|   I know CPR...This one's for you skeptic... | MonstrousMike
Jul 8, 2002 6:15 AM | | I'd like to re-visit a topic that Jon Risch quite nicely evaded and obfuscated recently.
Yes, this is about bi-wiring. The basic premise is that two separate cable runs to the speaker will reduce intermodualtion distortion (IMD) and therefore result in better sound.
IMD is caused by introduction of sum and difference frequencies into a signal which contains many different frequencies (like an audio signal). It is proposed that bi-wiring will separate the audio signal into two segments where low freqency signals are sent to the woofer and high frequency signals are sent to the tweeter. Since the impendences of these circuits are different, they will draw different current loads. So far, so good. Now, for the questions:
1. We all agree that a physical junction in a circuit will divide a current depending on the load of each segment. But how does the spectral content become divided? How can low freqs know to turn left and high freqs to turn right without a filter circuit or crossover device?
2. Even if the frequency content could be split into high and low, wouldn't all frequencies have to be isolated? After all, a bunch of high frequency components would modulate with each other as well.
3. If IMD is a problem in speaker wire, would it not also be a problem in interconnects?
References would be nice as would equations. Opinions and comments would add flavour as well, so please, fire at will. |
| re: I know CPR...This one's for you skeptic... | FLZapped
Jul 8, 2002 6:30 AM | | Well, first you must show that wires contain some non-linear element that creates the necessary non-linear response to generate the IM in the first place. So far, Jon has failed to do so.
-Bruce |
| re: I know CPR...This one's for you skeptic... | skeptic
Jul 8, 2002 7:45 AM | | OK, I'll bite.
The arguement is false. It is correct that a non linear element has to be introduced to create sum and difference frquencies as the result of "intermodulation." It doesn't happen. Here's a simple example that proves it. Look at a cable televsion signal. It carries dozens of video signals simultaneously all on different carrier frequencies being dynamically modulated. If the pheonomena were possible, there would be a lot of "heterodyning" which is spurious noise at many frequencies that would cause every television signal on that cable to interfere with every other one. It doesn't happen.
I could make a much better case myself for bi-wiring. You could theorize that the back emf from the woofer somehow interfere's with the siganl to the midrange and tweeter but if you bi-wire, the low output impedence of the amplifier between the two acts as a shunt to dampen it. Plausible, but personally, I don't buy that one either. |
| re: I know CPR...This one's for you skeptic... | FLZapped
Jul 8, 2002 9:10 AM | | [The arguement is false. It is correct that a non linear element has to be introduced to create sum and difference frquencies as the result of "intermodulation." It doesn't happen.]
Feel free to convince Jon of that then. He isn't listening to me...
-Bruce |
| re: I know CPR...This one's for you skeptic... | mtrycrafts
Jul 8, 2002 5:37 PM | | b So far, Jon has failed to do so.
What? Facts get in his way? ;) |
| re: I know CPR...This one's for you skeptic... | Bobby Blacklight
Jul 8, 2002 8:49 AM | | 3. If IMD is a problem in speaker wire, would it not also be a problem in interconnects?
Well if that's the arguement why not any piece or wire. Why not the internal wiring in the output stages of the amp???
1. We all agree that a physical junction in a circuit will divide a current depending on the load of each segment. But how does the spectral content become divided? How can low freqs know to turn left and high freqs to turn right without a filter circuit or crossover device?
The spectral content is divided by the crossover impeadances. At the amp you have the sum of each leg. The speaker crossover acts like the gate to split the signal for each leg. |
| re: I know CPR...This one's for you skeptic... | MonstrousMike
Jul 8, 2002 9:37 AM | | <<<< The spectral content is divided by the crossover impeadances. At the amp you have the sum of each leg. The speaker crossover acts like the gate to split the signal for each leg. >>>>
I'm not sure I follow you here. The crossover is in the speaker so nothing, other than current, gets "split" until the speaker.
In other words, each segment of a bi-wire contains exactly the same frequency components. It's on the other side of crossovers where there is different spectral content. |
| re: I know CPR...This one's for you skeptic... | Bobby Blacklight
Jul 10, 2002 11:11 AM | | I'm not sure I follow you here. The crossover is in the speaker so nothing, other than current, gets "split" until the speaker.
Yes that's it! The filtering/spliting is done by the crossover not the wire. Think os the crossover as a frequency sensitive splitter. The current flow throgh each leg is different. |
| re: I know CPR...This one's for you skeptic... | mtrycrafts
Jul 8, 2002 5:44 PM | | b The basic premise is that two separate cable runs to the speaker will reduce intermodualtion distortion (IMD) and therefore result in better sound.
But in 1998 he gave an AES conference presentation and this was included :) How is that possible? :)Flawd paper and presentation, of course as his measurements were at multiple points and tried to draw some conclusion, incorrectly. But he won't let facts get in his way though:) |
| re: I know CPR...This one's for you skeptic... | JoeW
Jul 8, 2002 9:37 PM | | >1. We all agree that a physical junction in a circuit will divide a current depending on the load of each segment. But how does the spectral content become divided? How can low freqs know to turn left and high freqs to turn right without a filter circuit or crossover device?
Obviously, they do not. But you keep pointing to load - which is different between woofer and tweeter. In some instances marginal, in others severe. I have yet to see anyone insist bi-wiring is automaticlly better. What I see over and over again is that it is system dependent. When it is benificial, is when the difference of the loads is significant. In this case, cabling per situation is reasonable.
In other words, use the one that sounds best with the woof on the woof, and the one that sounds best with the tweet on the tweet.
I have never seen Jon, or any other enthusiast, recommend a 'bi-wire' cable. If you have read Jon's posts, he's pretty clear about that.
Cables are to audio what salt is to a recipe: To taste |
| re: I know CPR...This one's for you skeptic... | mtrycrafts
Jul 8, 2002 10:05 PM | | b I have yet to see anyone insist bi-wiring is automaticlly better.
Read the yeasayer mail.
b What I see over and over again is that it is system dependent.
Another slither.
b When it is benificial, is when the difference of the loads is significant
More speculation.
b I have never seen Jon, or any other enthusiast, recommend a 'bi-wire' cable.
No? His cable doesn't qualify? Or comments on others? Of course he recommends, read more of his posts.
b Cables are to audio what salt is to a recipe: To taste
Oh, but these are testable, you think? Any evidence for the cables? I bet you would prefer salt one way or the other in a DBT taste test better than you would cables. |
| re: I know CPR...This one's for you skeptic... | MonstrousMike
Jul 9, 2002 5:25 AM | | The premise of the benefit of bi-wiring is that IM distortion is reduced. The only way this is possible is if the spectral content of the signal is divided where the bi-wire split occurs.
I can tell you with a lot of certainty that the only way to "direct" frequencies is to use filtering circuitry not a physical junction. Therefore the two inputs to a bi-wirable speaker will see the exact same frequency range albeit at different current levels.
If the premise of IM distortion is incorrect, then there must some other explanation as to why bi-wiring might be beneficial. This other explanation is what I am seeking. |
| re: I know CPR...This one's for you skeptic... | FLZapped
Jul 9, 2002 5:54 AM | | [ The premise of the benefit of bi-wiring is that IM distortion is reduced. The only way this is possible is if the spectral content of the signal is divided where the bi-wire split occurs.]
But first you have to identify some non-linearity that you are seeking to elinminate. That has yet to be adequately defined and quantified.
-Bruc |
| re: I know CPR...This one's for you skeptic... | MonstrousMike
Jul 9, 2002 8:36 AM | | Hey, I'm even giving them the benefit of the doubt on the premise.
If somehow a PhD shows how frequencies can be split without circuitry, then after I pick myself up off the floor from amazement, we can take your tack. |
| Probably repeating myself, but . . . | sam9
Jul 9, 2002 12:12 PM | | .. I strongly suspect that bi-wiring's principle (only?) benefit is that you get to feel and look cool without the effort and expense of bi-amping.
I don't know whether bi-amping has an audible benefit or not but a least the idea sounds plausible on the surface. And even then, the most plausible theory I've heard is that it is the required active XO rather than simply having two more amps to hook up.
To me bi-wiring doesn't even get past the prima facie silliness test. |
| bi-wiring with two amps | james geib
Jul 9, 2002 2:57 PM | | Does anyone disagree that bi-amping a speaker with two 200w amps is better than running single from one 400w amp?? |
| bi-wiring with two amps | mtrycrafts
Jul 9, 2002 5:01 PM | | You don't need nor can use such power in most tweeters, puff. :)
So, I'd prefer the 1 400 watt amp as that would have more power where it is needed. |
| bi-wiring with two amps | pctower
Jul 10, 2002 12:32 PM | | I pondered this question last time I replaced my amps (although it was a 200 watt single amp vs two 100 watt amps) and came to the same conclusion. |
| bi-wiring with two amps | mtrycrafts
Jul 10, 2002 11:36 PM | | There is an interesting 2/3 part article on how to design a multi amped system, how the amp powers are calculated for each speakers, etc. If you are interested, I will cite it for you :) |
| bi-wiring with two amps | FLZapped
Jul 10, 2002 7:57 AM | | This is a case where the math doesn't add up. The whole purpose of bi-amping is to get rid of the lumped crossover elements in the speakers and drive each with their own amp.
The amps power must be matched to the speakers being driven.
-Bruce |
| bi-wiring with two amps | sam9
Jul 10, 2002 9:09 AM | | I don't know about matched. Some people who do this use a tube amp for the tweeter-mid range and a SS amp for mid-low. I don't want to get into the tube vs SS thing simply because I think tube are an expensive nuisance regardless of any merits they may have. However, if one really likes and wants "the tube sound" but also likes heavy duty bass this appears to be a way to have your cake and eat it too.
Another motive is to be able to a class-A amps for the tweeter where the inefficiency of the class-A doesn't matter so much and class-AB for the rest.
Whether any of these scheme results in better sound is something I can't comment on. I just wanted to point out that one motive for bi-amping is the believe that using amps with differeng charateristics for the high and low rangers yields some kind of benefit.
Personally, I remain quite happy with a single ancient Carver "cube" amp with one channel for the right and one for the left. The nuances of bi-amping and XO schemes are too slight for me to worry about though they are facsinating toncontemplate in some regards. |
| Bi-wiring Q's | Jon
Jul 9, 2002 8:41 PM | | RE #1. I think that this is the big hangup for those not used to thinking in terms of EM signals.
An ElectroMagnetic signal consists of a voltage portion, and a current portion, or if you prefer, an E field, and a B field.
An electromagnetic wave can not exist without both present to some extent, other wise you have either extreme, an electrostatic charge, which isn't a pure voltage, but it comes the closest; and a permanent magnet, which represents a nearly pure magnetic field without an E field. These two ideal extremes are static and unchanging, essentially representing the ideal of a simple one factor presence. All by themselves, they can do no work, transfer no signals.
With bi-wiring, we have a voltage signal that is NEARLY the same in the single wire vs. the bi-wiring, BUT the currents are quite different in the two bi-wiring cables, and either one is different than the single wire cable.
Now some folks have tried to say that the superposition of the two bi-wire signals is the same as the single wire signal.
Unfortunately, this is NOT the case, and it is due to the actual circuit involved, as opposed to the wishful thinking of the naysayers.
The reason is the fact that each cable has resistance, inductance and a capacitive aspect, and that the amplifier tends to look like a (nearly) short circuit to ground, or what is known as a virtual ground, and the loudspeaker drive units, the woofer and tweeter, can both generate signals unrelated to the amplifier drive signal, that can get into the other driver via the crossover network.
In point of fact, having the "node" for the two crossover sections join back at the amplifier, brings the virtual ground of the power amplifier output in closer proximity to that node, while when the crossover is joined together at the speaker, the single cable is what EACH crossover section must get through to "see" the amplifier virtual ground. With single wiring, the crossover sections are connected at the speaker, the connection node is a single wiring cable away from the amplifier virtual ground.
Put another way, the drivers can talk directly to one another, with the only damper on the party a signle wired cable away, while in order for this to occur with a bi-wiring set-up, the driver interaction must furst pass through one of the bi-wired cables, and get PAST the amplifier virtual ground, and then back down the other bi-wire cable.
I have likened this to the difference between a "T" connection (single wired), vs. a "V" connection (bi-wired), where the top of the alphanumeric is the speaker, with the two crossover sections at either side of the top, and the bottom is the amp terminal.
Yes, this aspect WILL vary with the amplifier output impedance, and with the total LCR parameters of the cables, and even with the type of crossover the speaker uses, although this is not as much of a factor as one would tend to believe.
The least likely case where bi-wiring will help, is going to be an amp with a high output Z, and the use of high DCR and high inductance cables. The most likely case is going to be with an amplifier with a very low output Z, use of low DCR and low inductance speaker cables.
For the former, reductions in driver to driver interactions may only be on the order of 3 dB, while with the best case scenario, it might be as much as 30 dB or more. This kind of variation is likely responsible for the variability of the success of bi-wiring with certain speakers and systems.
However, the mere fact that the amount of potential distortion reduction varies, does not make the entire bi-wiring attempt wrong, or completely ineffective in all cases, for all speakers.
RE #2 Yes, even with biwiring, there is going to be some sort of IM or less than perfect signal transfer, but by reducing the bandwidth that the cable handles, we have cut in half the number of frequencies that have to traverse either bi-wire cable compared to the sin |
| Bi-wiring Q's, Part 2 | Jon
Jul 9, 2002 8:42 PM | | Part 2
RE #2 Yes, even with biwiring, there is going to be some sort of IM or less than perfect signal transfer, but by reducing the bandwidth that the cable handles, we have cut in half the number of frequencies that have to traverse either bi-wire cable compared to the single wired cable. This will result in a reduced number of total distortion products when summed acoustically.
RE #3 I have already mentioned this before, in the past, and recently at the AA Cable Asylum. Yes, IC's will also be subject to IM distortion arising, for various reasons. It is just a lot harder to properly measure these than for speaker cables, we are talking about 3 or 4 orders of magnitude advantage to measure it in speaker cables, and so, while it may be hard to measure, it is still possible that it is an audible issue.
Jon Risch |
| Bi-wiring Q's, Part 2 - wire IM | FLZapped
Jul 10, 2002 8:44 AM | | So prove it. -Bruce |
| Bi-wiring Q's, Part 2 | FLZapped
Jul 10, 2002 8:55 AM | | [Yes, even with biwiring, there is going to be some sort of IM or less than perfect signal transfer,]
All signal transfers through a medium are less than perfect, that doesn't make a case for IM unless you can clearly identify a non-linearity.
[we have cut in half the number of frequencies that have to traverse either bi-wire cable compared to the single wired cable.]
The applied voltage is exactly the same on both wires, it's a parallel circuit. The only thing that is different is the current and power density vs frequency due to the action of the crossover. So the entire spectrum is STILL there on both.
[IC's will also be subject to IM distortion arising, for various reasons.]
What are those specifically, how are they measured, how are they quantified? When do they become audible? You have all this data, right? Or is this more of your non-game playing speculation being passed of as fact?
[This will result in a reduced number of total distortion products when summed acoustically.]
[ It is just a lot harder to properly measure these(interconnects - Bruce) than for speaker cables, we are talking about 3 or 4 orders of magnitude advantage to measure it in speaker cables, and so, while it may be hard to measure,. . .]
Evidence?
[. . .it is still possible that it is an audible issue.]
Oh, of course, you don't really know, do you.
-Bruce |
| Bi-wiring Q's, Part 2 | Jon
Jul 10, 2002 9:26 PM | | [ The applied voltage is exactly the same on both wires, it's a parallel circuit. ]
No, it is NOT the same.
If you examine the voltage AT THE SPEAKER (where it matters), the voltage at the end of the single cable is NOT the same as the voltage at the end of the woofer bi-wire or the tweeter bi-wire.
The differences may be 40 dB down, or more, but as I have previously noted, these levels of signal abberations ARE at issue.
Jon Risch |
| Bi-wiring Q's, Part 2 | mtrycrafts
Jul 10, 2002 11:40 PM | | b The differences may be 40 dB down, or more, but as I have previously noted, these levels of signal abberations ARE at issue.
What's 40 dB down? IM distortion? And why maybe, didn't you do a good experiment? Of course you didn't, measuring at multiple points and trying to come to some conclusion, LOL. |
| Bi-wiring Q's, Part 2 | FLZapped
Jul 11, 2002 6:25 AM | | Jon, your twisting again, read it again....
The APPLIED voltage, Jon: no, you don't play games. Maybe you just have a processing problem.
[The differences may be 40 dB down, or more, but as I have previously noted, these levels of signal abberations ARE at issue.]
More randomly selected numbers and STILL no proof.
-Bruce |
| I'm proposing a new moniker for you... | MonstrousMike
Jul 10, 2002 6:21 AM | | Two Part Jon.
Instead of addressing every single point you make and risking becoming Two Part Monstrous Mike, I will address only the following at the risk of becoming Strawman Monstrous Mike.
<<<< I think that this is the big hangup for those not used to thinking in terms of EM signals. >>>>
I have been thinking in terms of EM signals for 23 years. And further, in my occupation I am concerned about the entire EM spectrum, not just audio bands.
<<<< I have likened this to the difference between a "T" connection (single wired), vs. a "V" connection (bi-wired)... >>>>
Actually the single and bi-wired connections are both "Y" connections with different lengths for each circuit segment.
<<<< The reason is the fact that each cable has resistance, inductance and a capacitive aspect, and that the amplifier tends to look like a (nearly) short circuit to ground, or what is known as a virtual ground, and the loudspeaker drive units, the woofer and tweeter, can both generate signals unrelated to the amplifier drive signal, that can get into the other driver via the crossover network.
In point of fact, having the "node" for the two crossover sections join back at the amplifier, brings the virtual ground of the power amplifier output in closer proximity to that node, while when the crossover is joined together at the speaker, the single cable is what EACH crossover section must get through to "see" the amplifier virtual ground. With single wiring, the crossover sections are connected at the speaker, the connection node is a single wiring cable away from the amplifier virtual ground. >>>>
What kind of nonsense is this? You're basically that, like a zipper, as the physical node moves from the speaker towards the amplifier, bi-wiring effects are taking place. That would mean the proportional dimensions of the "Y" connection have some meaning (assuming appropriate wire gauges and reasonable wire lengths).
Horse hockey. |
| Horse hockey? | pctower
Jul 10, 2002 12:38 PM | | Is that what you ice jocks play when you get too old to skate? |
| You are already Monsterous! | Jon
Jul 11, 2002 9:43 AM | | [ That would mean the proportional dimensions of the "Y" connection have some meaning (assuming appropriate wire gauges and reasonable wire lengths).
Horse hockey. ]
Now you have made a claim, yet again. When wil you provide ANY backup for this claim?
I think it goes without saying,k that if an individual does not believe that a single speaker cable per channel will not affect the sound,t hat they will also believe that bi-wiring has no effect. Can't imagine why.
I will show, via schematics and a simple analysis, one aspect of HOW bi-wiring can make a difference, and I will post this to my web site for all to see.
This is only one aspect, and would be a factor in conjunction with the other factors for bi-wiring.
I should be able to put something up in the next day or so, certainly by this weekend.
Jon Risch |
| You are already Monsterous! | mtrycrafts
Jul 11, 2002 11:09 PM | | What you should post is evidence that you can hear the difference, JOn, but that will never happen as you are on record you have nothing to prove to anyone but yourself. Therefore, you have no evidence, just empty claims, meaningless. |
| It has been out up at the web site. | Jon
Jul 14, 2002 8:06 PM | | See:
<a href="/crforum?14@@.ef8d3eb/13">Jon "Cumulative effects" 7/14/02 8:04pm</a>
Jon Risch |
| It has been out up at the web site. | mtrycrafts
Jul 15, 2002 2:58 PM | | Loved those references, Jon, based from 1884 science fiction story, LOL |
| Bi-wiring Q's | FLZapped
Jul 10, 2002 8:42 AM | | [For the former, reductions in driver to driver interactions may only be on the order of 3 dB, while with the best case scenario, it might be as much as 30 dB or more. This kind of variation is likely responsible for the variability of the success of bi-wiring with certain speakers and systems.]
Measurement data showing these number to be valid, please.
What about audibility?
-Bruce |
| |
