back at the other subject ....
Ummm.
Regarding directional couplers or bridges also with high vswr used in precision and high resolution measurements ......
A trick I was shown 45-50 years ago is to add resistive pads. As long as you can detect/measure the level, the loss does not matter. However, the vswr will be reduced as the reflected signal is attenuated as well. In fact ---
To reduce the reflected signal from the input or output of any device is to place an attenuator before or after the device. The attenuator reduces the reflected signal two times the value of the attenuation, while the transmitted signal receives the nominal attenuation value.
In practice, the padding on the source side seems to work a little better.
I have measured vswr to an Absolute level 1.001 at 1 GHz and 1.002 at 2 GHz this way 45 years ago. Thus, proving the spec of the GR slotted line in the standards lab at LLNL. Used the setup to calibrate precision microwave attenuators over a range of freq and verify an atten still met spec.
THx-RNMarsh
test and measurement junky
Ummm.
Regarding directional couplers or bridges also with high vswr used in precision and high resolution measurements ......
A trick I was shown 45-50 years ago is to add resistive pads. As long as you can detect/measure the level, the loss does not matter. However, the vswr will be reduced as the reflected signal is attenuated as well. In fact ---
To reduce the reflected signal from the input or output of any device is to place an attenuator before or after the device. The attenuator reduces the reflected signal two times the value of the attenuation, while the transmitted signal receives the nominal attenuation value.
In practice, the padding on the source side seems to work a little better.
I have measured vswr to an Absolute level 1.001 at 1 GHz and 1.002 at 2 GHz this way 45 years ago. Thus, proving the spec of the GR slotted line in the standards lab at LLNL. Used the setup to calibrate precision microwave attenuators over a range of freq and verify an atten still met spec.
THx-RNMarsh
test and measurement junky
Characteristic Z..... I described the characteristic Z value is where the Z flattens out vs freq. Plotted, it looks like this -->
THx-RNMarsh
THx-RNMarsh
The first image is unreadable.
For the second, you could simply provide the reference. it tells a different story than you are trying to convey here.
Transmission Line Zo
For the second, you could simply provide the reference. it tells a different story than you are trying to convey here.
Transmission Line Zo
My understanding is that scattering, defects, dislocations, impurities and direction contribute to current noise magnitude, spectrum and dynamics. It would seem that it would be odd if all conductors 'sounded' the same. In practice I find that introduction of known 'impurities' alters ASDR behaviours consistently and predictably, and that mixtures of known impurities causes IM of the individual ASDR behaviours, IOW pretty much any sound can be cooked up, ya just need the right base recipe and the right spices. In my setup for example I can run 'clean' sound or introduce the sound of gold or silver or both.....four different sounds on the fly and perfectly distinguishable.
Dan.
Hi Richard, AFAIK this is not quite right. Nominal Impedance is the manufacturer rated impedance of a cable, eg 75R and assumes operation in the TL mode frequency region. Characteristic Impedance is the actual impedance at any particular frequency for a given cable and will be same or greater value than the Nominal Impedance for any given cable.
Dan.
You are the one relying on bias/belief. Why do I say that? You haven't cited any proof to back up your claim.
I've already seen published results of level matched double blind listening tests of different amplifiers. Have you?
One question you haven't answered, which store bought amplifier has damping factor so low to cause audible difference these days?
So I will. Most tube amplifiers use matching impedance to maximize power transfer and thus reduce damping factor.
Then there are folks like Macintosh who have output transformers in some of their gear. But I have to admit I haven't looked recently at any of their stuff.
Then an interesting issue is damping factor vs frequency. As output devices got faster some amplifiers began to have effective damping on the midrange drivers.
Now in practice once damping factor gets above 10-20 it doesn't make much difference. Any effect will be below 1dB and the cabling, internal parts and wiring will also limit changes.
Now decreasing damping factor will certainly change the perceived performance of many loudspeaker systems.
There are probably a number of very good reasons ITD is rarely used in mixing. It only applies during the onset of a sound, that's why this example works so well Stereo Perception, Sound Localization & Auditory Cues
It can give a very accurate indication of the direction of a short, sharp sound (unlike ILD) which could be a threat.
Yes, we are all relying on bias/belief for a reason. Sometimes, belief that we don't want to back up with proof. So, what claims i have made? That amplifiers' DF affect sound perceived?
There's no reason you should doubt my experience, especially basic ones like reading level matched DBT results, reading Cordell, Self, Van Dickason, etc. I think people's experience can be seen from their posts.
Let me tell you that I was absent during the test you cited hehe. So excuse me, i reject the result (the 'conclusion', not the result). The test i proposed to you was different. I will not give you the result of ABX with any software but i will tell you which one of which. This is different. Listener should be able to describe what he hears. Correlating perceived sound with 'numbers' is what you do when designing amps with ears.
Aleph3 was not correct answer? Okay, any amp with one pair of latfet output stage.
Only two. Money and time (is money).
Money:
- You would need one delay by individual instrument that you want to put somewhere between center and side, to set a phase difference for this purpose. Compare with a single potentiometer.
- it will not help to sell one record more.
Time (and efforts): Tuning the localisation with the balance pot in the middle, using the delay. By pass the delay and do the same with the panpot alone. Add the two. Redo all this each time you want to change the localisation of an instrument. Or use a specialized device designed for this effect, that I had never seen for daily work in any pro studio.
Everybody consider levels based localisation works well enough.
No. As indicated by his name, it is a pure product of the listener, based on his head geometry.
Last edited:
I've seen many who did that. They sell expensive cables, DACs and/or amps.
You posted a claim that they sound different.
You should compare class A amp with another class A amp. It's a basic knowledge that harmonic signature is different by design between class A and class AB. Plus, your comparison method was just casual subjective bias uncontrolled listening, right?
Obviously. "No" to what?
If you adjust the different settings here you will see what I mean Interaural Time Differences
Last edited:
I suspect that you are not fully correct but quite close.
If I pan to a specific location, it will be the location that my ears are interpreting as such with no ITD. If I then added the required ITD, I change the parametrics even more. So the image will shift more and will not be as expected. In fact, it will add a change in depth, or distance. It is easy enough to calculate the required ILD and ITD for any point in space and apply it given the computing horsepower nowadays. What has held it up in the past is ITD will mess up a monophonic system, which shouldn't be a problem nowadays. Another problem is the equations require very specific speaker locations. And headphones which are lateralization, do not have the crosstalk of a speaker system. The initial ILD/ITD equations require a septum, crosstalk means adjustments away from the pure septum based equations.
Jn
As you gents are still discussing....
The beauty of my RF impedance only model is not that it calculates the system response for all audio frequencies exactly.. I can only calculate the absolute fastest the system will settle to value given the match ratio and the RF line prop delay.
As I pointed out 4 years ago, by adding in the cable Z at audio frequenciies (which from the recent posts is graphed for all), it is obvious that the mismatch becomes even worse.
All can argue that cables act differently at audio frequencies...yes, we know that. They become worse than the model I provided..I intended my model as an eye opener. And the ramifications to a stereo speaker system are not easy to test using electronics. Jn
No to your statement that it only applies during onset of the sound
Not sure what exactly that is showing but thanks for the links
Hi, jneutron.
As far as I remember well from the long gone away time I was exploring this, it was not my feeling. More a reinforcement of the localisation.
But I would not put my hand to cut that I did not do the opposite way: delay first, pan pot then.
All this in the most empirical way, as appropriate. ;-)
PS: To imitate fully the reality, it would be necessary in addition to that to alter the response curves of the instrument differently on the two stereo channels according to the response curve's directivity of the ears.
I don't fully agree on your "distance" change. In free air, whatever the distances (apart VERY close) , the delay will be near the same for sounds coming from the same angle. Just the distances differences of the two ears, not ?
About depth or distance evaluation, I believe our brain use the response curve in free air, and the amount of reverberation in rooms. Who knows !
Last edited:
On axis, there are no distance cues. Forward or back, both parameters are identical Off axis, say 30 degrees, a point source ILD changes as the source recedes. ITD will also change. Ah, almost forgot. ILD is also hugely dependent on the type of radiator. Point, line, and planar all have different equal time surfaces, so level falls as r squared, r, and none respectively for the three source types. Since listening is never head in vice, the choices are really random and what you like. How to incorporate all that into a mix console, I have no idea. Me, I just turn the knob on the strip. Nobody's in the sweet spot anyway for what I do.
Jn
Ed, it says both 1 trace and 1 to 8 ohms.
I see only one squiggly line.😕
Jn
I understand your point. I was just answering your question. Adding resistor at the output as suggested by DF96 is easy and have been used to compare tube amp versus solid state (You sure have read the report).
I have told you my experience. I build amps with ears. Take a simple example, a constant current source for a differential input stage. Read any electronics books regarding how to make a good ccs and see that everyone is talking about noise, output impedance, stability of current versus temperature, change in voltage, etc. But do they know how it affects the sound? If you look at their amp circuits, you will see that they have no idea.
Long time ago i observed that some good designers like to use special transistor in ccs, cascaded. These transistors have low Vcesat or good early voltage behavior (there is currently related thread by Mark Johnson). Of course, i tried it too, tried to listen to the result. And i concluded that it is not important. You see, my point here is, i not just compare things and say "hey, this is better or worse sounding than that!" but also "hey, i don't hear any difference!". You must be crazy to be proud of your measurement result or your so called hifi with 0.0001% THD if you can not even hear the effect.
But I wasn't quoted here and you've read it. Thanks for confirming my suspicion.
You've mistaken me with someone else. I was talking about objectivity in listening comparison method.
I don't know for you, but, me, when I hear a sound in the nature and try to localize the source's direction, my natural reflex is to turn my head right and left, while sweeping the scene with the eyes.
it is a very unnatural position that the one of the audiophile, closing his eyes, keeping his head motionless.
- Status
- Not open for further replies.