Measurements
Contents
As befits Alpha Audio, we also measured on the products. It is extremely difficult to measure and visualize differences in analog interlinks. For example, the measurement in terms of frequency response and phase shows nothing. And this while differences in timbre are indeed noticeable.
Where we do see differences is in the ‘impulse behavior’. We let a tone generator send out a square wave at 100 kHz. We set the output impedance to 50 Ohm. The input impedance of the scope is 1 MOhm. Unfortunately, with seperate outputs we did not get the measurements perfectly superimposed with our scope, but you get an idea, we estimate. Here we do see differences.
Square wave test – overlay – with splitter
We put a splitter on output-1 on the tone generator and connected two cables. These go to inputs 1 and 2 of the scope. This allows us to overlay the graphs of the cables exactly on top of each other. This does not work if we use two separate outputs on the tone generator. However, we did that too. Those measurements are shown below. You will see that the differences remain fairly equal. The cables do affect each other slightly through the splitter, so we did want to do two types of measurements.
You can see in the descriptions of the screenshots which cables we overlay. What is striking is how differently graphene measures. The deviation from the rest is large. The graphene with silver is almost the same as the silver cable. And the listening test showed that too: these cables did show similar characteristics. Mundorf and WE have more overshoot than silver. Lamp wire the least if we average everything.
Square wave test – Lamp wire versus the rest
Lamp wire = yellow
Block wave test – 5N silver vs. the rest
5N silver = yellow
It is quite striking how the response differs. Some cables show much more sway than others. And we wonder a bit if this is audible in practice.
Lamp wire and graphene are clearly more damped (graphene does dip a bit more). And in practice, these also sounded quite different from, say, silver. Silver sounded much richer. And compared to WE and Mundorf, the latter two were again more up-front. Those also show a greater swing out. Is there a relationship? Tricky, but it does intrigue us.
Phase and response
Now we actually expected to measure differences here. But the fact is: everything remained exactly the same. Whether we use lamp wire or silver: the phase and frequency response is the same in this measurement. Bit disappointing to be honest.
Cable properties
Except for conductivity, the properties are pretty close. This has to do with the geometry which we have kept the same – except for the standard lamp wire. However, the difference with the mundorf is striking. This mainly shows what wire thickness does in terms of impedance and phase.
We also see more sway in the low frequency domain with certain cables. That may have to do with coatings or other insulations. Or perhaps the mixing of the materials. Think copper / graphene. Copper / Tin. Or silver / gold.
“After we made the six cables, they all played for at least a few days”…
Absolutely not enough. A minimum of 10 days is necessary so that we can confidently talk about the differences. If you had followed that, the test would have been much more valuable.
R, L and C make complex or imaginairy numbers for Z.
But dos not matter if jou measure the amplitude and phase response or an impulse response and compare.
Another example, more advanced.
Like all those who believe that everything is reduced to the domain of time, without taking into account another form of analysis such as the revolution brought about by the Fourier Transform.
A transformation that many use but that very few really understand. With this I want to indicate that many engineers lack additional knowledge and try to understand reality from a limited academic training that prevents it.
https://en.wikipedia.org/wiki/Fourier_transform
– The End –
Fourier transform is nothing more than a mathematical transition to the frequency domain. Makes math easier but adds nothing more as signal in time domain.
My comment was not completely correct.
Fourier transform can lose time information:
https://en.m.wikipedia.org/wiki/Time%E2%80%93frequency_analysis_for_music_signals
Conclusion: time or impulse response is as important as frequency domain analisys.
Spectacular results. Anyone who still doubts that there are big differences is that they are fanatics or are deaf or, probably, have big bottlenecks in their audio systems that prevent them from appreciating the differences. Even so, the difference of the 5N silver cable with respect to the others is so enormous that it must be easily appreciated unless the small speakers of the laptop are used to do the test, I say (surely there will be not a few).
There are differences in the speaker cables too but not as huge as with that pure solid silver cable.
Many engineers just memorized knowledge without really understanding it, especially those who only studied electricity or electronics when more knowledge is necessary to deal well with the audio world (for example, wavesm electromagnetic fields, antennas, signal processing, electroacoustics…).
To understand us, many of them live in the world of mathematics of Real numbers, unable to understand that there are others, such as Imaginary or Complex ones. You know: Z = R + j X(w)
Always is all R, L and C to them, anything more.
—- —- —- —-
Greetings from Tarragona (Spain). If one day it occurs to you to come, give me a call and I will invite you “The Three Musketeers” to a good seafood platter or similar (if you are vegetarian, better, less onerous for me 😛 ).