The output zobel (typically roughly a 10 ohm R in series with a .1uF cap across the output terminals) helps make the phase margin of the amp more predictable, by effectively reducing the high frequency reactance of the speaker, so the amp may be less likely to spuriously oscillate. But an additional feature is that it shorts out any Rf picked up by the speaker cables acting as an antenna (which allegedly can happen), the energy of which would be delivered to the negative feedback input of the power amp circuit, if the damping factor of the output stage is weakened enough by the rolling off open loop gain. This can allegedly be a significant issue, especially with tube amps that often have significant output impedances at the higher frequencies (roughly 1 - 4 ohms, especially when negative feedback is low).
Do you think so? At HF you end up simply with the 10 ohms in parallel with the inductor and the 10 ohms in series with the cap. It gives just 6 dB att. I must admit a Z of 20 ohms is quite low compared to the antenna impedance of a speaker cable. But not that low, so keeping out HF? I don't think so. But for unconditional stability a CR + RL is a must.
The detailed answer is at
http://www.diyaudio.com/forums/multi-way/204857-test-how-much-voltage-power-do-your-speakers-need.html
I record stuff sometimes and often want to play my own recordings at near the original levels at home.
Beethoven piano sonatas will easily clip a 1000W/channel amp with 90dB/2.83V/1m speakers. With 200W, I don't usually notice the clipping which is most obvious on good small groups singing unaccompanied choral stuff.
I admit the stuff I listen to is different from modern 'pop' (music??) with 3dB dynamic range
Last edited:
Dynamic range has not much to do with it. For your needs a 50W amp is simply not appropriate anyway. But personally I don't appreciate a full symphonic orchestra or full blow choir in my living room. A 50W amps suffice 99.9% of the time. Note that the momentary dynamic range the ear can cope with is no more than 80dB. At loud average levels it regulates slowly down to protect itself.
Last edited:
Bob Cordell mentioned a recording which was substantially dialed back to allow for the snare drum "thwack" -- "Ghetto of my Mind" -- Ricki Lee Jones. See the BAF2016 video on youtube courtesy of Linearsystems. (About an hour and twenty minutes.)
The problem with overload from such a transient is that the amplifier locks up or rails, very unpleasant.
Bob auditioned his "Supergainclone" at our audio society several years back -- the circuit is in his book, but basically he buffers the input signal inverts so that the LM3886 can operate in inverted mode. He also employs a servo and his "klever klipper". The '3886 in his design is compensated similar to the design we are discussing in this thread.
Oh, it sounded really good on the SEAS Thors.
Using active systems, you also get free watts
The example:
100hz at 1 volt
4khz at 1 volt
Max amplitude 2 volts
Then sqare the amplitudes to get the watts x impedance
If the impdance is the same of the drivers at the frequencies you have saved half the watts...
Just standard maths but not so commonly pointed out.
The example:
100hz at 1 volt
4khz at 1 volt
Max amplitude 2 volts
Then sqare the amplitudes to get the watts x impedance
If the impdance is the same of the drivers at the frequencies you have saved half the watts...
Just standard maths but not so commonly pointed out.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.