circuit wrote:Why are there 3 current sensors instead of 2?
The controller looks OK as a testing platform, but I just can't see a product any time soon. I really want to, but can't.
Well, it's not my intention to build and sell complete controllers, at the moment I'm just in this for the
hobby. What I can sell is PCB's and programmed IC's so that people can build their own controllers. What John suggests
is very well possible. Or maybe someone who likes soldering and building can build and sell controllers based
on my IC's and PCB. My intention and interest is to develop my controller IC further and to include power
startup with full torque at motor standstill, without hall sensors. My interest is in the algorithm running on the
MCU, less in soldering everything together and building controller boxes. I build the box of the first post because
I want to try out my algorithm on a powerful motor, and as a demonstrator to show that the algorithm works.
About the current sensors, theoretically 2 sensors (lets say Ia and Ib) are enough because the current in the 3rd motor wire (Ic)
can be calculated with
Ic = - Ia - Ib
But typically the sensors pick up noise from their supplies etc, a noise which can be in a large part common in all sensors.
Ia = Ia_wanted + I_noise
Ib = Ib_wanted + I_noise
Ic = - Ia_wanted - Ib_wanted - 2 * I_noise
So with 2 sensors all the noise gets assigned to the 3rd motor wire.
With 3 sensors though the noise part common to all sensors (the I_noise) can simply be filtered out by the
algorithm running on the MCU....