Re-Inventing The Wheel-Part1-Clemente_Figuera-THE INFINITE ENERGY MACHINE

[conradelektro]

See the attached schematics for the test system I will build.

For an initial test I will wind all tree coils (primary 0 - secondary - primary 1) on a Ferrite rod next to each other (the secondary will be sandwiched between the two primaries).

Greetings, Conrad

http://www.adafruit.com/blog/2012/07/24/new-product-mcp4725-breakout-board-12-bit-dac-wi2c-interface/
http://at.farnell.com/texas-instruments/drv8834pwp/treiber-motor-dual-h-bridge-24htssop/dp/2115234
http://at.farnell.com/texas-instruments/msp430g2452in20/mcu-16bit-8k-flash-20pdip/dp/1865383
http://at.farnell.com/texas-instruments/msp-exp430g2/entwicklungskit-msp430-launchpad/dp/1853793

[TinselKoala]

There is a thing called a "motor controller shield" that snaps onto an Arduino, that incorporates 2 L293D quad half-H high current motor drivers onboard, plus breakout connectors. This shield interfaces between the Arduino proper and your motors. It will drive 2 RC servos, or 2 standard stepper motors, or 4 regular DC motors by PWM. You can run the motors off the Arduino power bus or from a separate power supply to the H-bridges. The 293s are in the standard 16-pin DIPs and are in sockets, so you don't have to mess with SMDs and hassle a lot when you blow a driver chip.

I have one that I use with the Osepp arduino clone. It works great for motor control and I suppose you can program any kind of signal to the two L293 drivers you like, maybe.

makeshields.com "full function motor control shield for arduino"

[conradelektro]

There is a thing called a "motor controller shield" that snaps onto an Arduino, that incorporates 2 L293D quad half-H high current motor drivers onboard, plus breakout connectors. This shield interfaces between the Arduino proper and your motors. It will drive 2 RC servos, or 2 standard stepper motors, or 4 regular DC motors by PWM. You can run the motors off the Arduino power bus or from a separate power supply to the H-bridges. The 293s are in the standard 16-pin DIPs and are in sockets, so you don't have to mess with SMDs and hassle a lot when you blow a driver chip.

Unfortunately the L293D is completely useless for micro stepping, it just can switch the transistor bridges on and off (no in between steps, no chopping).

This is what it is all about:

http://www.stepperworld.com/Tutorials/pgMicrostepping.htm

And it is even more complicated in case one wants that the phase difference is not locked at 90°.

Look at the circuit I posted in my last post. This is what one needs. Of course one can do it in many ways, but the circuit shows the principle, it shows the capabilities needed. Whatever circuit one builds or finds, it should match or even outperform the capabilities of my example circuit.

The components in my circuit are all very modern and therefore consume little power by themselves, which is essential for a self runner. There are microprocessors which have on board DA - converters (one needs two), but they need considerable power (in comparison to the MSP430G2xxx series). AD - converters (for sensing applications) have become common in microprocessors, but DA - converters integrated into microprocessors are still rare.

When looking for stepper motor drivers (or boards), look for at least 8 MICRO STEPS. The emphasis is on MICRO (in between steps). For these micro steps one has to vary the CURRENT through the coils along a Sinus wave form with a 90° phase difference. And only a few stepper motor drivers allow to even change the 90° phase difference and to realise any wave form for the CURRENT. Of course, all drives which can vary the current (which can do MICRO steps), do it by chopping.

An alternative way would be to just use two full transistor bridges and to do the chopping and phase control with the micro processor. But to do the chopping in any useful and fairly clean way, a lot of additional circuitry around the full bridges is necessary which I would hate to have to design. This is the reason why I looked to the stepper motor driver ICs, where they have solved all the little problems of "chopping" in the last 10 years.

I continue the (probably unwanted) lecture, just to make everybody understand:

To beat the "Figuera commutator + 7 resistors" one needs a stepper motor driver IC or board which can do at least 8 MICRO steps (driving two coils). For higher voltages and higher currents one can find stepper motor driver ICs, which can do only one coil each, but they start to have limitations (only 4 or 2 MICRO steps).

One can of course design a new "chopper circuit" based on full transistor bridges, but I wish you good luck with that, specially if you want high voltage and high current. Why do you think are the high power stepper motor drivers which can do at least 8 MICRO steps rather expensive and rarely versatile?

I want to test the Figuera idea, but I do not want to go into "chopper circuit design". "Chopper circuit design" has been solved in connection with stepper motor drivers which can do MICRO stepping (8, 16, 32 MICRO steps or even a lot of MICRO steps along an analogue wave form like the DRV8834 can do).

Greetings, Conrad

P.S.: I promise to stop my lecturing about stepper motor drivers and chopper circuits.

[norman6538]

Where is the original Figuera work? I can't seem to find that.
Perhaps another group?

Norman

[bajac]

Norman,

A sketch of the patent can be found in this Spanish article:
http://www.alpoma.net/tecob/?p=4005

And, this is a poor English translation of that article:
http://orbo.es/?p=26

Bajac