Hey everyone!
I am currently designing a 6 channel class D amplifier with DSP using the TPA3255 chips. One thing I am really concerned about with class D is the possibility of damaging the amplifier when no load is connected to the output of the LC filter. On the forums, I heard of people using load resistors to keep the amp loaded to some degree or using catch diodes to clamp the maximum LC filter to the voltage rails (PVDD and GND) so that the fly back effect does not go back in to the amp and damage the output stage of the IC. I am most keen on using the diode protection method, but where do I place the diodes? Do I place it before or after the LC filters? I seen Wondom amplifier boards place the catch diodes after the LC filter right before the outputs, but I also seen amps like the TDA7498E that place the catch diodes before the LC filters. Any suggestions?
I am currently designing a 6 channel class D amplifier with DSP using the TPA3255 chips. One thing I am really concerned about with class D is the possibility of damaging the amplifier when no load is connected to the output of the LC filter. On the forums, I heard of people using load resistors to keep the amp loaded to some degree or using catch diodes to clamp the maximum LC filter to the voltage rails (PVDD and GND) so that the fly back effect does not go back in to the amp and damage the output stage of the IC. I am most keen on using the diode protection method, but where do I place the diodes? Do I place it before or after the LC filters? I seen Wondom amplifier boards place the catch diodes after the LC filter right before the outputs, but I also seen amps like the TDA7498E that place the catch diodes before the LC filters. Any suggestions?
Catch diodes before the filters will protect the ic but not the filter caps. At certain frequencies an unloaded LC will produce voltage GAIN.
Thus taking out the caps.
Thus taking out the caps.
Placing these diodes before the filter would simply be 'redundant' as the MOSFETs already have body diodes built into their structure. I would place them after.
This is a known issue in 'pre-filter feedback' amplifiers.
I heard of body diodes somewhere else too (tpa3116 forums I believe) but if I placed them after the filter, should I place it before the zobel network or after the zobel. As you mentioned pre filter feedback, would I be able to crank my volume to the absolute max on unloaded channels as well without damage? I want my class D amp to behave like a class AB in terms of functionality. By the way my amp design does not use PFFB