Bi-wire/bi-amp

[Irvrobinson]

With passive bi-ampng, I agree in that each amplifier works on separate LF or HF portion of the spectrum and also some signal relative to the other part of the spectrum as filters are not brick walls and use filtering slopes.

I understand that my credibility might not be convincing to you, but I just did a search, and AH did an article with a similar description about four years ago:

https://www.audioholics.com/frequent-questions/the-difference-between-biamping-vs-biwiring

Passive Bi-amping
Passive bi-amping utilizes the passive crossovers built into the speakers, with each amplifier channel reproducing a full range signal to drive separate high and low frequency networks. Relative to active bi-amping, the benefits of passive bi-amping are much less pronounced (you might hear the derogatory term of “fool’s bi-amping” applied). Nonetheless, splitting the high and low frequency networks does have some effect, as the amplifiers each now see a different load than one would under normal circumstances when individually driving the entire speaker. Generally speaking, the individual networks are designed such that “out of band” frequencies will have a very high impedance relative to the expected pass band of the driver(s), and consequently will demand significantly less actual power from the amplifier at those out of band frequencies.

Because of this split, there are a couple practical advantages. Potential output ability is increased as there is little power wasted reproducing those aforementioned “out of band” frequencies. As with active bi-amping, the possibility of tweeter burnout due to amplifier overload is also reduced. However, in terms of basic audible differences, there’s not much to write home about, and what improvements do exist could typically be gained by simply utilizing a single more powerful amplifier with similar or better performance metrics than the two smaller amplifiers.

​

 

[highfigh]

With passive bi-amping, I agree in that each amplifier works on separate LF or HF portion of the spectrum and also some signal relative to the other part of the spectrum as passive crossovers are not brick walls and use filtering slopes.

What makes the amplifier channels act in different frequency ranges in passive bi-amping? Is the crossover now ahead of the amplifier? No, it's after, just like it was with a two conductor speaker cable. The amplifier is sending full-range signal to the crossovers.

Or, is this just semantics?

We're talking about:

Source--->AVR channel--->HP crossover input terminals--->Mid/tweeter
and separate AVR channel --->LP crossover input terminals---> woofer(s)

Is this correct?

 

[Verdinut]

Because an amplifier's output current at different frequencies is determined by the input impedance of the crossover at those frequencies. Assuming the tweeter's crossover has a high-pass filter at frequency X hertz of 24db/octave, the implementation of the crossover will cause the impedance the amplifier sees to be much higher at a frequency of X/2 than at X, so the amplifier will output much less current at X/2 than at X. Are you thinking that the amplifier power delivered to both crossovers is identical, and the current representing the lower frequencies are just dissipated as heat in the components of the tweeter's crossover?

In the tweeter's crossover, the series capacitor acts as the high pass filter and I never thought about heat dissipation because we seldom experience much heat dissipation in a home system environment. But what prevents the amplifier from sending the low frequency power to the tweeter apart from the capacitor? Then, where does that low frequency current go and heat dissipation? I would definitely benefit from your explanation. Thanks in advance.

 

[KEW]

What makes the amplifier channels act in different frequency ranges in passive bi-amping? Is the crossover now ahead of the amplifier? No, it's after, just like it was with a two conductor speaker cable. The amplifier is sending full-range signal to the crossovers.

Or, is this just semantics?

We're talking about:

Source--->AVR channel--->HP crossover input terminals--->Mid/tweeter
and separate AVR channel --->LP crossover input terminals---> woofer(s)

Is this correct?

LOL!
This is exactly why I chose Mechanical Engineering over Electrical!

The amplifier is sending full-range signal to the crossovers.

I think this is more properly stated as "The amplifier is making available a full-range signal to the crossovers"!
It doesn't send it unless it has a place to go and if the XO has a filter, some of it doesn't have a place to go!
So it was there, so where did it go? Now my head is starting to hurt and I am, once again, glad I chose Mechanical Engineering!

PENG or whoever, please correct me if I am mistaken on any of this, and if you can tell me what happens to the unused signal in terms dummied-down enough for me to understand it, that would be great!

PS: I think it terms of upstream and downstream, but with circuitry, I believe you have to fully understand what is "downstream" before you can know what is happening upstream!

 

[Swerd]

I think this is more properly stated as "The amplifier is making available a full-range signal to the crossovers"!
It doesn't send it unless it has a place to go and if the XO has a filter, some of it doesn't have a place to go!
So it was there, so where did it go? Now my head is starting to hurt and I am, once again, glad I chose Mechanical Engineering!

PENG or whoever, please correct me if I am mistaken on any of this, and if you can tell me what happens to the unused signal in terms dummied-down enough for me to understand it, that would be great!

I'm not PENG, and all the electronics I learned was in two semesters of physics in college, many years ago.

The available voltage from the amplifier may be high, but because of the XO filters, the current drawn may be high or low to nonexistent. If the current is blocked by a filter, it doesn't matter what the voltage is, no sound occurs.

 

[Swerd]

The OP’s question is whether there is a “frequency effect between low and high and the impact on a single cable feeding both”?

If I understand the question, it refers to the argument that back EMF from the woofer might stomp on the signals meant for the mid-range and tweeter. This idea has it that using two separate speaker cables keeps the woofer’s back EMF from interfering with the mid and high signals.

Is this back EMF from the woofers large enough to stomp on things? If I remember correctly, back EMF (electro-motive force) is a voltage that commonly occurs in electric motors when they start up. Relative motion between the motor's armature and the magnetic field produced by the motor's field coils, can act as a generator. The polarity of this voltage opposes that of the change in applied voltage, hence the name back or counter EMF.

If this is significant in larger electric motors, such as we have in air conditioning compressors, is the counter EMF from a woofer large enough to matter to the amplified signals that would drive the mid-range or tweeter?

I don’t know the answer, but this sounds similar to some other “audiophile” arguments that use a little bit of correct science to make an incorrect conclusion. I don’t know enough electronics to estimate the counter EMF voltage generated by a woofer and see whether it could stomp on higher frequency audio signals. Maybe someone would like to take a crack at that.

I do know that this can be answered by listening tests, those done with appropriate controls that account for possible differences in perception among various listeners, and for listeners whose responses are affected by sighted bias. Avoid those problems by doing the test with enough listeners (30 to 100) and do the test under blind conditions.

I strongly suspect listening tests to evaluate bi-amping have been done before, but with inconclusive results.

 

[Irvrobinson]

In the tweeter's crossover, the series capacitor acts as the high pass filter and I never thought about heat dissipation because we seldom experience much heat dissipation in a home system environment. But what prevents the amplifier from sending the low frequency power to the tweeter apart from the capacitor? Then, where does that low frequency current go and heat dissipation? I would definitely benefit from your explanation. Thanks in advance.

A simple capacitor will act as a 1st order filter, or 6db/octave. Most commercial speakers will use at least 2nd order filters, or more likely 3rd and 4th order filters to better limit the drivers to specific frequency ranges. With commercial speakers for a variety of reasons Linkwitz-Reilly 4th order filters (this is what LR-4 stands for) are preferred, but they are much more complex and expensive in implementation, especially if you want to implement them at low frequencies. (There are multiple inductors, resistors, and capacitors per driver or driver array.) LF sections use low-pass filters and Mid/HF sections may both use high-pass and low-pass filters.

Amplifiers provide current that varies with the load as "seen" by the their output stages, so capacitors acting as frequency dividers change the load as seen by amplifier (assuming a very simple circuit). The capacitor works by having an increasingly high reactance below the circuit's designed high-pass point, and the reactance diminishes above the high-pass point to where frequencies pass unimpeded.

Here's another well-written AH article on the topic of crossover theory:

https://www.audioholics.com/loudspeaker-design/crossover/crossover-cont

[Caveat: I am not an expert on the tradeoffs and success factors in crossover designs. Everything we're discussing here is relatively simple stuff about how amplifier output stages relate to the crossover's electrical load. How you actually design a crossover is a far more complex topic, and not my expertise at all.]

 

[Irvrobinson]

If I understand the question, it refers to the argument that back EMF from the woofer might stomp on the signals meant for the mid-range and tweeter. This idea has it that using two separate speaker cables keeps the woofer’s back EMF from interfering with the mid and high signals.

Is this back EMF from the woofers large enough to stomp on things? If I remember correctly, back EMF (electro-motive force) is a voltage that commonly occurs in electric motors when they start up. Relative motion between the motor's armature and the magnetic field produced by the motor's field coils, can act as a generator. The polarity of this voltage opposes that of the change in applied voltage, hence the name back or counter EMF.

If this is significant in larger electric motors, such as we have in air conditioning compressors, is the counter EMF from a woofer large enough to matter to the amplified signals that would drive the mid-range or tweeter?

I thought most large woofers these days have shorting rings to reduce/eliminate back EMF, so it's not a significant factor. True?

 

[Swerd]

I thought most large woofers these days have shorting rings to reduce/eliminate back EMF, so it's not a significant factor. True?

I thought the purpose of shorting rings on woofers was to limit voice coil excursion, keeping distortion minimal. If shorting rings also prevent back EMF, I wasn't aware of that. To go back to the OP's question, if shorting rings prevent back EMF, does that mean that back EMF is or was a significant problem in audio?

Most better quality woofers, but not all woofers, have shorting rings.

 

[Swerd]

I found a better explanation of shorting rings in woofers. See pp 4-7 in this link:
https://assets.sonicelectronix.com/manuals/nvx/xqw_sub_manual.pdf

Wikipedia had this to say about shorting rings:

The size and type of magnet and details of the magnetic circuit differ, depending on design goals. For instance, the shape of the pole piece affects the magnetic interaction between the voice coil and the magnetic field, and is sometimes used to modify a driver's behavior. A "shorting ring", or Faraday loop, may be included as a thin copper cap fitted over the pole tip or as a heavy ring situated within the magnet-pole cavity. The benefits of this complication is reduced impedance at high frequencies, providing extended treble output, reduced harmonic distortion, and a reduction in the inductance modulation that typically accompanies large voice coil excursions. On the other hand, the copper cap requires a wider voice-coil gap, with increased magnetic reluctance; this reduces available flux, requiring a larger magnet for equivalent performance.​

Neither of these mention back EMF.

 

[Swerd]

And, for those of us who abhor short answers… a 37 page paper on the subject.

The Effect of Faraday Ring (Shorting Ring) Usage on Voice Coil Impedance and Its Benefits

I only read the Introduction and the last section, The benefits of using shorting rings. Here is the introduction:

Some loudspeaker drivers are designed with a ring made from a nonferrous and high conductivity material that is placed in to the motor assembly with an orientation that it is coaxial with the voice coil. These rings are commonly called as Faraday rings, or short-circuit rings, or shorting rings. The name Faraday ring comes from the Faraday law in physics, which explains how the ring works in a speaker motor. Short-circuit ring or shorting ring names refer to the fact that these rings are single turn windings that are short circuited. But not only they are short circuited windings, they also short circuit the changing magnetic flux that passes inside them. The short-circuit and shorting ring name may also be referring to this aspect of them as well, which is actually a consequence of them being electrically short circuited windings. In any case, we will use the term shorting ring to refer to them here on.

In this document the effect of these shorting rings on voice coil impedance will be investigated by using simplified approximated models of the loudspeaker driver motor parts. A few examples with calculations of the voice coil impedance will also be given. Because these examples will be based on highly approximated and simplified models, they will be only for illustrative purposes to convey the idea of how they work, obviously not adequate for any real design work, which requires far more accurate modeling.

In practice, the exact placement, shape and the number of these shorting rings in a motor assembly vary from design to design. Some of them are on the base of the pole piece, some cover the pole piece as a sleeve, some are wide and cover the inner surface of the magnet, some are two rings that sandwich a T shaped pole piece from above and below, etc. A shorting ring which is a coaxial sleeve with its radius equal to inner radius of the voice coil and with a length that covers at least the over all excursion range of the voice coil, has the most effect on the impedance of the voice coil than other ring configurations that are used. Because of this and since this document is on the effect of shorting ring on voice coil impedance, the shorting ring investigated here will be of this kind. Understanding the working of such case of a shorting ring, will also lead to understanding the workings of the other configurations, because the basic physic principles that explain their working are all the same.

 

[PENG]

In the tweeter's crossover, the series capacitor acts as the high pass filter and I never thought about heat dissipation because we seldom experience much heat dissipation in a home system environment. But what prevents the amplifier from sending the low frequency power to the tweeter apart from the capacitor? Then, where does that low frequency current go and heat dissipation? I would definitely benefit from your explanation. Thanks in advance.

I already explained that to you, when I have more time I can try to include a diagram, as a picture = 1000 words.
For the time being, just imagine you have a simple circuit that only involves a single capacitor in series with the coil of the tweeter. Why would you expect low frequency current in such a simple circuit?

Just because there is measurable voltage at the output of an amplifier, it does not mean there is current flow, and just because there is current flow does mean the current has all the frequency components. Let's say for argument sake, if the load impedance is infinite at 100 Hz and below, and 8 to 20 ohms at frequencies above 100 Hz, then the current flow in that wire will only have frequency components above 100 Hz, and you really only need to know Ohm's law.

 

[Sparkus]

Sorry man. Don't take it personally. The weekend after talking with you I got so frustrated I made a bad decision...sent it back a few days later and I am shopping for a sub. After that mistake, I decided I should start at the beginning and went through all wiring, from the circuit feeding the HT to every signal and power cable in the system. So that's where I'm at...basics. Just on principle, I completely separated power and signal, changed some cabling, HDMI and RCA. That's where I hit the speaker cables and this.

I absolutely believe you guys...I think what changed it was a parallel run of wire. I probably needed 12 awg (10 wouldn't hurt) speaker cable. Which really makes me feel dumb...I ran 14 thinking it was plenty big. Distance for the mains is less than 15'. That's all it can be though...cable size. Basics.

 

[Sparkus]

I think your right.
I think what happened is by running 2 cables I doubled the size of my cable. If I use 12 or 10 awg and use one cable, I bet I'll get the same results. I used 14 awg, as an electrician I tend to think current, I'm guessing less resistance might be the answer here.
It was very noticeable, not subtle at all. What else could have improved the sound so much?

 

[PENG]

I get that the amp channel actually does only "see" what it reacts to on that driver and what is carried on each wire is slightly different as a result, but I still don't see that as having a significant effect, as I noticed in my own testing. This was with a bookshelf speaker and large amps though, not a huge reactive load, but still the difference was not really audible.

John, I have actually measured the currents in the two pairs of wires and the difference was not insignificant. The speakers measured were the Energy Veritas 2.3i, with the jumpers removed, the two bass drivers would be together as a group, while the tweeter and the mid convergent source module are the together as the other group.
As for audible effects, none to me for sure.

 

[PENG]

I think your right.
I think what happened is by running 2 cables I doubled the size of my cable. If I use 12 or 10 awg and use one cable, I bet I'll get the same results. I used 14 awg, as an electrician I tend to think current, I'm guessing less resistance might be the answer here.
It was very noticeable, not subtle at all. What else could have improved the sound so much?

There could be a number of things, but it is not due to the interference between the high and low frequencies and for 15 feet length, I doubt 14 AWG wires are to blame either. Regardless, if you are so sure it sounds so clearly better to you now, I would highly recommend you leave it alone.

 

[Swerd]

It was very noticeable, not subtle at all. What else could have improved the sound so much?

To be honest, we don't know. However, I doubt if it was for any reasons the bi-amping fans would suggest.

I can think of one trivial, even silly, story of how using new speaker cables made for a much improved sound. Someone replaced his rather old speaker cables with new monster-like cables, reporting a major improvement in sound quality. When pressed by doubtful readers, he retrieved his old cables from the trash and realized the bare ends of the old wire were badly oxidized and missing more than a few strands. He cut off those ends and stripped away some insulation to reveal shiny copper underneath. He tried that old wire with the newly exposed tips and they produced sound no different than his new purchase.

Sometimes, merely re-tipping old wire is all it takes .

 

[PENG]

What makes the amplifier channels act in different frequency ranges in passive bi-amping? Is the crossover now ahead of the amplifier? No, it's after, just like it was with a two conductor speaker cable. The amplifier is sending full-range signal to the crossovers.

Or, is this just semantics?

We're talking about:

Source--->AVR channel--->HP crossover input terminals--->Mid/tweeter
and separate AVR channel --->LP crossover input terminals---> woofer(s)

Is this correct?

No it is not correct but it is partially correct to say the amplifier is sending full range signal to a high pass or low pass filter but only because that would be true on transient basis when the source signal has to charge up the circuit's capacitance, and also because such crossovers are not brick walls so a high pass filter will allow some L.F. signal through. Once steady state is reached, the high pass circuit will contain predominately the high frequency signal currents. Predominately but not totally (or only) because analog filters for loudspeaker crossovers are not brick walls.

Again, once steady state (doesn't take long at all) is reached, the signal current will be the same whether it is in the wire between the amp and the speaker inputs, or between the crossover outputs and the drivers, that is, the wires for the tweeter will carry mainly H.F. signal currents. The same applies to the wires feeding the bass/mid drivers, i.e. the signal currents frequency spectrums as well as magnitudes will be different between the two pairs of feeding wires. This does not mean there will be audible effects, in fact I believe there is no audible effects, but it is incorrect to say the two amps (in bi-amp) will send the same full range signal to the speakers. It is correct to say the drivers in either case are practically getting the same signal current, bi-amp or not, because that's enforced by the same crossover, just that the HP and LP are separated by the removal of the jumper.

In case another diagram (in addition to those linked by HD and Irv) may help, consider the simplest high pass circuit using just a capacitor. You can view the current waveform in the wire part between the capacitor (filter),or after, it will be the same, and you won't find DC, once steady state is reached. Being a single capacitor in series with the tweeter's voice coil, one can easily see why the current would be the same before or after (physically) the capacitor filter. The same principles apply to the typically much more complicated crossover filters.

Too lazy to draw my own so I just copy/paste from a quick Google result:

http://www.cybermike.net/reference/liec_book/AC/AC_8.html

 

[Sparkus]

I do keep reading the arguments about the various high quality cables, I don't want to go that way. I'm just using zip wire and honestly I think for me right now that's probably fine.
Your post is exactly why I decided to start at the beginning and look at everything from the beginning. It's easy to miss the obvious.

 

[Beave]

Sorry man. Don't take it personally. The weekend after talking with you I got so frustrated I made a bad decision...sent it back a few days later and I am shopping for a sub. After that mistake, I decided I should start at the beginning and went through all wiring, from the circuit feeding the HT to every signal and power cable in the system. So that's where I'm at...basics. Just on principle, I completely separated power and signal, changed some cabling, HDMI and RCA. That's where I hit the speaker cables and this.

I absolutely believe you guys...I think what changed it was a parallel run of wire. I probably needed 12 awg (10 wouldn't hurt) speaker cable. Which really makes me feel dumb...I ran 14 thinking it was plenty big. Distance for the mains is less than 15'. That's all it can be though...cable size. Basics.

For runs under 15', there should be no (audible) difference between 12 gauge, 14 gauge, 16 gauge, and even 18 gauge.