Where speaker mets the air is the most important part, and that's what you are doing. I was just suggesting that measuring at the speaker terminals should show the same effect, but with less noise. Clean measurements are good. I'll be interested to know if you see that same thing via the microphone as via a purely electrical measurement.
I'm betting that you will.
You are suggesting I will see a difference from amp input to output with different drivers, or driver to resistive load? Yea, I would agree you will. But that is not at all the test I was doing. I was measuring the amp output vs the mic. I agree a pure electrical measurement would be best, but that is not what I was looking at. It is the AMP I am substuting in the test, not the load.
Sigh....
Yes, I know. I was just suggesting that using an electrical connection to amp/speaker junction will show the same thing the mic does, just cleaner. If different amps make the speaker behave differently, that HAS to show up at the speaker terminals. Where else would it come from?
But never mind, it's not that important. Please ignore the suggestion and proceed. Looking forward to your results.
We are using voltage amplifier with the assumption (faulty) it has the capacity to produce whatever current is required to maintain the voltage. The feedback tries very hard to make it so. So, I would not expect to see any voltage abnormality within the range of a scope on the output terminals. I might expect to see differences in current due to a lot of factors in the form of back EMF from the mechanical spring components to possible resonance or harmonic effects. Alas, no current probe yet. But by looking at the acoustic signal, I see the sum total of the effect. All the years of various tests that subtracted the output from the input, ala-Hafler and others would suggest the pure electrical answer is outside our measurements at this time. I agree totally it would be preferable. I don't know if this effect is related to the defect at all yet.
WEB research on the gender differences in uniform loudness did come up with one study I had not thought about: The quickness of the female ear to respond to stimulus. It seams the cochlea of the female is stiffer than the male allowing faster response times. Mostly only the abstracts that talk about sensitivity and frequency range differences were public. One suggested the variances in the pinna were greater than the differences in the cochlea.
WEB research on the gender differences in uniform loudness did come up with one study I had not thought about: The quickness of the female ear to respond to stimulus. It seams the cochlea of the female is stiffer than the male allowing faster response times. Mostly only the abstracts that talk about sensitivity and frequency range differences were public. One suggested the variances in the pinna were greater than the differences in the cochlea.
If there is no difference in the voltage then there can be no difference in the sound (caused by the amp). Any current problems will necessarily cause a change in voltage, unless the amp has zero output impedance in which case there can be no current problems. Pano is right, unless your setup does not obey the laws of physics.tvrgeek said:
I share Sy's opinion (back on page 1, post #3).
But there is one area of audio amplification where there seems to be very little available research on the correlation between measurements and sound quality. That area is tube guitar amps, which are highly non-linear by design.
Large amounts of distortions of various sorts are actually desirable in this application. But I'm not aware of any scientific body of work that has made any attempt to quantify or qualify the types and amounts of these desirable distortions. It seems a "good sounding" tube amp is still designed mostly by cut-and-try, or by copying a previous design known to sound "good".
It would be nice to see modern measurement and computer analysis methods brought to bear on this. Instead of mumbo-jumbo about magic sounding capacitors and resistors, it would be great to have something like a software analysis package which could import the actual transfer characteristics of tubes, correlate that with what is known of human acoustic preferences, and spit out at least some suggestions for operating points and other variables.
-Flieslikeabeagle
But there is one area of audio amplification where there seems to be very little available research on the correlation between measurements and sound quality. That area is tube guitar amps, which are highly non-linear by design.
Large amounts of distortions of various sorts are actually desirable in this application. But I'm not aware of any scientific body of work that has made any attempt to quantify or qualify the types and amounts of these desirable distortions. It seems a "good sounding" tube amp is still designed mostly by cut-and-try, or by copying a previous design known to sound "good".
It would be nice to see modern measurement and computer analysis methods brought to bear on this. Instead of mumbo-jumbo about magic sounding capacitors and resistors, it would be great to have something like a software analysis package which could import the actual transfer characteristics of tubes, correlate that with what is known of human acoustic preferences, and spit out at least some suggestions for operating points and other variables.
-Flieslikeabeagle
DF96
True if I was driving something that was not a motor with mass and springs and did not have feedback. This is what I understand. I have been wrong before.
Wave,
Ah, ya mean I got to go fix my old Heath generator to gate my HP. Yea, I could do that. Any hints on cheap current probes? Back when I was a lab rat, they cost $5000. I hope times have changed.
210
Human hearing is optimized most likely to keep us from being eaten by something faster than we are, our babies crying, or females when we say something we should not across the room from our wife at a noisy party. We men on the other hand seem perfectly happy to ignore voices, like "take out the trash". Actually, I was glad to see the more current research measuring stem activity rather that just "did you hear that". I suspect how we process what our ears pick up also varies with gender.
I hope my Amazon books get here eventually. I have been looking at lots of amp schematics. I get the basics on most of them, but there are a lot of features I don't get. I am gonna half to try my hand at this. I was comparing some of them to the earlier more simple op-amps. DUH. The obvious thing in the op-amp is not caps. Kind of makes picking to poles in the feedback internally a bit harder I guess. External only.
Anyway, back to looking for substitute mosfets. I got a tad carried away stressing the power supply in my freshly refurbed DH 120. Exicon looks like a reasonable part. Everything else in them are really, really generic: 2N2222's, 2N5550's and 2N5401's. Now THAT is learning to build with commodity parts.
True if I was driving something that was not a motor with mass and springs and did not have feedback. This is what I understand. I have been wrong before.
Wave,
Ah, ya mean I got to go fix my old Heath generator to gate my HP. Yea, I could do that. Any hints on cheap current probes? Back when I was a lab rat, they cost $5000. I hope times have changed.
210
Human hearing is optimized most likely to keep us from being eaten by something faster than we are, our babies crying, or females when we say something we should not across the room from our wife at a noisy party. We men on the other hand seem perfectly happy to ignore voices, like "take out the trash". Actually, I was glad to see the more current research measuring stem activity rather that just "did you hear that". I suspect how we process what our ears pick up also varies with gender.
I hope my Amazon books get here eventually. I have been looking at lots of amp schematics. I get the basics on most of them, but there are a lot of features I don't get. I am gonna half to try my hand at this. I was comparing some of them to the earlier more simple op-amps. DUH. The obvious thing in the op-amp is not caps. Kind of makes picking to poles in the feedback internally a bit harder I guess. External only.
Anyway, back to looking for substitute mosfets. I got a tad carried away stressing the power supply in my freshly refurbed DH 120. Exicon looks like a reasonable part. Everything else in them are really, really generic: 2N2222's, 2N5550's and 2N5401's. Now THAT is learning to build with commodity parts.
There were many researches about spectral dynamics and musical sounds,
I myself did such research when designed analog synth and found that nothing principally new can be designed that is totally different from traditional sounds of voices and instruments. Attack, decay, sustain, should present in any musical sound. Louder sound must have wider spectrum than soft sounds. You can mix and match parameters of known musical instruments and voices, but they all are known.
Yes, guitar amps are made depending on which sound was used by famous guitarists. All of them were already digitally modeled, and are available in different digital forms, including VST plugins.
Speaking about correlations of actual transfer characteristics of amps and their components, the whole thread is about it.
If you drive the same transducer with the same voltage then it will draw the same current and you will get the same sound, unless the transducer has a memory which stretches across the time it takes for you to change the amplifier. Possible, but unlikely, and in that case you are really testing the transducer memory rather than the amplifiers.tvrgeek said:
there are the "back emf" and "back driven by the room sound" memes floating around out there - its hard to convey the engineering understanding, perspective on these to the deliberately innumerate
none will admit there are relative magnitudes, we have to discuss any real and many imagined effects as if they all have equal weight
none will admit there are relative magnitudes, we have to discuss any real and many imagined effects as if they all have equal weight
There indeed have been things done in this area, as there's economic pressure to make transistor guitar amps with the same distortion/sound as tube amps (and of course with the lower cost and lighter weight of solid-state amps). These writings are the only public info I know of offhand, especially "Chapter 3 Transtube:"
Peavey.com
The "Line 6 Pod" is a commercially successful example of tube amp simulation/emulation.
I guess we need to talk more loudly about electrical and mechanical analogies and equivalents - get our memes out there too. Ultimately we are dealing with various forms of potential and kinetic energy, damped forced resonators etc. Show people a circuit and they will probably believe that same input gives same output. Unless there are significant latches etc. the same is true for a mechanical system. 'back emf' is just an inductor or LC resonator equivalent. Even room echo will be the same, unless the room changes too.
Ferrite ring around the conductor you want to measure current in, 10 turns or so of enamelled copper wire wound toroid style around the ferrite ring, and you are set to go. Might be a bit noisy, but good enough for FFT's or scope input to get the gist of a situation you'd like to understand better.
vac
I agree if the very same energy profile is delivered to the motor,you will get the same response. Both the amp and the driver are complex impedances with short term energy storage. The driver has mechanical forces trying to return it to rest. The amp, the servo effect of the feedback trying to make the voltage match the signal. So, current will not be exactly 180 degrees off voltage at all times. If I can get a current probe as has been suggested,then looking at the output voltage and output current may provide the clean measurement you are looking for. Instead of declaring defeat because I don't have the tools I could dream of, only means I need to use the tools I do have and figure out what that means. I am very accustomed to indirect measurements.
" All things we can count don't count, and we can't count everything that does count"
- Status
- Not open for further replies.