Are you saying that when your PWM output is connected to a PWM controlled device, the control does not range linearly over the available PWM output's range?

Or, are you saying you have performed an academic study of the PMW output disconnected from any load and are seeing odd behavior at the ends of the range?

vlad.colesnic

Dear experts,

Please suggest an advice on an issue regarding the behaviour of the PWM output with variable Duty Cycle configured on a S7 1212C PLC.

See attached pictures with the configuration Q0.4 integrated 20kHz PWM output mapped to QW1010.

When measuring the output with the oscilloscope, setting a duty cycle of 50% we measure approximately the same value (50%) but, when approaching the limits (5% or 95%) the gap between the setpoint and the measurement is very high.

For example for a 5% duty cycle setpoint we measure around 16%. See attached various measurements.

Have you ever faced this issue? Any idea could help! Thank you!

Best regards,

Vlad Colesnic

When you are measuring duty cycle of the PWM output, what are you measuring while holding the PWM output % steady?

See below for (hopefully) better explanation of my question. 

     ____      ____
____|    |____|    |____

    ^    ^    ^
    |    |    |
    |    |    -------- t2
    |    ------------- t1
    |----------------- t0

ON  TIME   = t1 - t0
OFF TIME   = t2 - t1
PERIOD     = t2 - t0 = ON TIME + OFF TIME
Duty Cycle = ON TIME / PERIOD * 100%

vlad.colesnic

Hello!

Thank you for your reply! It is both of them. In my project I am connecting the output to an IGBT driver. I am controlling the driver easily using an external pwm generator with 20khz and variable duty cycle and it behaves very well (i measure everything with the oscilloscope). Then, I disconnect the pwm generator and I connect the pwm output of the plc to the IGBT driver and, measuring again with the oscilloscope, during the adjuatment of the duty cycle I find  differences between the setpoint of duty cycle in the plc and measured duty cycle with the oscilloscope. 

To be sure, I disconnect the IGBT driver and I only add a 1 kohm resistor to the output and measure the dutycycle on it. Same strange behaviour. Setpoint 50%, meas 50%; Sp 40 Meas 42; Sp 30 Meas 36; Sp 20 Meas 30 and so on.

I thought about a Hw problem so I replaced the plc with another  same model but the issue persists!