Dear all,
As a fun project I set out to build a Class D amplifier (with some spare parts from previous projects) for my DIY sub woofer. First step was to create something functional and build furher on this.
I started with the IR2113 but since I had doubts about the stability of my dead time RC solution I switched to the IR21844 for 400 nS stable dead time. Please ignore the LTC7060 chip in below schematic (I took it as it was closest functionally for some initial simulation).
For now I hooked up the double bench power supply (current limited at 0.5 A) and a 100 Ohms load as speaker. Vcc is 12 V, PWM is 100 kHz 50% from a function generator. The unidentified diodes in the schematic are STTH3R02's. The Inductor is a 1 mH 2 A choke in Common mode (https://www.distrelec.biz/Web/Downloads/_t/ds/7446122001_eng_tds.pdf).
The problem starts when I hook up de LC filter with (or without) load and increase the supply voltage above 25V (while the simulation is working just fine). It seems like SW and TG are rising and falling more and more quickly within a period of 5 seconds up until heavy ringing occurs and the current limit kicks in. Looking at the scope both Mosfets seem to start causing shoot through due to the ringing but maybe somebody can explain better.
So far I tried to reduce the switching speed, use large dead time, moved from bread board to a soldered pcb, tried IRF610 Mosfets instead (lower gate charge would help?), use a simple air Inductor (18 uH) but still no luck.
With my limited knowledge it seems impossible to figure out the source of this issue. Any tips what could be the source of this issue and an explanation of the behavious of SW and HO with the LC filter attached would be much appreciated!
Orange: high side Vgs
Green: low side Vgs
Red: supply
Blue: SW
Yellow: HO
No load and no LC filter connected:
14V with LC and load, HO and SW are inceasing earlier now:
27V
27V a little later just before failure
Without the Math trace
As a fun project I set out to build a Class D amplifier (with some spare parts from previous projects) for my DIY sub woofer. First step was to create something functional and build furher on this.
I started with the IR2113 but since I had doubts about the stability of my dead time RC solution I switched to the IR21844 for 400 nS stable dead time. Please ignore the LTC7060 chip in below schematic (I took it as it was closest functionally for some initial simulation).
For now I hooked up the double bench power supply (current limited at 0.5 A) and a 100 Ohms load as speaker. Vcc is 12 V, PWM is 100 kHz 50% from a function generator. The unidentified diodes in the schematic are STTH3R02's. The Inductor is a 1 mH 2 A choke in Common mode (https://www.distrelec.biz/Web/Downloads/_t/ds/7446122001_eng_tds.pdf).
The problem starts when I hook up de LC filter with (or without) load and increase the supply voltage above 25V (while the simulation is working just fine). It seems like SW and TG are rising and falling more and more quickly within a period of 5 seconds up until heavy ringing occurs and the current limit kicks in. Looking at the scope both Mosfets seem to start causing shoot through due to the ringing but maybe somebody can explain better.
So far I tried to reduce the switching speed, use large dead time, moved from bread board to a soldered pcb, tried IRF610 Mosfets instead (lower gate charge would help?), use a simple air Inductor (18 uH) but still no luck.
With my limited knowledge it seems impossible to figure out the source of this issue. Any tips what could be the source of this issue and an explanation of the behavious of SW and HO with the LC filter attached would be much appreciated!
Orange: high side Vgs
Green: low side Vgs
Red: supply
Blue: SW
Yellow: HO
No load and no LC filter connected:
14V with LC and load, HO and SW are inceasing earlier now:
27V
27V a little later just before failure
Without the Math trace
Attachments
Last edited:
. TG would be Top Gate or HO and SW the (SWitching) output I guess.
