Hello everyone,
I've finally gotten around to building a ludicrously powerful single- ended class A MOSFET headphone amp. Because I really want to try building and listening to one, and it's a chance to make something audiophiley without breaking the bank. A single-ended class A amplifier capable of driving loudspeakers tends to be quite power- limited and huge and expensive and heat the room pretty well. Headphones are a bit easier. A watt or two of power into a typical
32-600 ohm load is more than anyone would ever need to use, and has a truly-hi fi level of headroom. Have a 30V supply, idle at 500-1000 mA, that's totally huge for this purpose! An amplifier with this much headroom is going to be running small-signal and be quite linear, so all feedback really has to do is provide damping, control the gain, and iron out the frequency response.Big problem with the oversized amp though - transient muting! When this monstrosity switches on and off it will fire pulses out capable of melting phones and the eardrums behind them. Especially the single- ended nature requiring an output coupling cap of decent size - turn it on and the cap is going to start charging up to Vcc/2 with consequent high current going through the delicate headphones threatening the irreplaceable eardrums behind them. So it needs to be MUTED when turning on and off - something needs to keep an eye on the Vcc rail and just slam down on the output until things have stabilized.
Was thinking a circuit sort of like this - this has been SPICEd but not built yet although I have plans to try it very soon:
The idea is that capacitor C1 keeps track of the trend in Vcc. Upon initialization of the supply, C1 charges up through the BE junction of Q1 and resistor R1. The collector current of Q1 activates the relay coil L1 causing the corresponding contacts to short the output. This mutes anything that might make it to the phones, and allows output coupling cap C2 to charge up very quickly. As C1's voltage approaches one junction drop of Vcc, Q1 turns off and releases L1, allowing the contacts to open up.
Upon turning off the supply, as Vcc drops, C1 discharges through diode D2 and the BE junction of Q3. D2 is present to prevent zenering of the BE junction during the initial charge-up of C1. The collector current of Q3 turns on Q2, again energizing the relay coil L1. Q3 and Q2 form a SCR-type circuit, and a PNPN device might well work instead of the transistors here, but unlike SCR operation the main current flows from what would be a 'gate' electrode in a SCR. Again, once Vcc and C1 have drained sufficiently relay coil L1 is de-energized.
Or at least, so I hope from SPICE. What I wonder is, would this circuit work as a practical device in the real world? Or will C1 leakage cause random 'mute' action? Are there other pitfalls I should be aware of?
Advice and alternatives earnestly solicited! Although bearing in mind that audio gear lasts a long, long time, that I've repaired amps older than my parents, I like simple things that use easily-obtained parts. Simple, rugged, reliable, that'd be awesome.