CB was mainly AM so you just need a slightly dirty connection to act as a xtal set for it. enough power and nothing else is needed. There was a stage when my phono stage could pick up french radio on a clear night!
RF anecdote again
Anecdote:
Another powered speaker design required the incorporation of a simulated surround processor following an invention by John Norris. I added a control approach by which the amount of the effect could be smoothly varied. To insure an adequate signal to noise ratio I used discrete bipolars and, based on the low source impedances anticipated, ran them at rather high currents to reduce voltage noise. The input stage was a complementary feedback pair but used in the manner of a cathodyne "phase" splitter, as I needed both polarities for the processing. The system worked as anticipated. But a subsidiary benefit that was not anticipated was immunity to RF in the vicinity of the unshielded processor and control board. Although there was a ground plane where possible, it was only a two-sided PWB, and the components were through-hole, garden variety 2SA1015 and 2SC1815.
In this case the noise constraint led to a much higher-than-usual gain bandwidth product, and as well with the high bias, the detected RF in the vicinity wasn't enough to provide significant detection. The requirements on immunity at the time for the category were somewhat relaxed compared to today, and minimal testing was done. But at a trade show, in the demo room, I noticed that I heard nothing when cell phones were being used, even in rather close proximity. All the other multimedia powered speakers were anything but silent under similar conditions. It was a pleasant surprise.
What is of course happening is that the nonlinearity of the base-emitter junction makes it an efficient AM detector, so the blowback through the feedback network of the RF results in envelope detection, thus audio frequency energy, which is then amplified in a conventional fashion. FET inputs tend to be a lot less prone to such detection, one number given suggesting by as much as 40dB or more (an old Analog Devices piece, possibly by Walt Jung if memory serves).
Anecdote:
Another powered speaker design required the incorporation of a simulated surround processor following an invention by John Norris. I added a control approach by which the amount of the effect could be smoothly varied. To insure an adequate signal to noise ratio I used discrete bipolars and, based on the low source impedances anticipated, ran them at rather high currents to reduce voltage noise. The input stage was a complementary feedback pair but used in the manner of a cathodyne "phase" splitter, as I needed both polarities for the processing. The system worked as anticipated. But a subsidiary benefit that was not anticipated was immunity to RF in the vicinity of the unshielded processor and control board. Although there was a ground plane where possible, it was only a two-sided PWB, and the components were through-hole, garden variety 2SA1015 and 2SC1815.
In this case the noise constraint led to a much higher-than-usual gain bandwidth product, and as well with the high bias, the detected RF in the vicinity wasn't enough to provide significant detection. The requirements on immunity at the time for the category were somewhat relaxed compared to today, and minimal testing was done. But at a trade show, in the demo room, I noticed that I heard nothing when cell phones were being used, even in rather close proximity. All the other multimedia powered speakers were anything but silent under similar conditions. It was a pleasant surprise.
Last edited:
I don't get the David Icke thing and find it inflammatory. There is nothing I've mentioned here that you can sort out for yourself with a little test equipment if you were to take the time. So please drop the snide comments.
Simply put if you add enough feedback to a tube amp you can get output impedances that approach that of solid state. Wherever you got your studies though they apparently are incorrect! Its easy to show too- with just regular test gear, nothing fancy. For example you can have two amps, one sounds bright and the other does not, yet on the bench it can be shown that both have the same bandwidth and FR. So why does one sound bright? Distortion- as the ear translates distortion into tonality. In fact the brain has tipping points and will favor distortion over actual frequency response errors.
Actually that is a good article and your assessment of it is incorrect. Perhaps you should read it again, in the light of: 'its not about " trying to come up with a way of making the driver behave in an open baffle"' The article was written by the head engineer of EV and does not focus on open baffle design. His comments still hold true today, for example no speaker made today needs more than 20:1 for a DF and there still are speakers that need less than 1:1 to sound right.
The output impedance is constant from about 2Hz to about 100KHz. So yes. It depends on the amplifier and how its measured as to what the actual value is, but unless feedback is applied they all are measured in ohms.
I get that, but did you know that there are two definitions? I am assuming you are only familiar with one. There appear to be several black box methods. The most common one I see (which is a voltage paradigm technique) is to inject a voltage into the box and see what happens. But that it not the only means. I already pointed out what the two definitions are. Both are correctly supported by math BTW; this only because electronics is based on math. more:
Actually, the paradigm does seem to make a difference (and so inside the voltage paradigm of course it is wrong)...
In the voltage paradigm we all know that an amplifier with a lower output impedance can drive a lower impedance load with more power. The proof is that an amplifier with a higher output impedance will be seen to have less power into lower impedances. Now let us assume that the higher output impedance amplifier also has no feedback. If we add feedback, it will have a lower output impedance.
So at that point with a lower output impedance, it should be able to drive a lower load impedance as well. But it can't. To do that, you have to add more output devices, a bigger power transformer, etc. because to drive lower impedances requires more current and that does not come from thin air. The problem is you come up against is something called Kirchoff's Law.
The issue has to do with whether you are measuring a voltage wherein the amp is really making no power at all (the voltage paradigm way) or if instead you are measuring the amplifier impedance while it is in fact making significant power (the power paradigm way). Here is the disconnect: adding feedback is known to reduce output impedance. But if that were really true it would violate Kirchoff's Law. So apparently adding feedback does not reduce output impedance even though it can result in causing an amplifier to behave as a voltage source. IOW the impedance of the output circuit and output impedance are not the same thing. Such is the nature of definitions.
This is why I like to use the word 'paradigm'. Paradigm refers to a platform of thought. Anything outside the platform is usually considered to be a form of blasphemy. That's how it works. But the laws of physics care nothing for blasphemy- that's a human problem and is not physical.
That depends on what is meant by good damping. Take another look at that article Banat linked in post #60. By asking that question, you are saying you missed something in that article.
The Ayre amplifier is an example of a SS amp with no GNFB and yet behaves as a voltage source. So another answer might be 'do it with care and you can pull it off'.
Actually it does not matter if the amp has RF bandwidth! RFI in audio circuits is often a problem and can cause any amplifier to sound bright. But when you look at amplifiers, how many actually pay attention to this problem?
Output impedance is defined as: Z = U / I. You can inject a current I and measure U or you can apply a voltage U and measure I. Both methods give the same impedance Z. There is no need for a paradigm here.
And the first one is much more common as it does not upset the steady state operating point.
Output impedance is not well defined for nonlinear systems.
Your amp can have very low output impedance due to feedback - but only until clipping.
And it does not violate Kichoff law.
And the first one is much more common as it does not upset the steady state operating point.
Output impedance is not well defined for nonlinear systems.
Your amp can have very low output impedance due to feedback - but only until clipping.
And it does not violate Kichoff law.
Last edited:
Most conventional SS amps have little gain at RF because of the dominant pole. However, in many cases this is not relevant because the RF signal has already been 'detected' in the input stage so travels on as audio.Fast Eddie D said:
Less conventional amps may have more RF gain, so may need more filtering than a conventional amp.
To answer this somwhat provokative question:
Yes, global feedback is a huge benefit for audio.
Nearly every aspect of amplifiers are improved by global feedback.
Even your own hearing use more than one form of global feedback.
Thanks. Yes, it is important to make the distinction between output impedance and output drive capability. Lowering the former does not entail increasing the latter.
It is difficult to know where to start in correcting such profound misunderstanding. You appear to be confusing output impedance with current driving ability. Adding feedback changes the output impedance - the real output impedance, the one you can calculate and measure in a number of different ways and always get the same result. Kirchoff's laws are not violated by feedback.atmasphere said:
I'm not sure what you mean by "the impedance of the output circuit" - maybe a naive summing up of component values without taking such issues as feedback into account? If so, this can of course be different from output impedance because it is not output impedance (except in trivial cases). Or perhaps you mean the output impedance of the output stage taken in isolation - in isolation from the driving stage (or not?) and in isolation from the feedback?
The 'paradigm' you choose to adopt can make no difference whatsoever in calculating or measuring output impedance. If it appears to do so, then you are not dealing with output impedance but something else which you have confused with output impedance. If an output stage is sufficiently nonlinear that output impedance appears to change significantly with signal level or output loading then it is sufficiently nonlinear that output impedance is not defined - except at the limiting case of vanishingly small signal.
Not a snide comment, but you threw the first brick calling me myopic rather than pointing me at some reasoned analysis. When you go to the 'I don't need to prove anything you can work it out for yourself' then that suggests this is a belief rather than a theory. It would be a lot easier for you to actually give some theory behind this than lots of empty words. I'm not looking for a fight, just for you to back up your assertions as they go against the accepted view.
As for distortion vs FR detectability any actual research here? Or your belief? I am happy to be proved wrong, but as I have already said and you have ignored, those who actually research this seem to disagree with you.
As for damping factors they seem to vary between 1.14 at the low end and 3000 at the high end, with a conjecture that, for non-pathological speakers anything over 50 should be inaudible. For Mk1 WATTs and apogee scintillas you need something a bit higher as the DF can drop to unity.
As mentioned above, output impedance and output power driving capability are two different things. I can get near zero output impedance using a DIP packaged opamp, but that doesn't mean it can drive a 1 ohm speaker to any level!
At low enough drive level, btw, it WILL drive such a speaker, it will just clip (or perhaps blow up?) if level is raised. An exception is if the amp has a DC offset, then it might get worked up with too low a DC load (but that's again just a case of its output current not being able to support the level and load impedance).
On transformer output amps, there are usually different output impedance taps, these are usually tailored to getting the maximum undistorted output (current and voltage CAPABILITY, not linear impedance) at each selected impedance. Changes in feedback and winding resistance will also cause the output impedance of the amp to change with each tap, but as far as I know, that's not the reason the taps are provided.
__
As Dick Pierce used to make a point of, damping factor of loudspeakers is less of a deal than it is made out to be. To be the "true" damping, the impedance (and damping formula) has include the voice coil resistance of the driver, which is no different than a series resistor or a large output resistance on an amplifier. True damping factors more than maybe 4 aren't likely to ever happen.
At low enough drive level, btw, it WILL drive such a speaker, it will just clip (or perhaps blow up?) if level is raised. An exception is if the amp has a DC offset, then it might get worked up with too low a DC load (but that's again just a case of its output current not being able to support the level and load impedance).
On transformer output amps, there are usually different output impedance taps, these are usually tailored to getting the maximum undistorted output (current and voltage CAPABILITY, not linear impedance) at each selected impedance. Changes in feedback and winding resistance will also cause the output impedance of the amp to change with each tap, but as far as I know, that's not the reason the taps are provided.
__
As Dick Pierce used to make a point of, damping factor of loudspeakers is less of a deal than it is made out to be. To be the "true" damping, the impedance (and damping formula) has include the voice coil resistance of the driver, which is no different than a series resistor or a large output resistance on an amplifier. True damping factors more than maybe 4 aren't likely to ever happen.
Last edited:
Another controversial article about amps & Neg. feedback , written by even more controversial scientist Mr. Bruce De Palma (RIP),
this article at certain time at least help me a lot to indirectly realize paradox why one crappy 2 x 7W -STK013 Sanyo chip-amp can reveal much more music enjoyment to me than many other mighty SS amps, very mighty in Watt`s & Specs like many Sony`s , Kenwood`s , Yamaha`s , .................
Analog Audio Amplifier Design
this article at certain time at least help me a lot to indirectly realize paradox why one crappy 2 x 7W -STK013 Sanyo chip-amp can reveal much more music enjoyment to me than many other mighty SS amps, very mighty in Watt`s & Specs like many Sony`s , Kenwood`s , Yamaha`s , .................
Analog Audio Amplifier Design
Last edited:
Bruce dePalma, a scientist? Really?
HAHAHAHAHAHAHA.... wait, I have to catch my breath.... HAHAHAHAHAHA
No.
HAHAHAHAHAHAHA.... wait, I have to catch my breath.... HAHAHAHAHAHA
No.
Yes that is correct, I am only familiar with one. Output impedance is the 'opposition' of the output terminal of a device (let's say an amplifier) to a current that an external device wants to impress on it. If the external device can easily impress a current into or out of the output terminal, the output impedance is low. If the external device cannot impress current easily the Zout is high.
So, for example, if you load an amp with a low impedance and the output voltage doesn't buck much, the external device (the load impedance) can easily impress current on the output and you will recognize an amp with low Zout. If you hang a load on an amp and the output voltage collapses, the Zout is consequently high. All very consistent and nicely corroborated by the real world.
I am unaware of any other definition, so pray tell educate me.
Jan
Just for lolz I decided to try and find the highest damping factor tube power amp currently on the market. Best I have managed so far is 9 for the VTL siegfried, which gives +/-0.5dB variation into a simulated speaker load. Certainly should be audible in AB against any half decent SS amp, but less than the variation a normal speaker gives you.
I think that the AR-1 liked a 4 ohm impedance, and I found that that was so. By listening, of course.
Unless the amplifier has a negative output impedance. However, we don't buy loudspeakers designed for use with such amplifiers, usually. And applying such an approach haphazardly is bound to entail oscillatory instabilities.
Uh Kirchoff, if anyone is listening... I'm not talking about non-linear systems so we can leave that out. If Y'all will go back and reread my post you will see that I am really not arguing with you. I am just pointing out how the voltage and power rules differ.
The voltage paradigm has taken over- we all know that. But unlike side valves in an engine, the prior art (power paradigm) didn't die and go away. Its still around and its rules are still in use. But not to my knowledge in any solid state amps.
What if the circuit employs no feedback?
Go back and reread the post... All I am doing is explaining the position. I know very well how output impedance is measured in the voltage paradigm; I've been building amps for decades and they have gotten good reviews and awards in the high end press for much of that time. While working on my EE I serviced consumer and pro audio gear for a living.
I refer you to the bit about paradigms: anything outside the paradigm is instantly regarded as heresy or blasphemy (IOW I hope you understand that I was expecting people to not grok the power paradigm at all, which is why I pointed this out in my previous post). If you want the historic perspective, how it was before there were the voltage rules, now you have it.
Actually Bill you were throwing bricks at me prior to that. Keep it down will you? I can't help it if you can't see some things. That does not change whether they exist or not. You didn't go look at that Fisher on Google did you? Did you see what it said on the damping control or will you take my word for it?
distortion vs FR detectability: Dr. Herbert Melchur but to my knowledge he has not published his findings yet. He has some very interesting numbers though on how the brain reacts to audio reproduction! Apparently there are tipping points wherein the brain transfers music processing from the limbic system to the cerebral cortex, if the brain detects what appear to be violations of human perceptual rules. IOW the brain will automatically stop treating reproduced music as it does real music. So he has objective numbers on the subjective experience. I am really looking forward to his paper.
DF is usually measured from a perspective of 8 ohms so yes, driving the old Appoges would need a higher DF, but if you look at it from the perspective of the speaker rather than 8 ohms, the DF never needs to exceed 20:1 on any speaker. I found that we could effectively drive the Appoge Full Range (1 ohm load, but really a fairly efficient speaker otherwise) with our 100-watt OTLs and a little autoformer we made for 'difficult' loads. Even then we were not hitting 20:1 but the owner of the speakers told me it was the best they had ever sounded- especially in the bass.
IME the simple facts are: 1) you will never get flat frequency response. I'm not saying don't bother, but I am also saying that the tipping points Dr. Melchur has shown do take over- so the ear can actually favor distortion as tonality over actual tonality. We see this in solid state amps 100% of the time. 2) damping factor isn't all its cracked up to be. Critical damping is good- again- read that article Banat linked in post #60.
Actually Jan I like your explanation quite well. All I was pointing out is that in the power paradigm the output impedance is measured while the amp is making power. In the voltage paradigm its done without power. That is why an OpAmp can have a low output impedance and no power at the same time (and just so we are clear, we are not talking open-loop...). FWIW I am not arguing with you. That strikes me as foolhardy.
9? Really?? Try the Wolcott, its a bit higher 🙂