The following two oscilloscope captures show the switching performance when a fast MOSFET and a heavy IGBT are connected in parallel. Both are turned on at the same time and the IGBT is turned off 500ns before the MOSFET in order to allow enough time for the current tail to extinguish.
MOSFET turn-off detail at 20ns/div:
Blue trace is Vce (and Vds) at 100V/div and it shows a 20ns crossover time for an over 600V swing
Red trace is Vge at 5V/div and shows how the fast turn-off transient tries to pull up the gate of the IGBT (despite the 2A driver IC and the 1A turn-off schottky paralleled to the gate turn-on resistor). The current being switched is around 14A.
IGBT turn-off detail at 200ns/div:
Blue trace is now Vce (and Vds) at 5V/div and it shows how the total voltage drop increases dramatically when the IGBT is turned off and the MOSFET is left conducting alone. Red trace shows IGBT Vge at 5V/div as in the previous capture, note the 500ns anticipated turn-off. The current being switched is also 14A, the Rds-on of my MOSFET at 25ºC is 0.16 ohms (over 0.4 ohm for 125ºC
), and the Vce-sat of my IGBT is around 1.6V for Ic=20A and 2.1V for Ic=40A over most of the temperature range (a heavy one 
).
Implementation is very simple thanks to gate driver ICs, as opposed to what some could think. I'm currently testing this approach as part of a 3KW boost PFC pre-regulator, but I will implement it soon on my high voltage +-180V +-30A class D amplifier. Efficiency seems quite improved...
Innovate and enjoy...
MOSFET turn-off detail at 20ns/div:
An externally hosted image should be here but it was not working when we last tested it.
Blue trace is Vce (and Vds) at 100V/div and it shows a 20ns crossover time for an over 600V swing
Red trace is Vge at 5V/div and shows how the fast turn-off transient tries to pull up the gate of the IGBT (despite the 2A driver IC and the 1A turn-off schottky paralleled to the gate turn-on resistor). The current being switched is around 14A.IGBT turn-off detail at 200ns/div:
An externally hosted image should be here but it was not working when we last tested it.
Blue trace is now Vce (and Vds) at 5V/div and it shows how the total voltage drop increases dramatically when the IGBT is turned off and the MOSFET is left conducting alone. Red trace shows IGBT Vge at 5V/div as in the previous capture, note the 500ns anticipated turn-off. The current being switched is also 14A, the Rds-on of my MOSFET at 25ºC is 0.16 ohms (over 0.4 ohm for 125ºC
), and the Vce-sat of my IGBT is around 1.6V for Ic=20A and 2.1V for Ic=40A over most of the temperature range (a heavy one ).Implementation is very simple thanks to gate driver ICs, as opposed to what some could think. I'm currently testing this approach as part of a 3KW boost PFC pre-regulator, but I will implement it soon on my high voltage +-180V +-30A class D amplifier. Efficiency seems quite improved...
Innovate and enjoy...
Yupp... I like it!
Basically we could have similar low conductive losses with two paralleled MosFets, too. But then the turn OFF transition would suffer quite badly from double reverse capacity...
Your approach has just for very short time full load in the MosFet. Great!
I guess you could use even a smaller MosFet in your circuit without much trouble...
Definitely a promising approach for high power fast switching!!! To be honest this idea is by far better than most patents.... ...only few golden nuggets there...
Basically we could have similar low conductive losses with two paralleled MosFets, too. But then the turn OFF transition would suffer quite badly from double reverse capacity...
Your approach has just for very short time full load in the MosFet. Great!
I guess you could use even a smaller MosFet in your circuit without much trouble...
Definitely a promising approach for high power fast switching!!! To be honest this idea is by far better than most patents.... ...only few golden nuggets there...
ChocoHolic
Hehehe...! By what I have seen in the Patent office is not difficult to have many thousands of different patents on the some circuit…
Yes I have a few patents but none about electronics.
I just posted the circuit to encourage Eva. I know that it works and is a very good and simple solution.
And as far as I know it is not patented…
As pointed out both are turned on at the same time and the IGBT is turned off xxxns before the MOSFET in order to allow enough time for the current tail to extinguish.
Hehehe...! By what I have seen in the Patent office is not difficult to have many thousands of different patents on the some circuit…Yes I have a few patents but none about electronics.
I just posted the circuit to encourage Eva. I know that it works and is a very good and simple solution.
And as far as I know it is not patented…
As pointed out both are turned on at the same time and the IGBT is turned off xxxns before the MOSFET in order to allow enough time for the current tail to extinguish.
Attachments
Hei Fredos...!!!
Are you trying to create me problems?
I never said or even suggest that the sch was mine.
Besides the partial sch I have posted directly infringes this patents here http://www.diyaudio.com/forums/showthread.php?postid=1083861#post1083861
Are you trying to create problems to PowerSoft?
I don’t know from where that sch comes from.
Only thing I know is that has some years old and ideas like that one coming from other people makes me a humble guy...
Other thing I know is that I know nothing… and dangerous comments of the kind of your post discourages any future posting from myself.
Are you trying to create me problems?
I never said or even suggest that the sch was mine.
Besides the partial sch I have posted directly infringes this patents here http://www.diyaudio.com/forums/showthread.php?postid=1083861#post1083861
Are you trying to create problems to PowerSoft?
I don’t know from where that sch comes from.
Only thing I know is that has some years old and ideas like that one coming from other people makes me a humble guy...
Other thing I know is that I know nothing… and dangerous comments of the kind of your post discourages any future posting from myself.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.