Townsend Isolda cable

[DonH56]

Its not clear, theres no mention if speakers or HPs whereused.

What John said. Post 671 describes the setup and was in reference to your post; post 682 was a follow up to John. Two speakers configured as a normal stereo pair. There were situated "normally" though we varied positions a little bit but were never real close. Maybe 6' ~ 8' apart with MLP at the apex of an equilateral triangle. It was around 1982~1984 so I may not remember all the details and again the focus was not audiophiles and hi-fi. I remember the numbers, just something I wanted to retain I guess, and remember the basic circuits used as it was a bit of work getting the test system all set up. I was tapped (as a student) because I had relevant experience including working with filters and all-pass circuits.

 

[Rock Rabbit]

Don't take this thread too seriously. Use commonsense, at audio frequency only resistance and current capability is important. Not even at high frequency where power is lower, calculating current with nominal impedance and max power then some security margin and that's all...enjoy the music.
If you feel curiosity, experiment with not equal lengths of cable, coiled cable...any decent amp/ speaker gonna sound the same

 

[Helicopter]

Don't take this thread too seriously. Use commonsense, at audio frequency only resistance and current capability is important. Not even at high frequency where power is lower, calculating current with nominal impedance and max power then some security margin and that's all...enjoy the music.
If you feel curiosity, experiment with not equal lengths of cable, coiled cable...any decent amp/ speaker gonna sound the same

I wasn't taking it seriously when we were talking transmission line theory for a 20 foot wire at 20kHz. Now that we are talking Special Relativity, less so. I guess not taking things seriously is relative too.

 

[Deleted member 23034]

I wasn't taking it seriously when we were talking transmission line theory for a 20 foot wire at 20kHz. Now that we are talking Special Relativity, less so. I guess not taking things seriously is relative too.

We are actually not talking special relativity. cbdb2 just pointed out some writings which appear to be a cut and paste job from the physics texts I learned from back in the 70's. As I said, it's just rehash, and has nothing to do with the discussion at hand. Read his post carefully where he calls it "off topic".
If you do not understand the gist of what I am saying, just ask. I don't model audio frequencies for t-lines, I just use the L and C that comes from a line Z that equals the speaker impedance. Nothing more, nothing less.

jn

 

[Rock Rabbit]

We are actually not talking special relativity. cbdb2 just pointed out some writings which appear to be a cut and paste job from the physics texts I learned from back in the 70's. As I said, it's just rehash, and has nothing to do with the discussion at hand. Read his post carefully where he calls it "off topic".
If you do not understand the gist of what I am saying, just ask. I don't model audio frequencies for t-lines, I just use the L and C that comes from a line Z that equals the speaker impedance. Nothing more, nothing less.

jn

That seems very interesting, how Z line very low value could have any effect on the very large L and C values on the electrical model of a typical speaker? For example
https://projectryu.com/wp/2017/07/23/electrical-model-of-loudspeaker-parameters/
In point 2 (fig. 6) where any C is near 1000 uF and any resonant L in the 10 mH order, and that C values don't even consider the box (or closed cavity for tweeter)...there must be an explanation

 

[Deleted member 23034]

That seems very interesting, how Z line very low value could have any effect on the very large L and C values on the electrical model of a typical speaker? For example
https://projectryu.com/wp/2017/07/23/electrical-model-of-loudspeaker-parameters/
In point 2 (fig. 6) where any C is near 1000 uF and any resonant L in the 10 mH order, and that C values don't even consider the box (or closed cavity for tweeter)...there must be an explanation

Interesting link..thank you
I'm actually not sure what your point was w/r to the values and an explanation. Can you elaborate?
interesting point in the paper....."why is mass an inductance".
below resonance, my HP meter considers it a negative inductance, thinks it's a capacitance.
Think of what is delivered to the terminals. You seem to be saying that if I were to for example, delay all midband content on one channel 10 or 20 microseconds, it doesn't matter because of the speaker model? Since others have shown that to be patently false through human tests, I think that can be discarded.

It is important to go beyond that. If something delays midband on one channel only, humans will hear it. The speaker model (which is a rudimentary one that entirely neglects what I've been speaking about in addition to many things Klipple speaks of) doesn't simply discard the timing of energy delivered to it's terminals.

Just look at the spice model (fig 7) of the fig 6 parameters you cite. 40 ohm peaks at 40 hz and 90 hz, back up to 40 ohms at 8 kHz. 8 ohms in between.
The settling time of that model using a 150 ohm 20 foot length of zip with an EDC of 5 will exceed human ITD thresholds at 8 ohms, scales back tremendously at the 40 ohm peaks. And, that is not considering dynamic modulation of the speaker impedance. Too may believe that graph is reality when there is program material driving the speaker. I have a bridge to sell..
What I speak about in terms of cables and speaker dynamic impedance is significantly beyond that work, the authors did not show that level of detail in their model. I'm sure they could not do so considering the audience they intended to reach. Honestly, I do not know if they even realize the complexity, I may be mistaken.
My recommendation to use a 4 pairs of twisted zip to bring the cable L/C down to the 25 ohm RF impedance centers the settling times halfway up that model, and that minimizes the midband ITD in the event there is significant impedance modulation. When that is done, impedance modulation no longer matters, as the impedance variation cannot produce delays which exceed ITD thresholds. And, its cheap, and it does not entail a significant capacitive penalty. It is a buffer against the speaker magnetic design flaws present in most designs.

I have no issue with them reaching out to me or using my material.. Hojo I believe did back in '04 (IIRC) for skin effect, I am happy he did.

jn



The bill comes due... always.

(just watched Dr. Strange)

jn

 

[audio2design]

ITD is based on the same sound reaching the ears at different times and there does need to be some isolation of those sounds to accurately detect position, i.e. you arent going to accurately pick up a light drum hit buried in a loud guitar solo ( if the timing is encoded at all). That likely means for practical purposes that dynamic effects as they apply to magnetics would be equal for both speakers and hence delays would be equal hence not shifting our perception of location of imaging.

 

[solderdude]

Could this only be a possible issue if substantially different cable lengths are used for R and L speakers? (fuzzier stereo imaging)
This is one Isolda's selling points b.t.w.

 

[Deleted member 23034]

ITD is based on the same sound reaching the ears at different times and there does need to be some isolation of those sounds to accurately detect position, i.e. you arent going to accurately pick up a light drum hit buried in a loud guitar solo ( if the timing is encoded at all). That likely means for practical purposes that dynamic effects as they apply to magnetics would be equal for both speakers and hence delays would be equal hence not shifting our perception of location of imaging.

If you re-read Dons initial post, the initial test regimen used pink noise IIRC which caused masking. the experimental design was changed as a result of that.
How you got from masking to "likely means for practical purposes that dynamic effects as they apply to magnetics would be equal for both speakers" is beyond me.
Trying to listen for any dynamic imaging artifacts by listening to Led Zepplin can only lead to heartbreak..

jn

 

[Deleted member 23034]

Could this only be a possible issue if substantially different cable lengths are used for R and L speakers? (fuzzier stereo imaging)
This is one Isolda's selling points b.t.w.

Only if the entire system is susceptible to what is being discussed. However, that would have to be proven. The discussion is about an entity that may or may not be existing in any particular system.
Did I post the two zip test on this forum?

jn

 

[audio2design]

If you re-read Dons initial post, the initial test regimen used pink noise IIRC which caused masking. the experimental design was changed as a result of that.
How you got from masking to "likely means for practical purposes that dynamic effects as they apply to magnetics would be equal for both speakers" is beyond me.
Trying to listen for any dynamic imaging artifacts by listening to Led Zepplin can only lead to heartbreak..

jn

If you mean post 671, I initially interpreted that as tone bursts. Now reading it again it sounds like continuous tones. You can't use continuous tones for ITD experiments with speakers as it is impossible not to have destructive and constructive waveform interference causing volume effects. You can test out by sweeping the continuous tone and noting changes in the itd threshold however even that has its issues because you are also sweeping through changes in our sensitivity based on frequency. To further complicate matters if you are using speakers you have to then add in frequency effects based on human body shape. All in all doing itd test with any sort of accuracy with speakers is quite difficult.

So that brings us back to transient events (with speakers) where you discuss rise times on the order of ITD delays due to cable LC. My point is ITD is based on the same sound reaching each ear at a different time, and since there can't be masking, the sounds must be similar, which means the rise times you discuss would be a symmetrical for both speakers and hence not an issue.

 

[Deleted member 23034]

My point is ITD is based on the same sound reaching each ear at a different time, and since there can't be masking, the sounds must be similar, which means the rise times you discuss would be a symmetrical for both speakers and hence not an issue.

I've no idea where you are coming form. I've also no idea why you are bringing rise times into this discussion.

You are confusing the step settle time analysis with an audio signal.
The step settling time is only a vehicle to show that the line to load impedance changes the response of the system based on the ratio between them.
It shows that for a speaker with a wide range of impedance, any frequency will have a (pseudo- LR) defined delay dependent on the impedance at that frequency, and that as the speaker impedance heads towards the line RF impedance, the settling time gets shorter. I had hoped to define a clear exp decay function based on line to load, but the best fit I could get was the 6th order polynomial. So the hopes of getting a second order lowpass which has its breakpoints dependent on line to load ratio was not possible.

It is important you make the distinction between the analysis to determine what the line to load ratio does to the system, and how the system can be tested with audio signals to determine if the effect is audible. As, I'm not sure what you are thinking.

Don and crew (undergrad I assume) didn't have the problems you are trying to push into the fray. Only that he learned about masking, then countered it. From then on, he reported 2 to 3 uSec ITD sensitivity using a position comparison between a fixed and a mobile sound from two speakers.
Since the test was successful, you saying it cannot be done for some reason or another is kinda moot.

jn

 

[audio2design]

I didn't bring up rise time you brought up rise time related it to itd thresholds.

Don and you are assuming he was measuring itd. I am stating it is not a given that he was measuring itd. He was introducing Channel delay and performed an experiment where humans were able to detect that channel delay. Don assuming that what detected was intraaural time delay. That is absolutely not a given not with speakers as the source as how the experiment was described. They had the problem of constructive and destructive interference as it would be impossible to avoid using speakers.

 

[Cbdb2]

I posted an audio file in post 661. I hear a big difference going from headphones to speakers, delay is still detectable but its a long way from 2us. Thats equivalent to turning your head 1degree. Or moving the source 5cm with a 3 m listening distance.

 

[audio2design]

If you had a single speaker and could create an experiment where you moved it 1 degree or 5cm, you would be able to detect it. 1 degree is considered the reliable angular resolution of our hearing for detecting angular position with optimal detection even less.

 

[Deleted member 23034]

I didn't bring up rise time you brought up rise time related it to itd thresholds.

Don and you are assuming he was measuring itd. I am stating it is not a given that he was measuring itd. He was introducing Channel delay and performed an experiment where humans were able to detect that channel delay. Don assuming that what detected was intraaural time delay. That is absolutely not a given not with speakers as the source as how the experiment was described. They had the problem of constructive and destructive interference as it would be impossible to avoid using speakers.

Actually, he repeated what was done previously.

If you choose to make up your own descriptors, have at it. Or say that avenues of research are not what they seem because you say so.

You have so far, not understood what was presented, mixing RF descriptors and audio descriptors I have presented. I believe it has to do more with the venue of discussion than anything else. Yet, I have been very clear in what I have done and said, which is a conundrum..

If you are willing to understand, I am at your disposal. If you choose to continue along unfruitful lines, that is your choice. I cannot make you understand, that is up to you.

I have to admit. It is the misrepresentation and misunderstanding you provide that soured my participation in diyaudio. I had hoped this forum was different. Perhaps I have made a mistake.

jn

 

[audio2design]

Not "because" I say so. But because it is true, and those skilled in this area know this to be a fact.

You can take your attitude somewhere else. I respect you are an expert at magnetics and obviously know more in his area than I do. However, when it comes to acoustics, and these experiments, obviously you are swinging outside your comfort zone. After you have read all these linked papers, maybe we can have a discussion: https://asa.scitation.org/doi/10.1121/1.4875714


P.s. I have understood what was presented, pointing out quite clearly the flaws in your L/C/Zo graph. However, in the 100+ messages, I missed that you were just using Zo to extract LC as the frequency component cancels out at RF frequencies. Sue me. I was not mixing up RF and Audio descriptors. Perhaps you need to be more concise and clear in your communication.


In terms of many of the experiments w.r.t. ITD as well as experiments to find the limits of non intra aural timing resolution, there are invariably assumptions made in the experiments that are not adequately justified, at least to more thorough researchers, though often the authors believe they are. We aren't solving world hunger here, doing work with high economic impact, or pushing the limits of human knowledge, our field does not have the same level of due diligence and scrutiny that exist in many other scientific fields.

 

[egellings]

Fancy cables and washing & waxing your car. The stereo seems to sound better, and the car just seems to feel like it's running better.

 

[Arno Fennix]

I just found this topic and with a big Smile I enjoyed the demo video by Max. Reminds me of his demo on vibration/damping during one of the audio shows. And the smile is for the down to earth approach, measure, demonstrate and listen avoiding to stick in a world of only theory and opinions.
My setup seems to be ultra sensitive to changes and small steps are audible. For Max' cables I would be sooooooo eager to give these a try, but it is not the easiest or cheapest trial with needing 1 cable to the filterbox and 3 cables going out to the RAAL tweeter, the TD2001 horn and the FE208 BLH...so buying it just for a trial is no option for me unfortunately. I do expect positive effects if I would try them on my system since the demo and cable theory/knowledge makes perfect sense to me (and I have been digesting al lot of cable modelling calculations myself (E-Eng background)..So, proof of the pudding...listening would replace the talking ...;-)

 

[tw99]

I just found this topic and with a big Smile I enjoyed the demo video by Max. Reminds me of his demo on vibration/damping during one of the audio shows. And the smile is for the down to earth approach, measure, demonstrate and listen avoiding to stick in a world of only theory and opinions.
My setup seems to be ultra sensitive to changes and small steps are audible. For Max' cables I would be sooooooo eager to give these a try, but it is not the easiest or cheapest trial with needing 1 cable to the filterbox and 3 cables going out to the RAAL tweeter, the TD2001 horn and the FE208 BLH...so buying it just for a trial is no option for me unfortunately. I do expect positive effects if I would try them on my system since the demo and cable theory/knowledge makes perfect sense to me (and I have been digesting al lot of cable modelling calculations myself (E-Eng background)..So, proof of the pudding...listening would replace the talking ...;-)

You obviously read this topic quite selectively then.