Author Topic: DC Motor Feedback without an Encoder  (Read 3628 times)

0 Members and 1 Guest are viewing this topic.

Offline pete2009Topic starter

  • Jr. Member
  • **
  • Posts: 22
  • Helpful? 0
DC Motor Feedback without an Encoder
« on: June 26, 2009, 02:40:43 AM »
Hi Guys,

Is that possible, if so how?

Offline billhowl

  • Supreme Robot
  • *****
  • Posts: 376
  • Helpful? 32
  • Your success is in your hands.
Re: DC Motor Feedback without an Encoder
« Reply #1 on: June 26, 2009, 03:34:44 AM »
Yes, it is possible, you can monitoring the current from the DC motor.

Read this
http://www.industrialcontroldesignline.com/howto/212000475
http://prism2.mem.drexel.edu/~rares/FB_DC.html

Offline Soeren

  • Supreme Robot
  • *****
  • Posts: 4,672
  • Helpful? 227
  • Mind Reading: 0.0
Re: DC Motor Feedback without an Encoder
« Reply #2 on: June 26, 2009, 03:46:02 AM »
Hi,

Yes, just monitor the drive voltage, it will dip slightly when meeting a pole, count dips, divide by number of poles and you have number of revs.
Regards,
Søren

A rather fast and fairly heavy robot with quite large wheels needs what? A lot of power?
Please remember...
Engineering is based on numbers - not adjectives

Offline guru

  • Full Member
  • ***
  • Posts: 59
  • Helpful? 1
  • Robotics, CNC, Electronics, Programming
    • ColinMacKenzie.net
Re: DC Motor Feedback without an Encoder
« Reply #3 on: June 26, 2009, 05:52:36 AM »
Some motor controllers measure back-EMF produced by the motor to measure speed and use it in a velocity feedback loop (so the position doesnt have to derive velocity and the system can be more stable.). Usually position is still measured, so there are two loops. Measuring BackEMF is complicated, it's like you trying to measure a spinning wheel while someone is vehomently trying to spin it. There is filtering that needs to take place, preferrably done in the analog domain. Which means you better know your differential equations.

In my opinion, I think you will still have your work cut out for you measuring the rotation with what Soeren said. If you are just trying to save money, buying the encoder is the cheaper route. You'll throw a lot of $$$ in development for non-encoder measurement.

The current is fairly easy, but still requires a filter, albiet a simple one. Otherwise, you could same during a dip in the current.

Take a look at http://www.openservo.com , they measure current, backEMF now too I think, and the schematics have some of the filters needed. You might even want to start from that platform and get where you want it. And they would be interested in any contributions you could make.

Colin

 

SMF spam blocked by CleanTalk