Every schematic of a SS amp that I've paid any attention to has the output transistors coupled to the output post via a low-value--usually 0.22Ohms or 0.47 Ohms--high-Wattage resistor. There's gotta be a better name for it than 'output-transistor output-coupling resistor'.
Often called the emitter resistor, even though for a CFP output it is in the collector circuit - it is still 'logically' in the emitter of the compound transistor. It has two main functions: limiting peak currents, and helping smooth crossover.
Yes, Jeffrey, Emitter Resistor (or Source Resistor if outputs are mosfets).
Sometimes more specifically referred to as the OS Emitter Resistors.
They have another purpose: to facilitate easy and stable adjustment of the output stage bias current. A good compromise is 0.22R; but if there are lots of output devices, I have seen values of up to 1R, which reduces the need for tight output device matching.
As a general rule, the bias current is adjusted by setting 26mV across the OS Emitter Resistor. So for 0.22R, this means around 120mA, for 0.47R, around 55mA. This voltage is chosen because the resulting current gives a Zout of the device which exactly equals the emitter resistor value, and this in turn halves the transconductance of the device with resistor at the crossover point, helping with a major issue in Class AB circuits called gm doubling.
Cheers,
Hugh
Sometimes more specifically referred to as the OS Emitter Resistors.
They have another purpose: to facilitate easy and stable adjustment of the output stage bias current. A good compromise is 0.22R; but if there are lots of output devices, I have seen values of up to 1R, which reduces the need for tight output device matching.
As a general rule, the bias current is adjusted by setting 26mV across the OS Emitter Resistor. So for 0.22R, this means around 120mA, for 0.47R, around 55mA. This voltage is chosen because the resulting current gives a Zout of the device which exactly equals the emitter resistor value, and this in turn halves the transconductance of the device with resistor at the crossover point, helping with a major issue in Class AB circuits called gm doubling.
Cheers,
Hugh
The quiescent current depends on topology. For Darlington it could be 100-200mA; for CFP more like 10-20mA. This is because the Darlington has a very gentle gm slope, but the CFP gain curve shoots up. Douglas Self covers this in his book, and I explained why in an Electronics World article (which unfortunately is not available on the web).
If you call them emitter resistors, most everyone will know what you are referring to, and no one will laugh.
We call them "ballast" resistors here. "Emitter resistor" refers to its place in the circuit, but emitter resistors do different jobs in different circuits.
We call them "ballast" resistors here. "Emitter resistor" refers to its place in the circuit, but emitter resistors do different jobs in different circuits.
- Status
- Not open for further replies.