Go Down

Topic: Controlling a mendocino motor (Read 710 times) previous topic - next topic

Tophness

Feb 11, 2018, 09:27 pm Last Edit: Jul 03, 2018, 05:07 am by Tophness
I'm trying to control a mendocino motor so that I can turn it to a particular position like a brushless dc motor, and then turn off and switch to using the panels to charge a battery or maybe a small supercapacitor capable of running a 3.3v arduino that can be used to control the position in future.

The problem is, the motor is designed to be always turning in order to generate the pulses for commutation, and there isn't enough power to run an arduino to start with, since there are 4 panels of 0.5v @ 0.05A each, each 2 wired in series to make 1v which runs 1 of the 2 coils, so running an arduino or step up converter would probably draw too much power to even make it turn, unless it first uses the solar panels to charge and then switches to letting it create an EMF normally.
I'm also just not sure what kind of motor this would be classified as. I think it's a 2 phase brushless inrunner? But apparently the solar panels are used like brushes.

Can I do this with a mosfet and a reed switch/ phototransistor and control the coils like an electromagnet?

Attached is the wiring diagram for the mendocino motor.

Robin2

Images from Original Post so we don't have to download them. See this  Simple Image Guide





...R
Two or three hours spent thinking and reading documentation solves most programming problems.

Robin2

I did not (do not?) know what a Mendocino motor is so I had a look on Wikipedia.

Is this tinfoil-hat territory ?

...R
Two or three hours spent thinking and reading documentation solves most programming problems.

jremington

#3
Feb 12, 2018, 03:45 am Last Edit: Feb 12, 2018, 03:50 am by jremington
Quote
the motor is designed to be always turning in order to generate the pulses for commutation
There is no commutation in a typical Mendocino motor. The PV cells are simply connected to the coils, and supply different amounts current depending on the light intensity at each cell. The permanent magnet(s) is arranged such the the cell receiving the most light forces the motor to turn away from the light source.

If you want to power an Arduino with solar energy, study this great tutorial: https://www.gammon.com.au/forum/?id=12821

Tophness

#4
Feb 12, 2018, 06:10 am Last Edit: Feb 12, 2018, 06:48 am by Tophness
If you want to power an Arduino with solar energy, study this great tutorial: https://www.gammon.com.au/forum/?id=12821
The problem with that is, it's not simply power coming from solar cells, the same wire coming from the solar panels are used to create an electromagnetic field used to push the motors.
I don't know what that would do to the arduino powering it straight from the same wire.

I could grab it before it reaches the coils, but then it would need some way of feeding back out to the coils. How do I make it switch back to powering the coils, and control the switching speed/direction?
An H-bridge driver needs too much voltage and would probably heat up the wires.
Would a mosfet work?
And could I power the mosfet using the solar panels or do I need to add a battery into the mix too?

Robin2

The problem with that is, it's not simply power coming from solar cells,
Any "magic" that you add is going to reduce the available power. Otherwise we would have perpetual motion and free electricity.

If you want to power an Arduino with a solar cell then use the solar cell to charge a battery and power the Arduino from the battery.

Of course this assumes that over the course of a day the solar cell generates enough energy to power the Arduino and make up for energy losses in the battery. If not, then you need a bigger solar panel.

...R
Two or three hours spent thinking and reading documentation solves most programming problems.

Tophness

#6
Feb 12, 2018, 12:49 pm Last Edit: Feb 12, 2018, 12:59 pm by Tophness
Any "magic" that you add is going to reduce the available power. Otherwise we would have perpetual motion and free electricity.

If you want to power an Arduino with a solar cell then use the solar cell to charge a battery and power the Arduino from the battery.
I don't subscribe to the "free energy" thing, I just happen to have an application that takes very light loads, and although it has very low torque, it can go pretty high RPM using very little power, and the magnets and coils would be perfect for integrating into a 3d print rather than mounting a whole motor, but I don't just want it to continuously spin, I want to control it to use it like a gimbal.

I can easily buy a ready made step up regulator to charge a battery and arduino, the problem is, at that point, the original pulses through the coils from the solar panels would have to be replicated with the arduino, or the arduino would have to switch the power from the solar panels to the coils coils on and off, and I'm not sure how to do that without cutting off the power to the batteries/arduino or short circuiting it (because of the EMF generated inside the coils).

jremington

Quote
The problem with that is, it's not simply power coming from solar cells, the same wire coming from the solar panels are used to create an electromagnetic field used to push the motors.
Not a problem. That is how all electric motors work.

Tophness

#8
Feb 12, 2018, 05:53 pm Last Edit: Feb 12, 2018, 06:08 pm by Tophness
Not a problem. That is how all electric motors work.
So if I tap into the same lines going to the coils to charge a battery, I'll just get whatever power is left? And I don't need to do any kind of isolation?
What about controlling the speed and direction of the coils? How can I control it if I'm charging from the panels at the same time?

jremington

Best to use separate PV cells to charge a battery. Great bargains here

Tophness

#10
Feb 13, 2018, 08:23 pm Last Edit: Feb 19, 2018, 05:18 am by Tophness
Best to use separate PV cells to charge a battery. Great bargains here
I understand it would be easier and I would get more power out of the panels using them directly, but that's not what I'm trying to do.
I've come up with another couple of ideas:
Using the EMF in the coils as an inductive charger and then separately drawing the power from them using another coil like a Qi charger, but controlling the rate it's being drawn in order to control the rate the coils can spin,
OR
Using the solar to power an RF relay and then outputting to the coils so I can control it remotely (this would be perfect for what I'm trying to do), but I'm not sure if there are even 1V versions of them
OR
Using larger voltage panels (this would be most efficient), so that there doesn't need to be any boosting circuit, and maybe control it with an ESC,
But then I'm not sure how to adjust the coils so they don't burn out.
Do it need more windings of the same coil or thicker wires with less windings? Can the magnets stay the same size?

I've bought a pre-made mendocino motor because I wanted to start with something that works. I read about people constantly struggling to adjust the magnet positions and such and wanted to skip that step.

I'd also like to support the end that's simply grinding into glass with a magnet or a bearing. Will this introduce cogging torque?

MarkT

#11
Feb 15, 2018, 01:49 pm Last Edit: Feb 15, 2018, 01:50 pm by MarkT
Your main problem is the 1.0V maximum voltage available - not many electronic devices can function at
that low a voltage.  Simple solution is change to 6V panels and use thinner wire for the windings.
Failing that you need a boost converter that runs from 1.0V input.  Bit specialist, and definitely only
available in SMT packages like: http://www.ti.com/lit/ds/symlink/bq25504.pdf
And a schottky bridge rectifier is needed to convert the motor ac to dc, losing most of your
voltage to start with.

6V panels are much much easier!
[ I will NOT respond to personal messages, I WILL delete them, use the forum please ]

Tophness

Thanks, I've already ordered 6V/200ma panels a few days ago cos I think you're right. Creating a new mendocino with bigger panels will probably be easier than getting a 1v one to convert efficiently.

As for converting the motor from AC to DC, wouldn't it be more efficient to charge from the solar panels directly and control the flow of electrons to the coils rather than having the panels run the coils and then have the coils convert back into DC?

The latter would achieve what I want easier - for the panels to flow to the coils at the natural rate to spin the motor, but if there's some way I can create a switch in between charging a battery to run an arduino to running the solar panels, and then controlling the on/off and direction to the coils I'd love to hear it.

I can't get my head around the circuit so I'm trying to make it as simple as possible to avoid short circuiting something and exploding a battery.
I think a ULN2003LV would work since it doesn't need huge transistors or fets to run the coils, but I'm guessing I need flyback diodes and such, and then there's the problem of the same power source (the solar panels) being used for the ULN2003 switching being the same power source connected to the battery charger/arduino, so switching a load on and off with it doesn't sound safe. I think there should be an additional switch to disconnect the panels to the battery when it's time to start controlling the coils.. I'm not sure how this would work.

Alternatively there are quad or dual SPDT switches that seem appropriate?
Maybe I could use the enable pin on the quad to only turn on when a battery charger (that automatically disconnects the source when a battery is full) signals it's ready.

Go Up