You're right, I admit it was me who defined the range of human hearing, or was it a dream............? No - it was your oversight of the audio band limit by using fast step risetime that led to some unnecessarily convoluted hybrid of control theory shoehorned with TL model to explain the simple and obvious.......... The clincher will be when you simulate lack of influence of C/length on latency for the audioband in your model with audioband risetime.
No, and he wasn't hamstrung by attempting to co-operate with your artificial load/source models for the sake of co-operative discussion either. Any real hf influence of cable C depends on the extent of shunt impedance provided by speaker and amp at audio hf. But that really doesn't matter as far as cable latency is concerned because cable latency is totally swamped by effects within the real speaker load as soon as real speakers are involved anyway.jneutron said:
Lack of reality is an artefact of your choice of artificial model, JN. But sticking with it because it's yours, you'll find for audioband risetimes, latency does not depend on C/length when you eventually get round to modelling it. LCR lumped is so easy even to do on the back of an envelope.
What an interesting strawman argument. It was not me who defined ITD...
He provided the information dude..
Again, there's that darn statement by Fred Davis that refutes that..
Man, who can ya trust nowadays...
I'm really wondering who you are arguing with now...First, you argue with yourself by claiming that wires cannot cause 5 uSec delays, as that would be "geological"...then you state that they can as you can do it without a calculator.
Now, you are arguing with Fred Davis, and Otala, and Huttunen.
Who else you going to argue with??
jn
You really have given up using TL arguments, JN........you may not realise but you finally get the LCR thing.........
I often measure loudspeakers on a tower outdoors so that the only thing in the measurement is the loudspeaker. We have mosquitoes at times where I live so spending an afternoon outside feeding them while taking a lot of measurements is undesirable enough to have prompted a solution.
I needed to have a 125 foot speaker cable that was invisible taking loudspeaker measurements including impedance curves of the raw drivers.
I will spare all the measurements and cut to the solution which was two parallel coax cables that were cross wired at the ends. I used 9913 coax which is a pair of about around a 10 or 12 gage conductors, the pair of cables in parallel, half that Rdc.
The cables parallel capacitance appeared to be about 2K Ohms at 20Khz, the series L was a tiny effect at 20Khz and the Rdc was less than the 10 foot lengths of the “woven” hifi speaker cables.
In other words, the measured effect of that cable at 125 feet was less than the alterations caused by short lengths of extension cord and the woven hifi cables and was “invisible” in the measurements I needed.. To be clear this was limited to the audio band, no higher than 24Khz.
Fwiw, at ham fests etc, one can buy RG-213 and variations, which is a flexible lower loss coax which makes nice speaker cables.
I bought a bunch out at the lake co fair grounds some years back.
Best,
Tom
Last edited:
Why wouldn't you mount the amp on the tower, then feed it from a balanced signal? 125 feet of cable is getting pretty extreme!😀
They've always been equivalent dude.
What, nothing about capacitance doesn't affect audio cables and 1 ohm doesn't exist??
Did you finally read Davis??
Nice try...the "look it's halley's comet" schtick didn't work.. try another strawman..
jn
Nice. So, you were driving the tower using a 25 ohm cable?
30 nH per foot and 48 pf per foot 84% lightspeed, so nice dielectric as well. It calculates out as foam, it that right?
I assume your speakers have reasonable impedance curves. Not some crazy 1 to 30 ohm stuff, right? I tend to like horns for that reason, no bandpass stuff.
From what I recall, you also do the multiple pass thing for waveform fidelity tests as well. Using a high z cable would certainly toast you if your impedance variation was across the range.. You clearly had issues with standard zip, that can run 100 to 200 ohms, major inductance, not enough c.
jn
Last edited:
Of course. You don't think you've 'been played' do you, JN?
I just don't get why not use the simplest form of analysis when it makes no difference and complexity stands every chance of misapplication.
How did we make the giant step from latency to all aspects of capacitance in speaker cables ? But yes, many aspects of speaker cable capacitance are altered or rendered irrelevant by source/load impedance shunting.
You did not answer as to what it physically means for |Z| for a load such as 1R to be below the DC resistance of the voice coil........?
My goodness, something that is actually, real life useful in this conversation! I'm truly amazed ...
And for the DIY'ers amongst us, 😀, here's a way of doing it ... Cross-Connected Belden 89259 DIY Hi-Fi Speaker Cables
Those instructions are phenomenal!
An externally hosted image should be here but it was not working when we last tested it.
I have to ask why bother.... Normal speaker wire for a standard living room would suffice....
Of course I don't subscribe to all this, I like my motor drives to be connected directly to the motor coils, so I use an active set up with minimal wire from the drive to the motor and no nasty passive components in the way........
Of course I don't subscribe to all this, I like my motor drives to be connected directly to the motor coils, so I use an active set up with minimal wire from the drive to the motor and no nasty passive components in the way........
Tom
The weenians used to have several hundred feet of cable in their test setup until someone made measurements using a resistor. They found out with that the setup had issues. So they made changes.
The improved measurement system did change the voicing of their products. The old ones worked better in large venues.
Now on one of my projects the EC sub changed the brand of wire used without mentioning it. (WPW to Gepco). As I use zobels this did have an effect. The wire specd was built for me initially and became a catalog item.
Es
The weenians used to have several hundred feet of cable in their test setup until someone made measurements using a resistor. They found out with that the setup had issues. So they made changes.
The improved measurement system did change the voicing of their products. The old ones worked better in large venues.
Now on one of my projects the EC sub changed the brand of wire used without mentioning it. (WPW to Gepco). As I use zobels this did have an effect. The wire specd was built for me initially and became a catalog item.
Es
You didn't ask.
All you said is you can't imagine.
So I pointed out What Fred Davis said, referencing Otala and Huttunen.
So ask them..
jn
Your gyrations are hilarious. I've always know that limiting bw makes them equiv, I've been saying that since before 2011. Re-witing history again dude??
I get it, I get it.. your mad because I got you to do the analysis for free..
For you, the simplest is the only thing you understand. So go for it..
Hey, Davis, Otala, greiner, they all used the simplest analysis they could use back in the day... But then, someone came up with the question" Hey, what about ITD in the 2 to 5 uSec range and non linear loads?..
OOps, time for a re-write.
jn
Last edited:
I'm not sure that's what tom was talking about. JR cuts the cables halfway, and by doing so forces the cables to also have a higher inductance. I recall performing the analysis of the inductance by figuring how much field was in the coax, then how much was in the field between. The number 70 nH per foot sticks in my mind. I did the calcs, and asked Jon to measure the inductance..it was exactly what I calculated..I think that was back in 2004, over on cable asylum.
If Tom truly paralleled but swapped core/braid at the ends, then he in essence parallels the coax. Reduces inductance to 30 nH per foot and cable impedance to 25.
jn
Last edited:
- Status
- Not open for further replies.