John Curl's Blowtorch preamplifier part II

[DF96]

Originally Posted by jneutron

Why do you believe the delay is more for low frequency audio?

Because the cable propagation speed is slower for low frequencies, as R dominates over L. In fact, it could be argued that for LF there is no longer wave propagation but diffusion.

Don't forget that the 1/sqrt(LC) figure is a high frequency approximation, valid typically from a few kHz (or 10's of kHz) upwards.

[PMA]

I hope everyone knows R/G for LF and L/C for high freq (rt)

[PMA]

Originally Posted by DF96

Because the cable propagation speed is slower for low frequencies, as R dominates over L. In fact, it could be argued that for LF there is no longer wave propagation but diffusion.

Don't forget that the 1/sqrt(LC) figure is a high frequency approximation, valid typically from a few kHz (or 10's of kHz) upwards.

Your formula is for resonance

[RNMarsh]

Originally Posted by Esperado

About the 'electrolitic cap' poison pointed by John, i believe we are back again in the black magic missunderstanding.
And i'm surprised to see that kind of assertion under the signature of an audio Guru.

An electronic cap present a (little) parasitic serial resistance and a parasitic serial inductance.
And minimal measured distortion, sometimes under the one of some film ones .
In fact a very good capacitance.
Dot!

When used in a network in serial with a > 1000 time inductance and a with a >1000 resistance, crossed by signals limited between 20 and 40Hz what the hell ?
When the speaker in // will naturally add a 12db/acoustic filter above 3Khz, when the previous filter is 18db/oct above 700Hz, what about supposed distortions ?
The only thing we can worry about is their life durability and precision.


What you see here is their internal resistance and inductance, refer in my above words.

I have no issue with your explanation for your situation. And, maybe not for a cap in the place of the circuit you use it (?). The cap wasnt as large value as the one you use.... yours would have much larger esr. The info/data may apply to other conditions than yours.

For example, some people have thier power amps very close to the speakers with lowest Ls and Rs and Cp from cables. Then esr would be a possible factor for them. Others have long runs around a room for surround or inwall cabling to distant speakers. Short length make the cable parasitics issues to be different from long cable runs. The importance of a paramenter is very situation dependant. Lets not assume each comment applies only to ones own condition/situation.... at least for my general comments. Thx-RNM

[DF96]

Originally Posted by PMA

Your formula is for resonance

If L and C are per metre then it gives the propagation speed when the high frequency approximation is valid.

[jneutron]

Originally Posted by DF96

Because the cable propagation speed is slower for low frequencies, as R dominates over L.

Yup, already said that as well. And, stated that it only makes the delay numbers worse than my intial approximation.
I was making sure you weren't assuming the load impedance peak in bass as a result of a reflex cab or something.

Originally Posted by DF96

In fact, it could be argued that for LF there is no longer wave propagation but diffusion.

One could..but then, what would be the "diffusion" velocity?

Originally Posted by DF96

Don't forget that the 1/sqrt(LC) figure is a high frequency approximation, valid typically from a few kHz (or 10's of kHz) upwards.

Already knew and stated that. So you agree that for the lower frequencies, my invokation of t-line is merely a lower bound on the delay, it will be significantly higher.

So then, where's the issue? You can't say there is no delay because the formula misses some entity, then ignore the fact that the entity makes the delay even larger..cake and eat it?

Originally Posted by PMA

Your formula is for resonance

He (DF96) is correct.

It's actually the equation for prop speed.

jn

[Esperado]

Originally Posted by PMA

No poison, please read Richard Marsh's and Walt Jung's paper on measurement of capacitor distortion and DA.

We had made such measurements long before Mr Walt and Jung published their owns.
Please, again, refer to the context in witch i use them.

Look at the jbl network. Here, we had the choice to use a 50µf film. No audible difference. blind tested with several people at various levels.
Note on this that the inductance Zobel is sophisticated enough to compensate flat the inductance impedance of the moving coil up to very High Hf (up to 1MHz). Just for you can feel the level of precision of our work.

[john curl]

Why did you NOT keep the 50uF films in the circuit? Then you would be beyond any criticism.

[Kindhornman]

"One of the biggest problems with frequency-dependent channel-channel phase differences lies in the loudspeakers, especially in the crossovers. This is doubly true for all-pass crossovers like the Linkwitz. Passive component tolerances can cause the crossover point in the two channels to be at slightly different frequencies, resulting in significant inter-channel phase differences in the frequency bands near the crossover points.

This is huge compared to any cable-related frequency-dependent differential phase effects."

Cordell.

Thank you for that one statement Mr. Cordell. I was thinking this the whole time this discussion of speaker cables was going on how much greater the crossover characteristics must overwhelm this argument. I was thinking of the LR 4th order passive crossovers thatI am using in my monitor speakers. The large inductors and multiple caps, all pp type crossed at 2.5Khz must be a factor greater at least than any cable interconnect must produce.

On another side note here that goes with this discussion. First what are we to do with the phenomena of the connection of these wires to the speaker cables when they meet the connectors at the loudspeaker. What I mean is who knows unless you've built the speakers as I have what the manufacturer has used for internal wiring. Is there some kind of reactance when the signal not only gets to the gold plated binding post but also meets the internal wiring and then on to the xo before the loudpseaker devices. If you had a choice of any type of wire construction and insulator material what would be the optimum solution. Do we go with copper wires and foamed Teflon or what? I'm sure that Mr. Kimber would love to sell me silver braided wiring but can't say that is a great idea with the above scenario. And I have not seen any radial lead pp capacitors that can be used in the high values necessary to build a passive crossover at these passive crossover frequencies myself.

[RNMarsh]

Esperado -- pls re-read my reply above yours. Some people use amp for bass and only passive on mid to high crossover - smaller values. Different confige etc. Info intended to be broad.


All can do this to tell them how their amplifier responds to cable type/ construction, length etal -->

Place the power amp within inches of the speaker input terminals and make a short jumper wire between the two. Listen and measure at the speaker terminals. This is essentially the sound without a speaker cable.... the best it will ever sound or measure. Without moving the amplifier... place a long speaker cable of choise between the two. listen and measure, again. Any difference? You are now comparing No cable to a cable... rather than one cable against another.

[you can do same with source to preamp/amp also]

Thx - RNMarsh