Let's return to the original topic of the thread and let this entertaining but seriously OT diversion go.. Thanks...
I'll jump back in (after that lovely Christmas diversion) to agree with Conrad. That's pretty much my experience exactly. Twin volume controls do work, but they aren't much fun to use. It's harder than you think to get matched levels. A good quality dual gang pot or a stepped attenuators is so much better.
Thanks Pano, Regarding level matching, does a balance controls have the capability to adjust levels in smaller increments than twin controls. ( I am not questioning your assessment regarding level matching with twin controls ) I'm just interested to understand why balance controls are better for level matching. I thought my DH-101 had concentric controls, the faceplate says balance. I removed the cover and found four ganged controls? I remembered that the balance control on my unit was never centered for balanced output.
Generally, the best way to allow the user to vary something is to have a control which varies that thing and makes little difference to other things. Hence to vary volume we have a volume control and to vary balance we have a balance control.
Separate channel volume controls means that to vary either volume or balance you need to adjust two controls and to vary them correctly - either in step (volume) or opposite (balance). This is just bad ergonomics, and quite unnecessary. It might (just) be tolerable if balance controls in themselves damaged the sound but they don't damage the sound if done properly.
Separate channel volume controls means that to vary either volume or balance you need to adjust two controls and to vary them correctly - either in step (volume) or opposite (balance). This is just bad ergonomics, and quite unnecessary. It might (just) be tolerable if balance controls in themselves damaged the sound but they don't damage the sound if done properly.
I believe the general consensus is that potentiometers are not the best choice for signal attenuation ( as noted in A previous post ) and I haven't seen forms of balance controls that select matched resistors like ladder or series type stepped attenuators. So DF, if you want another pot in your signal chain possibly by convention, that is OK. Instrumentation makers test gear that I saw being serviced used a form of wipers that slid past contacts on rotary switches to insure positive contact. the preamp that I listen to most has no balance or selector switch. I haven't used a balance control in twenty years. For all that posted statements about all sound differences can be measured, there is an article in a recent well known industry electronics trade publication called "audiophile heaven". Check out the post from the president of the Jensen transformer.
Test gear is often designed to feed 600 ohms or 50 ohms straight from the attenuator. Audio gear may have to feed 20-50k from the attenuator. Hence contact issues can be 2 or 3 orders of magnitude less of a problem for well-designed audio.yero said:
Most of the sound you listen to has been through pots and switches, probably at least several times.
Jensen would not be trying to sell transformer volume controls, would they? (Wild guess, having not seen the article).
I references his comments to counter many of the previous posts which stated that sonic claims have to be measured to be valid. Cnpope post # 27 and funk1980 post # 35 stating " sonic differences are always measureable" 
I never thought the day would come where EDN would entertain this subject.
I never thought the day would come where EDN would entertain this subject.
Last edited:
So, you'd rater use a common cathode triode gain stage then a cascode. Replace constant current sources with a resister and remove the feedback loops. That would remove about 1/3 of the parts
AN amp with those parts actually adds less of its own sound to the music than the very simple amp with less parts does. Those added arts reduce distortion.
Interesting the HiFi amps temps to be a little complex but the musical instrument amps are the simple designs. Musicians don't need to care about RE-producing the sound exactly as it was recorded and actually want to add the amp's own sound. HiFis want to sound like a "wire with gain" and need more parts and more expensive parts to do it.
"I references his comments to counter many of the previous posts which stated that sonic claims have to be measured to be valid. Cnpope post # 27 and funk1980 post # 35 stating " sonic differences are always measureable" I never thought the day would come where EDN would entertain this subject."
I don't see anything in those comments by Whitlock that in any way counter the assertion (which I would stand by) that sonic differences must be measurable with precision equipment. Again, one must be careful to distinguish two distinctly different questions here:
1) If two amplifiers produce audibly different sounds, then that means that their output voltages V(t) as a function of time MUST be different, and these differences must be measurable. This is just basic electrical engineering and science.
(Note that this is not saying that the first measurement someone thinks of, like say the THD or the IMD, is necessarily going to be the one that reveals the difference between the two signals. But there will be differences of some sort, and those differences will be discoverable by measuring appropriate characteristics of the two voltage signals.)
2) The question of what kind of coloration or distortion, relative to a "perfect" signal, sounds the most pleasing or euphonic to a human being is a totally different matter; this involves psychological issues and lies beyond the hard cold principles of electrical engineering.
It is assertion (1) that I am sticking by. And a corollary of that is that if someone claims to hear a difference between two signals and yet it cannot be confirmed by any precision measurements, then I would say that one should, at the very least, be highly sceptical as to whether their A/B comparison procedure is trustworthy. (Maybe it is influenced by observer expectation, for example.)
Chris
I don't see anything in those comments by Whitlock that in any way counter the assertion (which I would stand by) that sonic differences must be measurable with precision equipment. Again, one must be careful to distinguish two distinctly different questions here:
1) If two amplifiers produce audibly different sounds, then that means that their output voltages V(t) as a function of time MUST be different, and these differences must be measurable. This is just basic electrical engineering and science.
(Note that this is not saying that the first measurement someone thinks of, like say the THD or the IMD, is necessarily going to be the one that reveals the difference between the two signals. But there will be differences of some sort, and those differences will be discoverable by measuring appropriate characteristics of the two voltage signals.)
2) The question of what kind of coloration or distortion, relative to a "perfect" signal, sounds the most pleasing or euphonic to a human being is a totally different matter; this involves psychological issues and lies beyond the hard cold principles of electrical engineering.
It is assertion (1) that I am sticking by. And a corollary of that is that if someone claims to hear a difference between two signals and yet it cannot be confirmed by any precision measurements, then I would say that one should, at the very least, be highly sceptical as to whether their A/B comparison procedure is trustworthy. (Maybe it is influenced by observer expectation, for example.)
Chris
I references his comments to counter many of the previous posts which stated that sonic claims have to be measured to be valid. Cnpope post # 27 and funk1980 post # 35 stating " sonic differences are always measureable"
His comments don't counter anything! In fact, he seems to agree! Even calling our ears a means of measurement, which of course they are. Sure, same THD for two amps can sound different, but we all know there's much more to it than just some distortion figure.
Mr. Whitlock states it exactly how it is. Our ears are spectrum analyzers. Hair cells in our inner ear relay electrical pulses to our brain when triggered. Depending on the frequency, vibrations travel deeper into the inner ear so different hair cells get excited by different frequencies. It's that simple. No magic nor golden ears to best simple science. Else, explain to me how current measurement devices are incapable of matching that.
Mr. Whitlock states "two amplifiers with the same THD can sound entirely different, even to the untrained ear. IM distortion tests are a move in the right direction but barely scratch the surface". I can not be certain but can we deduce that he is implying other additional parameters yet to be identified or adequately defined determine our amplifier preferences? Ex. spectral contamination. then can we also deduce that if someone claims to hear a difference between two amplifiers and yet it cannot be confirmed by measurement conclude that our ears hear distortions yet to be quantified rather than to be skeptical of our perceptions?
Which is why people stopped characterizing amps in technical literature by THD alone many decades ago. The only people who cliam that engineers believe "THD is everything" are generally entirely ignorant of engineering. If I have an amp with 2% THD that is all second harmonic and another that has 2% THD that is all 9th harmonic, they will certainly sound different. There's absolutely no mystery in that. Two amps with the same harmonic structure but significantly different output impedances will usually sound different. There's no mystery in that, either. Two amps with different recovery from overload characteristics will sound different if they clip. There's also no mystery in that.
Audio comic books have damaged many minds. It's a pity.
Mr. Whitlock states "two amplifiers with the same THD can sound entirely different, even to the untrained ear.
Yero, you seem to be taking an extremely narrow view of what is meant by the word "measurement." Of course, if you limit yourself to a single parameter, like THD, then that doesn't tell the whole story. One could easily have circumstances, as SY observes, where two amplifiers with the same THD figure sound different.
When I, and probably the others you are criticising, speak of measurements, I mean to encompass all the possible ways of studying the output voltage in the time domain and the frequency domain. Measurements could include digitising and storing the two signals using a high-precision A/D converter, and then dissecting them bit-by-bit, Fourier analysing them, etc., etc. If the two signals genuinely sound different, then sufficiently meticulous analysis of them will reveal the differences.
A simple case to consider would be if the two output signals to the loudspeaker, call them V1(t) and V2(t), were available simultaneously, in real time. An obvious thing to do for starters would then be to subtract one of them from the other, and look at V1(t)-V2(t). If, to take an extreme example, V1(t)-V2(t) never differed from zero by more than a nanovolt, then one would be pretty safe in saying that the the two signals will sound the same. If someone was claiming they heard a sonic difference in those circumstances, one could be more or less certain that they were deluding themselves.
There are deep and subtle issues concerned with how the ear and the brain interpret sounds. But these should not be confused with the much more straightforward and mundane issue of whether different sounds must be caused by measurably different voltage signals.
Chris
Yero, you seem to be taking an extremely narrow view of what is meant by the word "measurement." Of course, if you limit yourself to a single parameter, like THD, then that doesn't tell the whole story. One could easily have circumstances, as SY observes, where two amplifiers with the same THD figure sound different.
When I, and probably the others you are criticising, speak of measurements, I mean to encompass all the possible ways of studying the output voltage in the time domain and the frequency domain. Measurements could include digitising and storing the two signals using a high-precision A/D converter, and then dissecting them bit-by-bit, Fourier analysing them, etc., etc. If the two signals genuinely sound different, then sufficiently meticulous analysis of them will reveal the differences.
A simple case to consider would be if the two output signals to the loudspeaker, call them V1(t) and V2(t), were available simultaneously, in real time. An obvious thing to do for starters would then be to subtract one of them from the other, and look at V1(t)-V2(t). If, to take an extreme example, V1(t)-V2(t) never differed from zero by more than a nanovolt, then one would be pretty safe in saying that the the two signals will sound the same. If someone was claiming they heard a sonic difference in those circumstances, one could be more or less certain that they were deluding themselves.
There are deep and subtle issues concerned with how the ear and the brain interpret sounds. But these should not be confused with the much more straightforward and mundane issue of whether different sounds must be caused by measurably different voltage signals.
Chris
Last edited:
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.