Several time ago I mentioned that in the near future the best SSA would not be the BIGBT HP or any other circuits in the thread.
The reason for that was because I found that the simpler topology of VSSA was better. Not only for performance but curiously stability.
Another reason was because I'm against the BIGBT or composite MOS-BJT for the purest amplifier implementation.
I have chosen this simpler topology without the input cascode (or worse VAS cascode) since the beginning, and believe that it is the "simplest and cannot be simpler" Einstein was talking about
My version is only a little different in the input current source (of course) but has higher power with double output mosfet. I believe that it is just a matter of time until LC or other experts build a HP version that will beat mine.
For the version with input cascode (SSA) I keep the Shaan version, only slightly different. I called it in my library SSA Shaan Mk1. Because it is the winner in one aspect compared to the rest.
I have had no access to my computer (hence the audio files and software) since 2 months ago and forgot where I put the amp (later I'll find).
Basically it is the same circuit as Shaan's (and is using 28V rail). I just removed the 1M resistor from input, and fine-tune the RC values from the input cascode (base) to ground (Shaan uses trimmer here, I change the cap value but don't remember).
I'm not sure about the 2 diodes for bias. Because in one version I have 3 diodes, one is in parallel with a resistor (may be because of higher rail voltage). But I think I don't change it in the Shaan Mk1 because 2 diodes is perfect.
- Thermal tracking/balance. Negative side is the thermal resistance.
- It can not hurt and, may-be, can have some advantages on the dynamic behavior, and it is easier to build a low noise power supply with low currents. Negative side is the complexity.
Last edited:
Don't have time to make calculations to answer your first question. Somebody build-it or simulate it ?
With a separate power supply with more voltage for the first stages, you can drive the power Fets with the same voltage to more output power.
Yes, the combined device have two discrete units on the same die, so they run closer in temperature with faster exchange.
About supplies, or you care inside the amplifier to filter ripples and noises in order to get sufficient PSRR, or you can let your amp simpler, with a low noise stabilized supply. Or you can cumulate the two.
Approximate 4.3 to 4.4 volt from rail.
This is what my simulation shows in SPICE.
With Laterals MOSFET output.
So somewhere between 4 and 5 Volts for a real circuit.