- Ask a related questionWhat is a related question?A related question is a question created from another question. When the related question is created, it will be automatically linked to the original question.
Tool/software: Code Composer Studio
Is a sample code available for High Resolution Period and High Resolution Deadband control for a Delfino TMS320F28379D. I have tried to modify the example code hrpwm_deadband_sfo_v8.c without luck. I need to operate a full bridge inverter (H-bridge with the second bridge 180 degree phase shifted) at 13.56 MHz with 50% duty cycle and deadband control. The frequency and deadband need to be able to be finely adjustable with the high resolution of the Delfino controller. Thank you.
Hi Devin,
Yes, I could not get the high-resolution period control to work. I have abandoned trying to modify the example code. Instead, I'm using simulink to build the code. I have been able to generate a PWM with an average frequency of 13.57 MHz (shown in black in the attached images), but a 5 ns jitter is seen that changes the instantaneous frequency and duty cycle to 13.36 MHz and 40.78% (shown in blue in the attached image) and 14.33 MHz and 43.93% (shown in blue in the other attached image). I have attached the simulink model and c code generated. Is there a selection in simulink or a line needed in the c code to alleviate this?
Hi Brandon,
You can get jitter if you generate a sync out pulse. It looks like in your code, SYNCOSEL = 3, which means that the sync out signal is controlled by TBCTL2.SYNCOSELX. I didn't see the configuration of this register in the code you attached, so you might want to check if the code is setup to generate a sync out on CMPC or CMPD.
Hi Devin,
Thank you very much. I made SYNCOSEL = 0 by changing in the Simulink model in the ePWM7 block under the General tab, setting Synchronization output (SYNCO) to 'Pass through (EPWMxXYNCI or SWFSYNC)'. Doing that plus setting under the HRPWM tab the High resolution PWM (HRPWM) Load mode on ePWMxA: 'Counter equals to zero (CTR=ZERO)', High resolution PWM (HRPWM) control mode on ePWMxA: 'Duty or Period control mode', and High resolution PWM (HRPWM) edge control mode on ePWMxA: 'Rising Edge', fixes the jittering frequency. All that is left the duty cycle is jittering from 37% to 43% (shown in the attached figures) even though it is set at 50%. Any suggestions on how to resolve this, preferably in simulink? Thanks so much!
Hi Brandon,
13+ MHz is definitely pushing what this module is designed to do. Does this run OK if you reduce the frequency by a factor of 10?
In the TRM, we loosely state that 5MHz is about the max for 100MHz ePWM clock.
Hi Devin,
I ran the controller at 1.355 MHz and the PWM looks much better. I noticed there is still a jitter on the PWM duty cycle, 48.82% to 49.41%, as shown in the images. This of course is negligible. I zoomed in on the waveform though and it looks just like the jitter observed at 13.56 MHz, as shown in the image. Is this jitter always present, but is just negligible at lower operating frequencies? Thank you for looking into this.
