The datasheet figures of trr/qrr are completely unrealistic for class D audio. First they are rated for very long forward conduction befor recovery, in reality the forward charge must build up so that the recovery charge happens in the other direction.
If you don't let the steady conduction of body diodes, the trr is always shorter than the dead time.
Secondly the 100A/us slope is unrealistically slow.
Third thing is you must take into account the soft switching phenomenon which happens here and there.
Forget about silly equations, like losses = Vsupply*Qrr*fsw
The figures are only good for direct faster/slower comparison.
If you don't let the steady conduction of body diodes, the trr is always shorter than the dead time.
Secondly the 100A/us slope is unrealistically slow.
Third thing is you must take into account the soft switching phenomenon which happens here and there.
Forget about silly equations, like losses = Vsupply*Qrr*fsw
The figures are only good for direct faster/slower comparison.
Qrr is a good starting point, but be careful, because it is specified at different forward current, and Qrr increases with If! IRFB4227 is specified at 46 A, while is IXFH40N50Q2 specified at 25 A. So IRFB4227 is not only 1000nC/420nC = 2.4 times better, but approximately 4 times.
Diode bypass is compulsory at this high voltage.
Although I used 1 uC device (PSMN070-200) without any trick in a ClassD amp, at 250 kHz, but efficiency was quite low (around 80-85 %).
Diode bypass is compulsory at this high voltage.
Although I used 1 uC device (PSMN070-200) without any trick in a ClassD amp, at 250 kHz, but efficiency was quite low (around 80-85 %).
I was reading all the posts that was written so far and I think that I didn't say it clearly what my intention is.
I want to make a class-D amp operating just in the LF range. More precisely to use for Sub woofer application, let's say with a frequency range of 10Hz - 250 Hz.
The choice of using IGBT was, as far I remember, when the post started there was no one with an working IGBT amp in Class D. And I thought that IGBT was more robust to handle the high current and voltage ratings needed for such an amp.
(And please don't ask again for what I need multiple kW to drive a Loudspeaker)
What I'm thinking...
- for this specific application I think that I don't need to worry much about ultra low THD.
- using low switching frequency, lets say 35kHz, seems to me to be enough
- High efficiency that will be my target
- And trying to have low collateral damages, with the fewest IGBT's blow ups as possible
The positive inputs I have still now...
- IXYS IGBTS IXGH60N60C2
- 3-level PWM
- IR2110 or 2113
I want to make a class-D amp operating just in the LF range. More precisely to use for Sub woofer application, let's say with a frequency range of 10Hz - 250 Hz.
The choice of using IGBT was, as far I remember, when the post started there was no one with an working IGBT amp in Class D. And I thought that IGBT was more robust to handle the high current and voltage ratings needed for such an amp.
(And please don't ask again for what I need multiple kW to drive a Loudspeaker)
What I'm thinking...
- for this specific application I think that I don't need to worry much about ultra low THD.
- using low switching frequency, lets say 35kHz, seems to me to be enough
- High efficiency that will be my target
- And trying to have low collateral damages, with the fewest IGBT's blow ups as possible
The positive inputs I have still now...
- IXYS IGBTS IXGH60N60C2
- 3-level PWM
- IR2110 or 2113
"The choice of using IGBT was, as far I remember, when the post started there was no one with an working IGBT amp in Class D."
I ve one self-built 20kHz IGBT amp running on 600V Bus (APT 1200V thunderbolt IGBTS) since late 2002 (!) - driving 2 x 18" 8Ohm subs - still working but finally retired and replaced by XTI6002.
http://www.diyaudio.com/forums/class-d/87913-class-d-amp-photo-gallery-23.html#post2170284
I ve one self-built 20kHz IGBT amp running on 600V Bus (APT 1200V thunderbolt IGBTS) since late 2002 (!) - driving 2 x 18" 8Ohm subs - still working but finally retired and replaced by XTI6002.
http://www.diyaudio.com/forums/class-d/87913-class-d-amp-photo-gallery-23.html#post2170284
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.