Hello everyone...

For some time now I've been thinking about developing a new way to control/command, sets of thyristors in anti-phase, using pulse width modulation (PWM), with zero-crossing synchronization (ZC).

In fact, I had already used such resources, in temperature control for large industrial electric furnaces, especially for heat treatment and aluminum casting. Photos of the equipment already in use are attached, as well as a brochure with the respective datasheet.

PS: The equipment is old and was produced in Brazil, therefore with information in Portuguese.

If you observe the details you will understand that the equipment received an analog signal, from the temperature controller (PID) with a normalized range of 4-20 milliamperes; then the signal would be compared to the internal ramp voltage, determining an enable signal for firing the thyristors (with two phases in anti-parallel), always at the zero crossing.

The idea is to apply the same concept (PWM + ZC) to the speed control of induction motors; replacing noisy thyristor circuits with phase control. I believe the control could enable each half-cycle of the mains voltage, in a sequence of up to 127 half-cycles; because the mechanical inertia of any induction motor, allows the use of a train of half-cycles without any problem; in addition to canceling that characteristic 60 Hz hum (or 50 Hz depending on the mains).

To date, the only specialized IC I know for integrating a thyristor firing control is the TCA785; and the application requires phase control, which I intend to avoid.

 

If the thyristor is always triggered at a zero crossing then how is the power going to be varied to adjust the speed. The issue of "Noisy Thyristors" is usually solved with a bit of filtering. Aside from that, thyristor motor speed control is a rather mature technology. Zero crossing switching works very well for temperature control of processes that do not change rapidly.. Motor speed control is a bit different.

 

yes this is already used to control loads... specially for large loads to fight interference that occurs when load switched on a. the idea is to turn the SCR just after crossing zero. then SCR stays on until next zero crossing. so during 1/2 of period, duty is 100% but one can choose which half periods (or periods) will be on. ideally one would shuffle them a bit - creating more or less evenly distributed patterns to reduce load variations. this can be easily done with small controller - read setpoint value and use counter to keep track of intervals that are on and off...

 

To control the speed of an induction motor you need to vary the frequency not the voltage. Switching the power on and off over a period of 2 seconds is probably the easiest way to destroy the motor.

 

If the thyristor is always triggered at a zero crossing then how is the power going to be varied to adjust the speed. The issue of "Noisy Thyristors" is usually solved with a bit of filtering. Aside from that, thyristor motor speed control is a rather mature technology. Zero crossing switching works very well for temperature control of processes that do not change rapidly.. Motor speed control is a bit different.

If the thyristor is always triggered at a zero crossing then how is the power going to be varied to adjust the speed. The issue of "Noisy Thyristors" is usually solved with a bit of filtering. Aside from that, thyristor motor speed control is a rather mature technology. Zero crossing switching works very well for temperature control of processes that do not change rapidly.. Motor speed control is a bit different.

Hello misterbill2... I understand your question, the difference in using phase control is that we start using something similar to frequency modulation, delivering more or less a half-cycle, over the time of a train of half-cycles. I am busy with other tasks, analyzing legal processes for technical expertise. But I will post a table explaining the concept, you won't regret waiting. I promise.

Yes, the noise must be treated by filters, however the audible hum, always at the mains frequency, remains.

I agree that the illustrated application works very well for thermal equipment, which has enormous inertia. However, I believe we can adapt the ZC concept to asynchronous induction motors.

 

However, I believe we can adapt the ZC concept to asynchronous induction motors.

We wait with bated breath.

 

yes this is already used to control loads... specially for large loads to fight interference that occurs when load switched on a. the idea is to turn the SCR just after crossing zero. then SCR stays on until next zero crossing. so during 1/2 of period, duty is 100% but one can choose which half periods (or periods) will be on. ideally one would shuffle them a bit - creating more or less evenly distributed patterns to reduce load variations. this can be easily done with small controller - read setpoint value and use counter to keep track of intervals that are on and off...

Exactly, panic mode. I've already committed to posting a table explaining how to generate this FM, over, for example, 127 half-cycles.

As I explained above, I will continue developing the table. We'll have news soon. Big hug and thanks for the motivation.

 

if you are going to work on it, perhaps make sure that you try it first using no randomness - just count half-cycles on zero pulse.
if counter reaches 100, reset it.
if count is less than preset do nothing.
if the counter is greater than preset, trigger SCR.
you can implement this in software (or hardware if you want that retro look...)

 

if you want randomness it is almost the same process (even without lookup tables)

reset counter and memory bits representing lookup table (100bits or 100 elements array)
generate 100bit random sequence (populate the lookup table or array bits)
on zero pulse count half-cycles
if the bit position (or array element) is set, trigger output, otherwise do nothing
if counter reaches 100, start from begin