I made some mods to the schematic per comments.
*changed transformer to isolate the output power from the communication power
*Added two HCPL-2201 ICs so the PIC can communicate with the outside world.
The PCB Artist file is attached. I am unable to attach the PDF. It seems that our forum will only allow a 19.5kB PDF file and my optimized file is 117kB. I tried to convert to a JPG but it was too fuzzy and unreadable.
edit: attachments removed, new ones can be found below
hi Firegod. The design looks great only one concern though. You are running Diodes in series with the outputs of the PIC to the Gates of the IGBT's. Im wondering if you will get any issues with noise or slow turn off issues with them as the capacitance of the gates may cause a slow turn off of the IGBT's. I would have put a resistor to gnd on each gate.
What frequency are you PWM the IGBT's with?
Wombat
Wombat wrote:hi Firegod. The design looks great only one concern though. You are running Diodes in series with the outputs of the PIC to the Gates of the IGBT's. Im wondering if you will get any issues with noise or slow turn off issues with them as the capacitance of the gates may cause a slow turn off of the IGBT's. I would have put a resistor to gnd on each gate.What frequency are you PWM the IGBT's with?
Wombat
If he is using IGBT's (it might just be the symbol of choice to get the pacakage he wants), the device technically will never shut off once turned on. Leakages might do it after a while...but definitely not at a quick rate. Because of the higher capacitance on a IGBT, they often require very STRONG gate drives in order to cause them to turn on/off to minimize switching losses (which can be severe). For lower currents a 200 mA gate drive should be enough to minimize switching losses (keep them at sub 10% of continuous losses)
