Let's Make Robots!

DC 20.6V DC 6616-006B

Hi all, 

I have some basic question about this DC motor. I found this motor have 5 Wire.

RED,BLACK,BLUE,YELLOW & ORANGE, image

The wire RED and BLACK wire are connected to motor coil directly and can be used just to simply run it.

BLUE, YELLOW & ORANGE are connected to an magnetic rotary Encoder.

There is an Megnatic Rotary disk attached to the other end of motor shaft and is fixed in such a way that it rotates along with the rotation of the Shaft. ImageA ImageB

The Circuit of the Three Wire image is hereA hereB

I would like to know how to use this DC6616 Motor and what applications/Purpose this can be used for.

Further Info: i used the normal Read analogvolt Sketch on Arduino whereby i connected the 5v to VCC and VOUT to A0 and GRD to GRD on Atmega2560.

I let the motor run with seperate powersupply using 12V on Red and Black wire and found the Analog readout was 2.5V after several interval of 0 eg 0,0,0,0,2.5,0,0,2.5,0,0,0,2.5

I tried to find after how many interval it repeates the 2.5V reading and failed to do so as the repeatation is sometime after 4 0's sometime after 2 0's sometime 3 0's 

Also i disconnected the powersupply to motor and rotate the shaft manually by hand and found on slow rotation it reads nothing only 0's

when i give it a sudden rotation then it spikes to 2.5V and then back to 0.

 

i hope the above info would be helpful for guys out there who exactly know what this motor encoder is.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.

The magnet passing over the zigzag trace creates an ac wave.  Since there are 60 (?) zigzags it will be 60 times higher than the RPM. Assuming the circuit traced by Torrentula is correct, and it most likely is, The Vout will simply be an amplified version of that, between GND and Vcc.  So at say 6000 RPM (100 Revs per second) the output frequency would be 60 times that, or 6000 Hertz.  Your inconsistent readings of the voltage are probably because you aren't measuring it fast enough.  You would have to measure "at least" twice that fast (12000 times per second in the example) to follow the waveform (Google for Nyquist sampling theorem).  The magnet and zigzag trace works like a generator, the faster it spins the higher the voltage.  That is why you don't see anything when you spin it slowly.  You can feed that ac signal into a timer/counter on the arduino to measure the frequency.  You may be able to do it directly or you may need a comparator.  If you have an oscilloscope you can see the frequency directly.  Hope this helps.

Here's my go at a schematic for the circuit (I can't guarantee that it's correct!):

http://eliaselectronics.com/?attachment_id=927

 

That squiggly line represents the trace on the PCB that goes a round the edge.