On Tue, 27 Apr 2010, Andy Pugh wrote: > Date: Tue, 27 Apr 2010 20:51:22 +0100 > From: Andy Pugh <[email protected]> > Reply-To: "Enhanced Machine Controller (EMC)" > <[email protected]> > To: "Enhanced Machine Controller (EMC)" <[email protected]> > Subject: Re: [Emc-users] Microcontroller motor drive [Was: Resolver to > Quadrature Convertor] > > On 27 April 2010 17:25, John Kasunich <[email protected]> wrote > >> On Mon, 26 Apr 2010 11:22 +0100, "Andy Pugh" <[email protected]> > >>> The power modules I have been looking at recently >>> seem to work best at about 10kHz PWM frequency, well within the >>> capability of EMCs existing software PWM generation. >> >> 10kHz is not really within the capability of software PWM. > > You beat me to it, the error in my reasoning occurred to me on the > plane this morning. > 10kHz update frequency is easy, a useful 10kHz PWM is a rather different > thing. > It is also rather out of the scope of the AVR microcontrollers, which > have a max PWM rate of (Base Clock / 512) > > I should also point out the other error I made earlier in the thread. > Holding the low-side drivers of that module I pointed out low all the > time will not have the effect I said, it does need 6 drive signals. It > might be possible to drive it with a logic-level inverter. I confess I > am rather unsure what a sinusoidal three-phase PWM signal looks like, > ie what the relationship is between high-side active and low-side > active for each phase. > > -- > atp
Its can be pretty simple, at idle (0V) the high and low sides are ~50% duty cycle square waves (the low side being the inverse of the high) the deadzone is subtracted from the active widths to prevent "shootthrough" so in a actual system the high and low side drives might be active 48% of the time or so at idle. Voltage output across the motor windings is proportional to the difference in the three PWM values (deviation from idle = 50 %) and the DC bus power. The three PWM values for sinusoidal drive are just three sine wave values with 120 degree phase differences with magnitude determined by the desired motor voltage and offset 1/2 way up so 0 = 50 % duty cycle PWM. (glossing over second order troubles like deadzone compensation) Often theres a limitation on full scale (high side on) PWM values for systems that use boostrap type gate drivers (Small IGBT modules often use boostrapped gate drivers) so max PWM may be 98% or so. Minimum PWM (low side on) is not limited since the high side gate drivers are charged when the low side is on. Peter Wallace ------------------------------------------------------------------------------ _______________________________________________ Emc-users mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/emc-users
