It won't be a power source, either, since the speaker impedance isn't constant.
Speakers measure 92dBSPL 1W, 1m, 4ohms,, chair is over 2m away. Amp needs 1.5Vpk to reach 5W, I was giving it about 6dB below that (driving directly out of a dac, so peak possible output was 1.4 and gain was set around -6). So, peaks at 2m would be around 91dB, average maybe 10dB below that. Not high.
I guess, maybe, is the AE meant to make something seem loud? I do sense a tendency here for some to enjoy the name-calling, "effects box", " low tech aural exciter". It would be as accurate and less inflammatory to call such amps "amplifiers that arent QUITE as distorted as the speakers most people use". Maybe a dsp solution could be found to get similar distortion character from a uCD amp?
Actually it will be remarkably close. A power source will (attempt) to put out constant power into the load just as a voltage source will (attempt) to produce constant voltage.
Of course no zero NF tube amp is capable of true constant power but in the case of the AR1 1/2 db is really not a problem.
Yes. The terms used are a bit different (e.g., "really pops the vocals out of the mix!"), but that's what it comes down to.
I've often believed that, yes, it could be accurately simulated digitally, but it will be difficult to sell since it lacks the mystique.
I think your argument is that such amps WILL have more distortion than the speakers, at least more than high quality speakers.
How so? If the speaker's impedance is varying between 4R and 25R (not unreasonable numbers, and I'm not taking reactance into account), and the amp has a 7R source impedance, the power won't be anything like constant. Here's the impedance of the first multiway speaker I pulled off Stereophile's website:
Tekton Design Enzo XL loudspeaker Measurements | Stereophile.com
What complicates things even further are the effects of the varying load impedance on the output stage and the generally rising source impedance at higher frequencies.
But also more uniform distortion, as speakers selectively distort when stimulated within relatively narrow freq ranges (usually but not always near the bottom of each drivers' range or near regions of radiator breakup). Ears likely don't do that at high volume in the same bands.
Not really my 'argument', still just a supposition.
I though AE was meant too make voices sound 'richer', usual mentioned application was on Linda Ronstadt's voice? Was AE also used on the orchestra?
That circuit is configured as a unity gain buffer.
One advantage of this configuration is that back emf and other nasties are not fed back into the input of the amplifier and amplified. It's unity gain.
Nakamichi employed this principle in a much better way in some of its amplifiers. (No tone controls.) I own one and it is very powerful for its claimed rating. I have other more conventional designs that sound cleaner to me.
I think it's interesting that people generally will think a playback done at a slightly higher level will sound better than one at a little lower level (hence the abx level match requirement). And generation of in-ear distortion product energy (which, with music, will be a. 'floor' of not-originally-there energy, not discrete tones) happens more as the program gets louder. Seems these two bits of knowledge might go together and at least partly explain the perverse-seeming allure of things like SET amps.
Funny. I always thought that the purpose of the feedback loop was to feed back the nasties back into the input of the amplifier - in opposite phase - so that the effect of the nasties at the output is cancelled. It's what we want, no?
Jan
Whatever fancies you, but in my book this is written in capitals: RED HERRING.
The latter is not guaranteed- for example its not really a variable in our amps.
Regarding how so?: A lot depends on the tube amp in question. Nor is 7 ohms anything like a target (that just happens to be the output impedance for which the AR-1 was designed).
What you want to keep in mind is that the power paradigm speaker design rules are in fact different. For example the port for a woofer is likely tuned to a different frequency to take advantage of the higher output impedance (and if done correctly can result in greater bass extension). You see, the reason the word 'paradigm' is used is because its a different way of thinking, and if you use voltage paradigm rules you will not be able to see how the power paradigm works. You'll just think its non-sense. But the 1940s and 1950s in fact happened and how it worked was with power paradigm concepts (which are still around today; ex.: 1 watt/1 meter, SETs, horns etc.).
So to continue, without feedback the output section of the amplifier is going to behave according to Ohm's law- with lower impedances there will be more current but not as much as in a voltage source- when you work it out on paper it works out to a very close approximation of constant power as long as you stay in certain limits (4 ohms to 25 ohms being pretty easy- go below 4 ohms though and you run into problems). At higher impedances the power simply does not decrease the way it does in a solid state amp (this BTW is why tube amps play better bass on ESLs that have higher impedances in the bass, such as the older Quads).
Again, Google 'Fisher 55A' and take a look at the YouTube image that pops up. At 4:06 in the YT video you get a good look at the damping knob and how its labeled. This is an amplifier that was made at a time when the industry was doing its transition from the power paradigm to the voltage paradigm- it can operate in both worlds.
Mike I don't follow you, sorry. FB DOES use all available loop gain, doesn't it? Or do I read you wrong?
Jan
Just because you don't see it does not mean it is not so- it simply means you don't see it. When you operate from the voltage paradigm (which is why the word paradigm is being used here) you will have difficulty with this.You are correct that one amp does not make 'a transition' but the Fisher was by no means the only line of amps at the time that employed that kind of 'damping' control. At the time the industry was struggling with the lack of a standard- on the one hand you had amps with damping controls, some with feedback and some with none. OTOH you had speakers that had level controls on their drivers for the same reason. By the end of the 1950s voltage feedback was taking over although even at the time (if you read Norman Crowhurst) it was understood that it was not the perfect solution either.
BTW if a tube amp has a lot of feedback it will have a low output impedance that will be a fraction of an ohm; that is how it can behave as a voltage source even though its all-tube.
Something you may find interesting is the definition of output impedance, which is different with the voltage rules as opposed to the power rules. With the voltage rules the output impedance is often a measure of the servo feedback of the feedback loop in many cases; IOW how much the amp actively reacts to the load (which it can do also do to a lesser degree without feedback though Ohm's law). In the power paradigm the output impedance is the actual impedance of the output section. These are two very different numbers!
Ralph (Atmasphere)
I believe that Fischer 55A use very similar basic principles showed in this basic schematic ,
where amp operator can manually to adjust 25R potentiometer and to select ratio between two different Global-FB loops , CFB/VFB ratio , regulating in that way relative level of damping factor , or amplifier output Z .
I believe that Fischer 55A use very similar basic principles showed in this basic schematic ,
where amp operator can manually to adjust 25R potentiometer and to select ratio between two different Global-FB loops , CFB/VFB ratio , regulating in that way relative level of damping factor , or amplifier output Z .
Attachments
Last edited:
Yes- it was voltage and current feedback on the same control. Current feedback increases output impedance; voltage feedback decreases it. In the middle the two cancel as if there is no feedback at all.
Thus the control was marked 'constant voltage', 'constant power' (at the noon position) and 'constant current' fully clockwise.
No. The purpose of the feedback is to control the amplifier. You do not want to amplify external artifacts!
If the output impedance is a "perfect" zero at all frequencies, then there is absolutely no advantage. If you're driving a lot of current into a speaker with very a powerful motor - like a subwoofer - then this topology might offer an advantage.
It's not my topology. It's been used in some very popular amplifiers. And I already said that subjectively I prefer some more conventional amplifiers I own, like the amp in my old Luxman receiver. The Nakamichi amplifier can play speakers way louder subjectively, with less than 2x the rated power; but the Luxman sounds clearer to me.
I was merely pointing out what other designers have done in the past. Doug Self licensed the circuit to Nakamichi. Take it up with him.