Congratulations on the license and becoming "official". I assume this means you are in the Netherlands or perhaps Germany. Maybe I will hear you on the air via SDR.
Well at first I tested many Class-D modules. Some of them failed and some did not have very good response. Modulated RF output had some distortion (modulation triangle on my scope was rounded) at higher power levels etc oddities.
I have some Class-AB modules too so I tested them and all problems gone. Modulation is now very linear up to 100%.
OK. I understand. There are many low-quality Class-D amplifiers available out there. Keep in mind that this sort of transmitter requires that the module produce quite a bit of DC output (several Amps)
in addition to the audio output on top of that, just to keep the transmitter on the air. Without it the final transistor (in your case the IRF530) will not have sufficient drain voltage. I have found that not all the available Class-D modules, especially the inexpensive ones, can do this well, as it it is a bit above and beyond what it is supposed to do for just routine audio amplification. I believe that this may be part of what you were seeing.
You mentioned RFI susceptibility earlier. I can't tell if that is (also) what you were experiencing just based upon what you wrote, but there are some very simple steps to reduce the likelihood of RFI (grounding, shielding, ferrite beads, increased physical separation, etc.).
I have had good results with Sure (Wondom) Amplifiers, which is why I recommended them in the original document. These are Class-D.
Discrete Class-AB is just more robust anyway
I'm sorry but that's simply not true. You can make a crappy, non-robust Class-A, Class-AB, Class-C, Class-(take your pick) amplifier just as you can make a good one. It all depends upon whether it is appropriate for the application, "fit for purpose", as they say in the UK.
Just because you had bad results with some cheap Class-D amps absolutely does not indict all Class-D amplifiers. That's like saying that because you got food poisoning from some incorrectly-prepared chicken that
all chicken is poisonous. It just does not stand up to the evidence.
The same goes for RFI susceptibility. Susceptibility to RF is a very niche need and it's not surprising that the makers of the very cheap Class-D amplifiers, who are under pressure to reduce every fraction of a cent in "extra" cost, did not worry about trying to reduce injection of ~6 MHz into their amplifiers.
Also, to be clear, it's not clear to me that your problems were not the result of things
external to the Class-D amplifiers you tried.
