View Single Post
  #6   Spotlight this post!  
Unread 10-10-2012, 19:50
tsaksa's Avatar
tsaksa tsaksa is offline
Registered User
FRC #0997 (CHS Robotics)
Team Role: Mentor
 
Join Date: Jul 2011
Rookie Year: 2011
Location: Corvallis Oregon
Posts: 203
tsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond reputetsaksa has a reputation beyond repute
Re: 110V AC to 12V DC suggestion

Quote:
Originally Posted by AdamHeard View Post
A power supply with enough current to run a robot system (with motor) will be very expensive.

A battery is the best cheap option.
That is quite true if you really need or want to run a whole robot. But if you really need or want a non battery system you may be able to get one to work with a stripped down version of the robot. This can also come in handy for training members of the electrical team. Our team is building a development station right now that we intend to run from a DC power supply. (Although we were particularly lucky to get an unusually large supply donated, if you are careful a mor modest supply might work.) We have had some luck previously with some scaled down sub components. If you build up a system with just the CRio, a power distribution board, digital side car, and modem, but not much else, you will have much better luck powering it without a battery. Just do not expect to run a compressor or any CIM motors from an average size DC power supply. Even small DC motors under load are probably a bad idea for this approach.

In place of motors and motor controllers you can use some servos with little arrows on the actuator pointing to a home-made gauge marked with Forward, OFF, Reverse. This only simulates the motors and does not really replace the need at some point to test the code on a real robot. But it can give the software team a lot of debug capability without access to a full robot. Just an idea.

You can possibly add some Jaguars, victors, and spikes without taking a big hit on additional power requirements if you use small toy motors that are not under much load. It all depends on what your intent really is. There are times when it is really great to let the software team or electrical team have access to something that does not need a battery, and is safer than a full robot even if it is not fully representative of the final product. It is also nice to have options if you do not always have access to a place where it is safe to operate a real robot, or if you have a hard time keeping the software team from killing your batteries.

A while back I built a small sub-portion of this that you might consider as a way to try out this idea on a much smaller scale. I mounted a toy motor, victor, and encoder together on small block of wood with an old 12 volt laptop battery charger as the power supply. Plug in the supply and this setup is ready for stand alone training in PWM, encoders, PID, etc. You can hook it up to a PWM output or PWM signal generator to demonstrate how the motor controllers work. Or you can hook the encoder up to a scope or digital input of the CRio to show how encoders work. Or you can hook both to a digital side car for an exercise in PID control. It is not a good representation of the real world torque, rotation inertia, or time constants you will encounter in a real robot. But it is simple and easy to use, and gets the basic concepts across. It worked for me. Your mileage may vary.
__________________
This is the zeroth day of the rest of your life.