|
|
coupling afforded by the capacitor C1. Then, the feedback strives to eliminate any discrepancy between the two inputs to the Op Amp by driving the MOSFET's gate in response to the error signal. The MOSFET then conducts more or less current in response. The end result is a regulated output that is rock steady and as accurate as the part tolerances will allow. The 100 pf capacitor that connects the output of the Op Amp to its negative input is there to remove the MOSFET from the feedback loop at extremely high frequencies, thus enhancing stability. (One design trick would be to insert a resistor from this capacitor to the MOSFET's source. This resistor working into the capacitor would serve to de-couple the feedback from the output at high frequencies, which would leave the MOSFET working as a simple source follower, while maintaining full feedback at low frequencies, below 200 Hz let's say. The result might be the best of all possibilities: tight, solid bass and liquid mids and highs. If you try this, let us know your results.)
Disadvantages
|
|
 |
|
|
The disadvantages of this design are that it is quite complex, that it requires a separate winding for the IC supply (if a tube rectifier is not used, the 5 VAC winding is perfectly usable), and that the low voltage parts might fool someone into believing that they are safe to handle when in use: they are not safe to handle.
Observations
One critical construction technique must be observed: Make sure that capacitor C2 is of high quality and that it is connected as closely as possible to the negative supply pin of the Op Amp, as most Op Amps have a much poorer PSRR figure at the negative supply relative to the positive supply pin. C2 helps shunt any high frequency noise from feeding into the Op Amp, which it would then pass it along to its output.
The control element IC was not specified, but just about any high quality, low noise, unity gain stable device will work. This circuit has been built using a OP-27 from PMI. Yes, this device is ancient in solid-state terms, but it is also extremely quiet, unity gain functional, and readily available.
The MOSFET used in this circuit must have a high breakdown voltage and sufficient power handling to survive both the idle and the peak dissipations required of it. Lower gate-to-source capacitance is a very desirable feature in this device.
No output capacitor strapped across the output appears in the schematics. Purposely. This circuit has been built with and without a capacitor; it worked in both cases, but the sound was preferred in the second case.
|
