Question about fault control in 30F5015

stoyanoff · Joined: 20 Jul 2011 Posts: 375

Hi! I have a problem with 1 30F5015. I`m controlling IGBT bridge with it and I have a protection connected to FAULTA pin. So when I`m starting the bridge the big current consumption in the begging is disturbing the fault pin and activates the interrupt. In the INT_FAULTA I`m stopping the PWM outputs. So I need to make a simple software protection to avoid this ENP. BUT if I enter in this interrupt the fault control is stopping the pwm outputs immediately - Cycle-by-Cycle mode(30F5015 datasheet page 105). So could you give me an advice how to deal with this problem?
Thanks!

Mike Walne · Joined: 19 Feb 2004 Posts: 1785 Location: Boston Spa UK

Soft start.

Mike

stoyanoff · Joined: 20 Jul 2011 Posts: 375

Yes! But before I start the bridge the filter caps are charged to voltage above work values. So on first pwm pulses I`m decreasing the duration the dv/dt is too big and it creates a powerful EMP. As far as I remember in the older version in 18F4431 the fault pin is waiting for few cycles and if the port is low again it triggers the interrupt.

Ttelmah · Joined: 11 Mar 2010 Posts: 19515

Generally a current limit fault, _wants_ to trigger if the current exceeds the safe level, without exception. You say you are 'decreasing the duration'. Is it the minimum it can be?. The normal practice would be to program the shortest pulse possible, and send this for a number of cycles before starting the PWM control.
If the fault is on 'per cycle', you will get a truncated pulse from the PWM, which will stop where the current limit exceeds, which is ideal for start up.
Fault doesn't delay on the 4431.

Best Wishes

stoyanoff · Joined: 20 Jul 2011 Posts: 375

Is there a way to trigger only the Fault interrupt

Ttelmah · Joined: 11 Mar 2010 Posts: 19515

These are not separate.
The fault interrupt, triggers the outputs to go to a safe condition.
No separate 'interrupt'.
You can turn off the switching to the safe condition in the PWM setup. and effectively use the fault input as a 'normal' interrupt, but then your driver risks overloading.

Best Wishes

stoyanoff · Joined: 20 Jul 2011 Posts: 375

OK! I want to suppress my EMP disturbance by adding a small delay (1us) and checking again the fault input. If it`s low again -starting the protection. There is one detail - how can I check if the pwm counter is in active state or is inactive?
Thanks!

notbad · Joined: 10 Jan 2013 Posts: 68

are you having a problem with charging of capacitors?
or the current surge when starting the motor?
if you increase the duty cycle slowly from zero there should be no problem.

stoyanoff · Joined: 20 Jul 2011 Posts: 375

I've tried this - no success! I added duty cycle correction depending on the voltage on the bridge - no effect. So I need to know how can I check if the pwm pulse is in active state (1) or is in inactive state(0) till it's counting?
Thanks!

notbad · Joined: 10 Jan 2013 Posts: 68

You can check if timer value is greater than ccpr.
But I think you're going in wrong direction.
Are you sure your circuitry for current threshold is correct?

Ttelmah · Joined: 11 Mar 2010 Posts: 19515

Agreed.

I'd actually suspect something really dangerous is going on (like both sides of the bridge switching on at once), and as such the fault input is saving things....

Best Wishes