Feedback controller
NOTE this is not a pulse-controller!
(https://forum.009society.com/proxy.php?request=http%3A%2F%2Fhnorwood.co.uk%2Ffiles%2Fimages%2Ffeedback.gif&hash=9bfc6a1baf8540804aae278e25d66c14e288d0a7)
This circuit is based on the cassette tape drive speed stabiliser by Linsley Hood, but I have seen it used elsewhere so I'm not too sure who originated it. I came across it many years ago and about 4 or 5 years ago decided to see if it would work as a train controller.
I brought it along to the 2013 AGM, and Roger kindly let me try it out on his layout, after he mentioned that some feedback controllers were not kind to some modern motors, causing them to run erratically. This unit drove Roger's Graham Farish- based diesel around at a slow smooth crawl.
Now, I'm not too sure exactly how the circuit works, but this is my take on it:
Normally, the voltage at the top of the motor should be the same as at the base of the 2N3904. If the motor is under load, the voltage across it will drop as it tries to draw more current. This causes the emitter of the 2N3904 to drop, turning it further on, which further turns on the TIP42, allowing more current to the motor. As the motor comes off load, the reverse happens. (At least, that's what I think happens...)
The zener diode allows me to switch the controller to a lower output range. I put in the 7812 regulator to provide current limiting to the device.
Once again, this is presented as an exercise in circuit design and development. I do not make controllers except for my own enjoyment and curiosity.
I must give this circuit a try. It looks very interesting.
I think you hit the nail on the head in your interpretation of the circuit! Yes the voltage across the motor is applying negative feedback to the 2N3904 transistor.
I have seen circuits that take the bemf from the motor in a similar way and feed it back to a pulse generator so the feedback controls pulse width. I am not certain but I suspect that it is this type of feedback controller that causes problems with some motors.
There is another sort that doesn't use pwm but nonetheless disconnects the power transistor briefly and measures the bemf generated by the motor. In this way it is just the bemf that is being measured whereas in your circuit it is the sum of the bemf and the V = IxR from the current flowing through the motor.
Then you can get really clever and start looking at what the commutator is doing:
http://www.ideadvance.com/sites/ideadvance.com/files/files/WhitePaper2.pdf (http://www.ideadvance.com/sites/ideadvance.com/files/files/WhitePaper2.pdf)
If you are keen to try new circuits, have you considered using Yenka/Crocodile Clips (free for home use) to simulate circuits to complement practical experimentation?
http://www.yenka.com/ (http://www.yenka.com/)
We will have to twist your arm to join MERG: http://www.merg.org.uk/ (http://www.merg.org.uk/)