Bi-wiring

[Titanium Dome]

Okay, here is one of JBL's more comprehensive statements on bi-wiring. I'm only focusing on bi-wiring, so the bi-amplification portion is not included. Bi-amping is far more complex and not easily attainable (or practical) for most members here, and I'm including myself.

So, if you want to discuss bi-amping, start another thread please.

From the L7 owner's supplement:

[Titanium Dome]

More from the supplement:

[Titanium Dome]

And more from the supplement:

[Titanium Dome]

So there are four possibilities for bi-wiring according to this information:

One amp (single amp) bi-wire
Two amp bi-wire
Three amp bi-wire
Four amp bi-wire

Thus, JBL presents bi-wiring as a very flexible and progressive activity, suggesting that even the owner of a single amp can make appreciable gains by running two wires per speaker from the single speaker terminals on an amp to the separated terminals on the speaker.

JBL's comments on the effects of HF/LF separation resulting from the running of two wires even from a single source reflect this representation from John Risch's site:

http://www.geocities.com/jonrisch/page7.htm

We always need to bear in mind that the L7 network is not a simple or typical one, and that JBL's comments, while general on the benefits of bi-wiring, in this context are specific to the L7's network and its particular adaptations for this usage.

[Titanium Dome]

For your reference, the L7 network.

[Titanium Dome]

Here is a physical view of the networks.

Top: HF
Middle: LF (side view)
Bottom: LF (top view)

[Titanium Dome]

I've seen lots of flaming discussion on the 'Net, and none are worse than bi-wiring. The acrimony with which naysayers attack those who endorse it is epic. Some of the more nasty, insulting, and immature arguing I've ever read is between those who oppose the idea.

I don't know why this is such a big deal to some, but it is.

IME, sometimes it works well and sometimes it doesn't. It seems to work remarkably well with my L7s and S2600s. It seems to be built in to a stacked PT800/PS1400 Performance Series topology. OTOH, it makes little difference on an XPL200, certainly not of the magnitude of improvement one gets with JBL's optional external crossovers.

So, have you bi-wired anything successfully or unsuccessfully? Do you think it was the wire, the amp(s), and/or the internal crossovers that made it work or not work?

[Fred Sanford]

There are parts of this that make no sense to me as they're illustrated in these scans- if there's no active x-over dividing frequencies before the power amps, how would dual-wiring or even dual-amps reduce the intermodulation issues they refer to here?


"By using separate conductors for high and low frequencies (drawing L-C), unwanted treble modulation is avoided. Bass flows through one, treble through another."

Each cable/conductor would be carrying identical information (one-amp), or each amplifier would be amplifying the entire frequency range (all other scenarios). This one paragraph makes me question the validity of the entire article, but I'll consider the source and figure I'm reading it wrong...

je

[Zilch]

The HF goes through a separate set of wires, and is not subject to the varying voltage drop across the wire resistance (and other allegedly audible effects) in those driving the higher-current LF.

The frequencies are split by the separate passive crossovers (HF vs. LF) "drawing" the signal selectively through the two sets of wires....

[MJC]

It seems to be built in to a stacked PT800/PS1400 Performance Series topology.
So, have you bi-wired anything successfully or unsuccessfully? Do you think it was the wire, the amp(s), and/or the internal crossovers that made it work or not work?
I don't know if sub woofers really count when you talk about bi-wire, being subs have their own amp anyway. Either with a plate amp, as the PS1400, or as I have the SUB1500s powered by a Crown K2, but the signal is coming from the sub out on the receiver, not the speaker outs.

But if the subs and mains are stacked, as can be done in the PS and as I have with the PS/SUB1500, it is kind of like being bi-wire, I guess.
I kinda think of my stacks as full range as all frequenies radiate from the same point, though.
Too bad JBL didn't setup the PT800s like the L7 or L880/890, where the PT800 could be bi-wired.

[speakerdave]

How nice. Thank you, T-dome.

[Titanium Dome]

I don't know if sub woofers really count when you talk about bi-wire, being subs have their own amp anyway. Either with a plate amp, as the PS1400, or as I have the SUB1500s powered by a Crown K2, but the signal is coming from the sub out on the receiver, not the speaker outs.

But if the subs and mains are stacked, as can be done in the PS and as I have with the PS/SUB1500, it is kind of like being bi-wire, I guess.
I kinda think of my stacks as full range as all frequenies radiate from the same point, though.
Too bad JBL didn't setup the PT800s like the L7 or L880/890, where the PT800 could be bi-wired.

When the PS is stacked, I don't consider the LE14H-3 as a sub, rather a woofer. The crossover is up to 130 Hz, rather than the 80 Hz (THX Standard) or 100 Hz I might want for a sub. However, the feed is a single wire, not a bi-wire, so... :dont-know

[just4kinks]

From the other thread...



If the nominal impedence of a speaker system is 8 ohms and you then separate the woofer from the tweeter/mid, by bi-wiring, do you still have an 8 ohm impedence, or would it be more like 4 ohm?
Two 8 ohm loads, in parallel, become 4 ohms to an amp.

Don't forget that neither component is a perfect resistor with flat 8 ohm resistance across the entire frequency range. With the crossovers in place, the woofer has very high impedance at tweeter frequencies, and vice-versa. It will vary from system to system, but the overall impedance will be roughly 8 ohms through most of the frequency range.

[Fred Sanford]

The HF goes through a separate set of wires, and is not subject to the varying voltage drop across the wire resistance (and other allegedly audible effects) in those driving the higher-current LF.

The frequencies are split by the separate passive crossovers (HF vs. LF) "drawing" the signal selectively through the two sets of wires....

That's the part I'm just not "picturing", and I'll assume it's either a gap in my learning or a matter of me misunderstanding semantics. Passive crossovers, at the speaker side of the cable, "drawing" only select frequencies through the amp-to-cabinet cable...I'm just not getting it. I'd see that the reaction of the loads would differ between x-over/LF and x-over/HF in multi-amp scenarios, maybe it's the single amp having the wires common at the output posts that makes it a tough sell for me...but, maybe, with more time trying to picture it, it's looking like there's a difference. Hmmmm...I'll be quiet and watch the proceedings for a while.

je :blink: :dont-know :help: :montyp:

[JBL 4645]

Science be damned, use yer ears, man! THATS the tool! :D


What’s a matter Heather, you seem to disapprove:D. I think that’s a good start please continue Dome, just make it a little more exciting.:bouncy:

[hjames]

Originally Posted by hjames http://audioheritage.org/vbulletin/images/buttons/viewpost.gif (http://audioheritage.org/vbulletin/showthread.php?p=188939#post188939)

Science be damned, use yer ears, man! THATS the tool! :D

What’s a matter Heather, you seem to disapprove:D.
I think that’s a good start please continue Dome, just make it a little more exciting.:bouncy:

Disapprove???
Ashley, can you not read the words I type on a page?

it was SARCASM, I even labelled it so to avoid misunderstanding!
It wasn't meant for a fight, its meant for a smile, ya dingdong!

Nevermind - I've deleted my post just for you.
Go start a fight somewhere else

[Ducatista47]

That's the part I'm just not "picturing", and I'll assume it's either a gap in my learning or a matter of me misunderstanding semantics. Passive crossovers, at the speaker side of the cable, "drawing" only select frequencies through the amp-to-cabinet cable...I'm just not getting it...



Just an ignorant stab here, but I think it is a passband thing. If an electronic or electrical device does not permit a signal to pass through it, and it is part of a series circuit, that signal will not flow and therefore will not be drawn (from the current source - the amp). In the case of a crossover, that signal would be whatever band(s) - frequencies - are not allowed to pass to the speaker component down the line.

I don't think the blocked signal is converted to heat in a crossover, at least not most of it, but I could be wrong. In most electronic circuits, resistors are there to create a needed condition, usually a particular voltage value, not to convert signal to heat. You know, Ohm's Law.

Clark, Probably Ignorant in Peoria

[rek50]

Thanks for posting the info, Dome.
The one amp/bi-wire idea is almost like saying, the frequencies are different, because they're separated together. BUT, one amp = ALL the frequencies, through both sets of wires, independent, of where or what they're connected to. I think it's in the ohms load.

Years ago, I tried the one amp/bi-wire set up on a pair of McIntosh SL-4 speakers. The SL-4 does have the separate inputs for LF/HF, as well as the strapping bars. I used Monster cables, 100WPC Denon receiver. I connected the cables to A and B speaker outputs, ran A to the LF and B to the HF, removed the straps. I worried about trying this as the SL-4 is rated at 4 ohms, but WTF, here goes.

It sounded louder and brighter, NOT better!. I figured it must be because of the two 4 ohm loads running in paralell.

Granted, SL-4 is NOT JBL, but ohms law is independent of a name plate. Maybe the results would be different with a pair rated at 8 ohms. The difference might just be due to the lower ohm load, when using the ONE amp/bi-wire set up. But then again...............:)

[Valentin]

I think Ducatista47 is on the spot



If an electronic or electrical device does not permit a signal to pass through it, and it is part of a series circuit, that signal will not flow and therefore will not be drawn (from the current source - the amp).


and whatever passes will become heat

[Earl K]

Hi


- I don't "Bi-Wire" because I Biamp most everything.
- If I didn't Biamp my first inclination would be to BiWire using 2 separate amps.


That's the part I'm just not "picturing", and I'll assume it's either a gap in my learning or a matter of me misunderstanding semantics. Passive crossovers, at the speaker side of the cable, "drawing" only select frequencies through the amp-to-cabinet cable...I'm just not getting it.

- To convince yourself that this can and does happen ( ie; a passive Xover "telling" the amp what frequencies to transmit ) ;

(i) setup the input of an RTA across the outputs of an amp ( use appropriately sized, high value isolation resistors to protect the RTAs' input circuitry and minimize loading for this test ) . The amp needs to be a typical DC coupled transistor type ( ie; voltage follower ) .
(ii) load the amp with an passive Hipass circuit ( or Lowpass ) connected to an appropriate sized, 20 Watt, dummy "load resistor" . The amp needs to directly "see" the LC elements for this test to work .
(iii) send a full bandwidth pink noise signal into the amp and observe the RTAs' display . You should see ( if you followed my abridged directions ) that the shape of the pink noise output has been roughly attenuated to the expected shape ( according to what one would expect from the LC elements in the passive filter )




I'd see that the reaction of the loads would differ between x-over/LF and x-over/HF in multi-amp scenarios, maybe it's the single amp having the wires common at the output posts that makes it a tough sell for me...but, maybe, with more time trying to picture it, it's looking like there's a difference. Hmmmm...I'll be quiet and watch the proceedings for a while.

- I agree that using a double set of wires on just a single amp challenges ones' comprehension of the "available benefits" . As a result, it seems to me that this particular configuration comes down to, empirically searching for a different "sound of wires" that is the most pleasing .

:)

[Titanium Dome]

Well done, as usual, Earl.

[MJC]

When the PS is stacked, I don't consider the LE14H-3 as a sub, rather a woofer. The crossover is up to 130 Hz, rather than the 80 Hz (THX Standard) or 100 Hz I might want for a sub. However, the feed is a single wire, not a bi-wire, so... :dont-know
But don't you have the option to wire the PT800 and PS1400 with separate wires from the amp/receiver? And/or running a coax from the sub out to the PS1400?
I know the PT800 is down 6db @ 80 htz, so I guess only down 3db @130 htz? The manual shows the PT800 with FR from 80~22K (-6db)

[Ian Mackenzie]

"The majority of current flowing between an amplifier and speaker is devoted
to bass reproduction. In fact, 60% or more of an amplifier’s output is destined
for the woofers. When current flows through a wire, it produces an
electro-magnetic field (EMF) that expands and collapses at a rate equal to
that of the music’s complex frequency components. If a single speaker
wire must conduct the full musical frequency, the preponderance of lowfrequency
information can interact with or modulate high frequencies.
The resulting intermodulation can create audible changes to the high
frequencies even before they reach the loudspeakers.
By using separate cables for high and low frequencies, unwanted high-
frequency modulation is avoided. Low-frequency information flows
through one cable while high frequencies flow through another.
Having the opportunity to use separate low- and high-frequency cables
allows you to use cables best suited to each frequency range. Optimally,
you should use high-quality, audiophile-grade speaker cables for both lowand
high-frequency conductors. However, let your ears guide you to the
Frequency-optimized Cables.""


From the JBL 4344Mk11 owners manual.


IMHO if you had a blameless amp and blameless cable none of this would be an issue. Keep your cable short and stout and use an amp with a high damping factor at all frequencies, not just tthe low bass at 20hz.


Some amps will react to the back emf anyway and this is one of the key reasons systems with large woofers seem to benefit from true biamping.


The back emf of the woofer can also make its way into the mid range crossover filter in full passive mode and cause intermodulation issues unless carefully designed. (most of the 43XX series).


Its a case by case thing.


Ian

[toddalin]

Hi


- I don't "Bi-Wire" because I Biamp most everything.
- If I didn't Biamp my first inclination would be to BiWire using 2 separate amps.



- To convince yourself that this can and does happen ( ie; a passive Xover "telling" the amp what frequencies to transmit ) ;

(i) setup the input of an RTA across the outputs of an amp ( use appropriately sized, high value isolation resistors to protect the RTAs' input circuitry and minimize loading for this test ) . The amp needs to be a typical DC coupled transistor type ( ie; voltage follower ) .
(ii) load the amp with an passive Hipass circuit ( or Lowpass ) connected to an appropriate sized, 20 Watt, dummy "load resistor" . The amp needs to directly "see" the LC elements for this test to work .
(iii) send a full bandwidth pink noise signal into the amp and observe the RTAs' display . You should see ( if you followed my abridged directions ) that the shape of the pink noise output has been roughly attenuated to the expected shape ( according to what one would expect from the LC elements in the passive filter )





- I agree that using a double set of wires on just a single amp challenges ones' comprehension of the "available benefits" . As a result, it seems to me that this particular configuration comes done to, empirically searching for a different "sound of wires" that is the most pleasing .

:)

If this truely is the case, it seems to me a very simple test would suffice.

Bi-wire the amp from a single channel to the LF and HF sections of the speaker.

Play a very low sine wave (<60 hz) though the amp.

Put a voltmeter across the HF wire. If no current flows because of the noted series circuit, voltage should be 0.

Repeat the test, but put an ampmeter in series with the HF wire.

Play the low tone. If no current flows because of the series circuit, no amperage should show on the meter.

Seems pretty simple to me. ;)

[Earl K]

Bi-wire the amp from a single channel to the LF and HF sections of the speaker.

Play a very low sine wave (<60 hz) though the amp.

Put a voltmeter across the HF wire. If no current flows because of the noted series circuit, voltage should be 0.

Repeat the test, but put an ampmeter in series with the HF wire.

Play the low tone. If no current flows because of the series circuit, no amperage should show on the meter.

- This looks to be a viable test ( to show / or not ), that the reactive impedances of the individual LC circuits, dictate the flow of voltage ( & current ) into their respective sections ( in a parallel type crossover ) .

- Are you going to try out this test Todd ?

:)

[toddalin]

- This looks to be a viable test ( to show / or not ), that the reactive impedances of the individual LC circuits, dictate the flow of voltage ( & current ) into their respective sections ( in a parallel type crossover ) .

- Are you going to try out this test Todd ?

:)

No, I can't easily access the back of my amplifiers as they are crowded within the entertainment center. (It's a two person job to open the entertainment center to access equipment.) Additionally, my cabinets are not set up for bi-wireing.

Actually, giving this a little more thought, the voltage test won't work. If I stick my voltmeter in a wall socket, it reads 120 volts, even if no current is flowing because it's reading the potential and not what's actually flowing. Of course this also infers that voltage from the low tone is also present at the HF terminals.

But, the series ammeter test should be valid. This should be done for both the LF and HF sections and the results compared.

But there are obviously several here who could try this very easily and report back. :blah:

[Zilch]

Todd's test is flawed.

If you run a LF test tone through the amp, that voltage will be present at the HF crossover input terminals, even if no HF current is flowing.

It's the current that will be zero, not the voltage....

[toddalin]

Todd's test is flawed.

If you run a LF test tone through the amp, that voltage will be present at the HF crossover input terminals, even if no HF current is flowing.

It's the current that will be zero, not the voltage....

Zilch: Look up!

[1audiohack]

How about an example from another angle.

Have you ever seen an extension cord on the floor do a flip when a sizable capacitor start motor is swithed on? A visual manifestation of counter electromotive force! Don't think for a second that does not distort the wave form in the conductor, or inductively effect other conductors in close parallel proximity.

Commercial power supplies that run shake tables, linear motors and the like must deal with large flyback voltages and currents. This is usally dissipated as heat by the output stages. This must be done effectively as to not allow the counter EMF to modulate / distort the power supply output waveform.

In audio we call that the dampening factor. The higher the number, supposedly the better equipped
is the amp to deal with counter EMF. Its the whole mass in motion thing, the woofer becoming a generator as its phisical motion goes out of phase with the output waveform.

Woofers can and do receive, and generate large amounts of current.

The bi-wire sceme wishes to let the amp deal with the counter EMF, as the shortest path to ground state, leaving the HF cables out of the picture, and the signal less effected,,, even though it is still connected to the same amp.

For the same reason, phones and computers have their own tap off the step-down transformers, somewhat isolating them from all the high current antics caused by the machinery on the shop floor.

1audiohack.

[Earl K]

Todd's test is flawed.

If you run a LF test tone through the amp, that voltage will be present at the HF crossover input terminals, even if no HF current is flowing.

It's the current that will be zero, not the voltage....

Hi Zilch,

- Well ( re; Todds' test with a HiPass filter in place on the HF circuit ), a voltage drop ( referenced to the voltage present at the amps' output terminals ) will be measurable at the HF drivers' input terminals ( as opposed to a measurement taken at the HF Xovers' input terminals ).

- Todds' statement ( test ) is sufficiently vague ( maddening so ), as to allow at least a couple of interpretations as to where the test leads would be placed . I can easily understand your interpretation of his words .
- Yes, it's sloppy wording / but I was hoping he would tighten it up after doing his own bench testing .

- for review ;


If this truely is the case, it seems to me a very simple test would suffice.

Bi-wire the amp from a single channel to the LF and HF sections of the speaker.

Play a very low sine wave (<60 hz) though the amp.

Put a voltmeter across the HF wire. If no current flows because of the noted series circuit, voltage should be 0.

Repeat the test, but put an ampmeter in series with the HF wire.

Play the low tone. If no current flows because of the series circuit, no amperage should show on the meter.

- Checking for current flow is still relevant to this discussion on Bi-Wiring ( in the wires leading to the HF portion of the crossover while a 60 hz test tone is played ) .
- ie; once it's determined ( & accepted ) that little LF content ( even in the single amp model ) flows into the HF crossover / then one can start thinking about empirically choosing cable types that "enhance" only the HF content ( a bit like choosing capacitors for their "sound" / and just as expensive in a "hit & miss" sort of way ) .

:)

[Ian Mackenzie]

In audio we call that the dampening factor. The higher the number, supposedly the better equipped
is the amp to deal with counter EMF. Its the whole mass in motion thing, the woofer becoming a generator as its phisical motion goes out of phase with the output waveform.

Woofers can and do receive, and generate large amounts of current.

The bi-wire sceme wishes to let the amp deal with the counter EMF, as the shortest path to ground state, leaving the HF cables out of the picture, and the signal less effected,,, even though it is still connected to the same amp.

For the same reason, phones and computers have their own tap off the step-down transformers, somewhat isolating them from all the high current antics caused by the machinery on the shop floor.

1audiohack.

I think your closer to a plausible theory than perhaps some of the other responses.

Its important to realise even a speaker cable has finite impediance.

If you have a bunch of filters daisy chained in parrellel along that length of cable with their loads producing a back e.m.f then the farthest point on the cable with a filter is most likely to be a point that is modulated by the previous load along that length of cable.

By offering each load its own dedicated cable back the reference voltage source (zero ohms) then in theory there is less possibility of modulation of the other loads. Of course the whole idea falls over if the amp reacts to the back emf.

I should point out solid most amps with large amounts negative feedback have a low output impediance at low frequencies. They can however be influenced by load characteristics because of stability or what might be call transient response damping to reactive loads as a result of large negative feedabck ratios.

I am less convinced of the idea of seperate cables for corresponding frequencies unless you have the crossover filter at the amp end of the cable.

In reality I confess to using Zilch's cat 5 twisted ends cable to biwire my JBLs when I am too lazy to biamp.

Ian

[jblnut]

I think the largest gains to be made here come from being able to use different amps for the highs and lows.
I'm not alone in using a large transistor amp for the lows and a smaller (two mono amps in my case) tube amp for the highs. It's great that many of the larger, newer JBLs allow for this type of configuration. I've not swapped out my speaker wire or tried anything different on the high end. I'm not a huge "cable guy" and I feel the money is better spent elsewhere.

jblnut

[Harkness]

it's a bit of a strange article, as it seems to suggest some of the benefits of bi-amping will occur with bi-wiring.

personally i think bi-wiring only makes sense in a single (stereo) amp situation. otherwise you are better off splitting the signal before it gets to the amps to avoid wasting amp power on frequencies that will be filtered out at the crossover.

no point in using a sweet little tube amp(s) for mids and highs and then sending a full-range signal through it. the bass range will eat up a lot of the amp headroom unnecessarily.

[jblnut]

Is this really the case ? My knowledge of such things is admittedly limited, but I believe that tube amp will not be called upon to deliver bass frequencies because the crossover network that the mids and tweeter are attached to will filter out anything that low. I can tell you that in practice the setup works extremely well. The tube amps are able to keep up with the transistor amp driving the low end, even though they only have 1/4 of its rated power. When I drive the system full range with only the tube amps, the output is far lower becuase they are now acutally powering the woofer.


jblnut



it's a bit of a strange article, as it seems to suggest some of the benefits of bi-amping will occur with bi-wiring.

personally i think bi-wiring only makes sense in a single (stereo) amp situation. otherwise you are better off splitting the signal before it gets to the amps to avoid wasting amp power on frequencies that will be filtered out at the crossover.

no point in using a sweet little tube amp(s) for mids and highs and then sending a full-range signal through it. the bass range will eat up a lot of the amp headroom unnecessarily.