back EMF. How you step down the voltage is where you will see different practices. Especially considering that microcontrollers can "sometimes" run off a variety of input voltages and uses very low current. For running "just" the microcontroller a linear regulator should work fine. I think the poor efficiency will be nothing compared to that of the motors.

\$\endgroup\$
1
\$\begingroup\$

Both the batteries and controller uses same power source except they'll have separate voltage regulators. In your case you can use 7805 IC which takes 12V input and delivers 5V output which can be connected to your controller.

\$\endgroup\$
1
  • 3
    \$\begingroup\$ You'll want to be very careful using a standard 7805 for 12V to 5V, that's potentially a lot of heat to dissipate! \$\endgroup\$ – David Sep 21 '14 at 21:23

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.