Audio buffer stages.

[catman]

G'day all, just gauging a few thoughts and opinions on this. Modern gear is generally designed these days with ultra low audio output impedances aimed at improving 'voltage' drive and 'audio quality' generally.

In the simple audio gear I designed I've always felt that a simple output buffer stage like a FET source follower or bipolar transistor emitter follower (preferably) was a worthwhile refinement. Any thoughts or comments on this? Regards, Felix aka catman.

 

[dshoaf]

I see them in a number of devices, Catman. For instance, the McIntosh C2200 tube preamp has an output impedance of <200 ohms. Another example is the Bose 901 Series II equalizer that's rated at <100 ohms output impedance - this about 1973 vintage or so.

On the tube side, the McIntosh MX-110 tuner/preamp I have has a cathode follower stage that buffers the high level mic/phono stage to the tone control stage. It is there in order to drive both the tone control stage _and_ present a low-impedance for the Tape Output jacks as well. I note there has been a rather lengthy thread over at Steve Hoffman's web site about this very model's buffer. BTW, its a really simple cathode follower using 2 triode elements of a 6D10 tube.

As to whether it is a worthwhile refinement, the are camps who believe that :

- Adding yet another active stage - whether it has gain or not is another discussion! - introduces another opportunity for distortion and noise to be introduced.

- Lowering output impedance in the last stage offers a way to improve the quality of the sonic presentation.

Frankly, the engineer in me says, "don't try to fix what ain't broke". That is, use them whenever they're required but not just for fun. Why run up the cost of a device to use an active stage that doesn't have gain? There are other engineering approaches that would not cause you to need a buffer. Matching gain, impedance, noise floor, distortion and making the tradeoffs is what good engineering is all about.

Cheers,

David

 

[catman]

G'day mate, I can see both sides to this. In my present case I think adding a buffer stage to a high quality FET infinite impedance AM detector I've built, although the output impedance is always quite low to start with but limited output 'drive' without the buffer is a handicap, along with benefical buffering of the AM detector stage itself. Regards, Felix aka catman.

 

[catman]

An interesting experiment.

G'day all, inspired by this idea I actually added a simple emitter follower buffer to one of my infinite impedance detectors, and yes it certainly helps. 'Smoother' audio with improved 'definition'.

Interestingly enough, the RF performance of the detector stage seems to have improved too with better 'sensitivity'. I don't have the test equipment to measure these things but my reasonably critical ears certainly detect an improvement. :yes: Regards, Felix (vk4fuq) aka catman.

 

[catman]

G'day mate, very interesting. I'm using a simple BC549C emitter follower with a 2.2 Kohm emitter resistor into a preamp stage set at 100 k though interconnects in the 200 pf range.

Subjectively the quality of the bass extension seems to have improved giving the sound more overall 'weight'. It sounds very agreeable to these ears. :yes: Regards, Feliux aka catman.

 

[catman]

G'day mate, well that's interesting. Perhaps what I'm hearing is lower distortion. :scratch2: There line preamp is a DIY high quality OPA2134 based preamp set at 100 k input impedance. http://sound.westhost.com/project88.htm and this is the emitter follower circuit I'm using (attached). Regards, Felix aka catman.