IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Bi-Amping: basic questions and reality check
wgscott
post Feb 14 2011, 01:24
Post #1





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



I have a pair of Bowers and Wilkins speakers (CM7) that come with two sets of posts with a removable metallic plate that bridges them. When I first got these, I naively bi-wired them, and discovered the reason why I could not hear any improvement. (I used left-over non-designer 12-guage wire, so at least I didn't pay a huge financial penalty for my ignorance.)

One often hears "Bowers and Wilkins speakers are made for bi-amping," and so I've done a bit of reading up on horizontal and verticle bi-amping, and have tentitively concluded the following, but would like a reality check from those with more experience:

(1) For decent results, one should use two copies of the same amp, preferably vertically.

(2) You need to employ active crossovers between a common pre-Amp and each amp.

(3) You need to disable the passive crossover in each speaker (i.e., removing that metal plate isn't sufficient).

(4) Passive bi-amping is no better than bi-wiring, if you normalize for gain.

Assuming I have this right, does (3) involve a huge undertaking with large potential to damage the speakers, and does anyone actually do this? Does the sound significantly improve in quality compared to simply using a more powerful convential amp?


(Currently I drive these speakers with a Peachtree Nova, nominally 80Wpc at 6 Ω, and the recommended power is 50 to 150 Wpc at 8 Ω, so I am at the low end of the threshold and have a large room, so I am looking for possible improvements).

This post has been edited by wgscott: Feb 14 2011, 01:35
Go to the top of the page
+Quote Post
Xenion
post Feb 14 2011, 01:44
Post #2





Group: Members
Posts: 1041
Joined: 23-May 02
From: DE
Member No.: 2107



I have the B&W 703s and drive them with a Rotel 1080. I tested a bi-amping setup with a Rotel 1070 for mid-hi and the 1080 for the bass. I could not hear any improvement but i don't listen at high spl. i did not abx this but to me it was not worth the extra money. get a decent amp and you're fine (my opinion)
Go to the top of the page
+Quote Post
wgscott
post Feb 14 2011, 01:55
Post #3





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



QUOTE (Xenion @ Feb 13 2011, 16:44) *
I have the B&W 703s and drive them with a Rotel 1080. I tested a bi-amping setup with a Rotel 1070 for mid-hi and the 1080 for the bass. I could not hear any improvement but i don't listen at high spl. i did not abx this but to me it was not worth the extra money. get a decent amp and you're fine (my opinion)


Thanks for the quick reply. Did you use active crossovers?
Go to the top of the page
+Quote Post
bryant
post Feb 14 2011, 02:01
Post #4


WavPack Developer


Group: Developer (Donating)
Posts: 1291
Joined: 3-January 02
From: San Francisco CA
Member No.: 900



Well, I have no experience in this area, but I can offer a little common sense advice. The first thing is that yes, (3) sounds a little risky to me. Obviously accidentally hooking up the low-frequency amp to the high-frequency speaker (with its crossover removed) would be a disaster, but even the turn-on thump from an amp (or some other component) might cause damage to the tweeter. And Iím not sure exactly how you would wire this...if the CM7 was a 2-way speaker it would be straightforward, but since itís a 3-way you would still need some of the internal passive components in place.

The other thing Iíll comment on is (4). Since the separate amps are driving through the passive crossover, they do see an easier load than a single amp would. In other words, the high-frequency amp is not going to have to deliver a lot of low-frequency power because the impedance of that input is very high at low frequencies (or something would have to get hot inside, which I doubt). So that would be better than simply bi-wiring a single amp, but it would probably not be as good as a single amp that was twice as powerful (but if you already have one, it might be cheaper).
Go to the top of the page
+Quote Post
Xenion
post Feb 14 2011, 03:02
Post #5





Group: Members
Posts: 1041
Joined: 23-May 02
From: DE
Member No.: 2107



I did not use an external crossover. Using an external crossovers has major advantages in some points but if you don't have IIR tables from your manufacturer this can be quite tricky. It's not only about just setting a crossover frequency with a certain characteristic. If you want something with an active crossover and maybe some tweaking options, i would have a look on studio main-monitors (ks digital, klein+hummel etc.) BUT the CM7 is a very talented speaker that sounds quite good. I would use them in a passive setup with a powerful amp. Note that most B&Ws have a minimum impedance of 3,1 ohms which makes them a 4ohms speaker according to din/iec (B&W says 8ohms)
Go to the top of the page
+Quote Post
wgscott
post Feb 14 2011, 03:21
Post #6





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



OK, you both have convinced me not to bother. Mind if I hijack my own thread and ask if you think I would likely realize an improvement just adding a standard stereo 150 W power amp? I'd prefer class D, but of course don't want it to sound worse than what I have now.
Go to the top of the page
+Quote Post
Xenion
post Feb 14 2011, 14:28
Post #7





Group: Members
Posts: 1041
Joined: 23-May 02
From: DE
Member No.: 2107



QUOTE (wgscott @ Feb 14 2011, 03:21) *
just adding a standard stereo 150 W power amp?


If adding means bi-amping: make sure that your additional amp has the same amplification factor. Generally speaking: upgrading from undersized amps to something better can make a huge difference but i don't know your amp, so i can not tell you.
Go to the top of the page
+Quote Post
2Bdecided
post Feb 14 2011, 14:48
Post #8


ReplayGain developer


Group: Developer
Posts: 5104
Joined: 5-November 01
From: Yorkshire, UK
Member No.: 409



QUOTE (bryant @ Feb 14 2011, 01:01) *
So that would be better than simply bi-wiring a single amp, but it would probably not be as good as a single amp that was twice as powerful (but if you already have one, it might be cheaper).
There are supposed to be at least two theoretical advantages in terms of quality, beyond the fast that passive crossovers waste power (which doesn't really matter in itself IMO)...

The first one is that you can use more accurate components in an active (low voltage, before power amplifier) crossover than in a passive (high power, in the speaker itself) crossover. You can even do it digitally if you wish. Anything is possible. You can avoid component value drift.

The second one is that removing the components between power amp and speaker allows much better control of the speaker itself. Speakers are electromagnetic devices - things like overshoot (for example) send signals back into the amp, which a decent amp will fight against. Putting passive components in the signal path reduces the efficiency of this control.

In terms of ABXing any differences - no idea.

In terms of overall quality (subjective or objective) at a given price point - extra amps add to the cost - so at a given overall budget this approach might be worse than the simple one.

In money-no-object scenarios, then of course you'd do the cross over digitally and have one discrete amp per speaker cone/driver. You'd also use digital technology to improve the speaker's frequency response. More importantly, you'd choose and design the physical characteristics of the speaker far more carefully in the first place, making any improvements from all this other fancy stuff much harder to detect!

Cheers,
David.
Go to the top of the page
+Quote Post
wgscott
post Feb 14 2011, 17:41
Post #9





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



QUOTE (Xenion @ Feb 14 2011, 05:28) *
QUOTE (wgscott @ Feb 14 2011, 03:21) *
just adding a standard stereo 150 W power amp?


If adding means bi-amping


Sorry, to clarify, this would mean using a standard stereo ~150W (preferably class D) power amp instead of the built-in 80Wpc amp. (The nice thing about the Peachtree Nova is you can use it as a stand-alone DAC, a stand-alone DAC+pre-Amp, or as DAC/preAmp/Amp.)

eg: If I got, say a Rotel 1562 (100 Wpc) or 1572 250 Wpc, or similar...

This post has been edited by wgscott: Feb 14 2011, 17:42
Go to the top of the page
+Quote Post
DonP
post Feb 14 2011, 20:52
Post #10





Group: Members (Donating)
Posts: 1471
Joined: 11-February 03
From: Vermont
Member No.: 4955



QUOTE (wgscott @ Feb 13 2011, 19:24) *
(1) For decent results, one should use two copies of the same amp, preferably vertically.


I'm not sure what you mean by vertically.

I've always put one amp on bass and one on treble.
Depending on the crossover frequency (but usually) you will need more power for the bass half, so put the bigger one there if the power doesn't match.

I've been using various biamped systems for about 15 years. Mostly using 4 or 5 channel amplifiers (made for home theater, they go cheap once the new models have more channels), and all with line level crossovers. You say your speaker was made for biamping. If that means the crossover in the speaker can handle it (presenting high impedance for out of band signal) that should work.

Go to the top of the page
+Quote Post
wgscott
post Feb 14 2011, 21:42
Post #11





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



QUOTE (DonP @ Feb 14 2011, 11:52) *
I'm not sure what you mean by vertically.


One stereo amp for left, another for right.

QUOTE
You say your speaker was made for biamping.


I've heard it claimed. I don't know. It seems to me if it were "made for" it, they wouldn't require you to modify the speaker. So I am asking if there is any merit to the claim...
Go to the top of the page
+Quote Post
mixminus1
post Feb 14 2011, 23:30
Post #12





Group: Members
Posts: 688
Joined: 23-February 05
Member No.: 20097



Bi-amping with the speakers' internal passive x-overs is of dubious value as the amps still have to "deal with" the passive x-overs. As has been noted, true bi-amping means connecting the drivers directly to their own dedicated amplifier channels and doing all crossovers and equalization at line-level before hitting the amps. There are real improvements to be realized there, but you have to not only bypass any passive x-overs, you must also know the exact performance specifications of each driver in order to set x-over points and equalization correctly - IOW, it's not something that can readily be done to an existing commercial loudspeaker.

To give you an idea of the range of opinions on bi-amping amongst the loudspeaker manufacturers themselves, consider that I've seen $300/pair speakers with bi-amp connections (the connector cup took up nearly the entire back panel of the enclosure), while Dynaudio's top-of-the-line (3-way) Evidence Masters, at $80K/pair, have only a single pair of binding posts.


--------------------
"Not sure what the question is, but the answer is probably no."
Go to the top of the page
+Quote Post
Roseval
post Feb 15 2011, 00:08
Post #13





Group: Members
Posts: 482
Joined: 26-March 08
Member No.: 52303



I agree with mixminus1
Assuming a 2 way system:
1 - Amp to binding post= amp drives 2 drivers through a low pass and a high pass filter
1a Ė Bi-wiring= amp with 2x2 wires to binding posts
2 - Bi-amping=1 amp drives a driver through a low pass filter + 1 amp drives a driver through a high pass filter.
3 - Active: 1 power amp is connected straight to the woofer, 1 amp is connected straight to the tweeter. X-over is done on the signal before it enters the power amp.

1= normal ďhifiĒ
1a= audiophile (doubling the wire budget)
2= more audiophile (doubling the amp budget + wire budget for marginal improvement)
3= studio monitors (standard in the pro-world, maximise transparency by lowering distortion). Most of the time much cheaper than option 2.

Bit more on X-over: http://thewelltemperedcomputer.com/KB/Crossover.htm


--------------------
TheWellTemperedComputer.com
Go to the top of the page
+Quote Post
wgscott
post Feb 15 2011, 01:11
Post #14





Group: Members
Posts: 22
Joined: 10-May 10
Member No.: 80534



Thanks everyone. I think the genral conclusion is that it would be audiophoolish for me to do this with my current system. I appreciate the feedback (and money saved).

In the unlikely event I ever do this in the future, it seems a digital approach would probably be superior. I guess that means two DACS, etc, as well, and appropriate software. Some other lifetime...

This post has been edited by wgscott: Feb 15 2011, 01:16
Go to the top of the page
+Quote Post
STORMINORMAN
post Mar 2 2011, 02:04
Post #15





Group: Members
Posts: 1
Joined: 26-February 11
Member No.: 88515



Have a Rotel pre/processor with discrete outputs. Also a B & K pre/processor as well, if that might work better for some reason. My plan is to use a Yamaha MX-1000 Power Amp (260 RMS per) to run the LF & RF speakers, and a separate 100 RMS (or so) power amp for the LR & RR speakers. I want to know if I use a B&W speaker that is set up for what they refer to in the manual as "bi-wire" and take the pre/processor's center channel RCA output through an adaptor (male to 2 female) then a standard male to male RCA cable to a third stereo amp (probably also a Yamaha, M-50 or MX-630 @ approx. 135 RMS X 2) and thus bi-amp the B&W, while maintaining coherent speaker wiring, vis-a-vis (+ vs. -) , will I run into any major problems? These amps have individual gain controls on their front panels to aid in setting a correct volume level. I believe the pre/processors also have provisions for that as far as the individual channels are concerned, but not as far as this proposed "bi-amping" of the center channel speaker w/b.

It would seem to me that the B&W is made to have all the signal carried to the all speakers because of the bridge between the two sets of terminals, therefore the crossovers should protect the mid & tweeter from low frequency signals and vice-versa for the woofer????

I appreciate any help you can offer.

Regards,

Norm
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 2 2011, 13:52
Post #16





Group: Members
Posts: 3700
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (wgscott @ Feb 14 2011, 19:11) *
Thanks everyone. I think the genral conclusion is that it would be audiophoolish for me to do this with my current system. I appreciate the feedback (and money saved).

In the unlikely event I ever do this in the future, it seems a digital approach would probably be superior. I guess that means two DACS, etc, as well, and appropriate software. Some other lifetime...


If you were serious about bi-amping a given set of speakers, here is how I would proceed:

(1) Make such internal modifications to the speaker systems as would be needed to terminate each driver's voice coil externally.

(2) Obtain one of the many "Loudspeaker management systems" that are sold for professional use, starting with the Behringer DCX 2496 at the low end. (The DCX can be very effective).

(3) Obtain a complete loudspeaker analysis system - measurement mic(s), preamplifier(s), audio interface, computer, software etc.

(4) Hook up speakers, amps, management system etc.

(5) Make logical adjustments to the speaker management system to obtain the desired system performance in accordance with measurements and subjective judgements.

(6) Enjoy!

In short, throw away most of the design of my existing speakers other than the drivers and box design, and design new speakers for myself almost from scratch.

Or, just cut to the chase and build new boxes and select new drivers and start with that.
Go to the top of the page
+Quote Post
sshd
post Mar 2 2011, 14:34
Post #17





Group: Members
Posts: 207
Joined: 16-June 03
Member No.: 7218



I have modified my Snell CV for tri-amping with external crossover.

Setup like this:

* Bypassed passive crossover and put new connectors on the back of speakers: One set for low, mid and high frequencies.
* Meassured room and speakers with Audiolense
* Playback chain: foobar2000 with foo_convolver_gapless (FIR filter from Audiolense) > Soundcard > External DAC > dbx Driverack PA+ crossover > Rotel RSP 1068 8 channel pre-amp > 3x Rotel RB-1090/1070 power amps > Snell CV speakers

The modification made a huge impact on the sound. Hard to describe in no-nonsense terms, but everything sounds less distorted. Bass is a lot less muddy. Apparently 2x200W is not enough to drive these speakers with passive crossover.
I replaced the tweeters, so it makes no sense to talk about this improvement.

I initially had a stereo pre-amp between the DAC and crossover, but much better to put a multi-channel pre-amp after the crossover. This is probably because the crossover does it own analog->digitial->analog conversion after the DAC, so volume changes before the crossover affects the available bit depth quite a lot. Also there was a lot of annoying hiss in the speakers.

Going to change the setup someday so the digital signal is fed directly into the crossover.

Quite expensive project, but well worth it.
Go to the top of the page
+Quote Post
BearcatSandor
post Mar 22 2011, 22:39
Post #18





Group: Members
Posts: 149
Joined: 18-May 10
From: Montana, USA
Member No.: 80732



I hope this isn't a thread-jack, if so please cut it off and accept my apologies.

A few times people have mentioned the benefits of bi/tri-amping as far as load and control, but are there audible benefits?

Is this why a lot of studio speakers are active and have one amp per driver? What's the relationship to that?


--------------------
Music lover and recovering high end audiophile
Go to the top of the page
+Quote Post
Roseval
post Mar 22 2011, 22:59
Post #19





Group: Members
Posts: 482
Joined: 26-March 08
Member No.: 52303



Assuming a 2 way system
Bi-amping refers to driving a passive speaker having 2 x 2 binding post with 2 amps.
The passive crossover remains in place.
Bi-amping doubles your power amp budget.

In pro-monitors the amp is connected directly with the driver, without any filter.
The needed filtering is done before the signal enter the power amp.
This is called active speaker.
Active monitors are claimed to be more transparent then passive ones.
Although they have an amp per channel, prices of pro-monitors are low compared with traditional HiFi.

Have a look at Adam, KRK, Genelec, Dynaudio, etc



http://thewelltemperedcomputer.com/KB/Crossover.htm

This post has been edited by Roseval: Mar 22 2011, 23:00


--------------------
TheWellTemperedComputer.com
Go to the top of the page
+Quote Post
DVDdoug
post Mar 23 2011, 00:27
Post #20





Group: Members
Posts: 2566
Joined: 24-August 07
From: Silicon Valley
Member No.: 46454



QUOTE
A few times people have mentioned the benefits of bi/tri-amping as far as load and control, but are there audible benefits?
Well, you can't say one approach is always better... You can build a good sounding passive-crossover system or a good sounding bi/tri-amped sytem. Just as there may be benefits to a 3-way system over a 2-way system, but you can't always say one sounds better.



QUOTE
Is this why a lot of studio speakers are active and have one amp per driver? What's the relationship to that?
wink.gif Marketing might be involved... It might actually be cheaper to build a bi-amped active monitor. Electronic parts are fairly cheap, and if you put two amps on the same circuit board, powered by the same power supply, two small amps might be about the same price as one big amp. And you save the passive-crossover cost. Good quality coils & and large-value (non-electrolytic) capacitors for passive crossovers are relatively expensive. And, you can make a more precise (or more advanced) active crossover than passive crossover.

Cost is always a major factor. You want the best sound for a given cost, and/or low cost compared to the selling-price for maximum profit.

QUOTE
Bi-amping refers to driving a passive speaker having 2 x 2 binding post with 2 amps.
The passive crossover remains in place.
Actually, with speakers that are designed only for bi/tri amping, there is no passive crossover. Large-scale P.A. systems usually have only active crossovers, although there might be a "protection" capacitor on the horns.
Go to the top of the page
+Quote Post
BearcatSandor
post Mar 24 2011, 01:38
Post #21





Group: Members
Posts: 149
Joined: 18-May 10
From: Montana, USA
Member No.: 80732



Thanks for the responses. That makes sense. Those brands are on my short list of items to audition. I'm especially interested in the Genelec DSP models though they are pricey, especially since i want to set up an ambisonic system.

This post has been edited by BearcatSandor: Mar 24 2011, 01:38


--------------------
Music lover and recovering high end audiophile
Go to the top of the page
+Quote Post
thesurfingalien
post Mar 25 2011, 20:30
Post #22





Group: Members
Posts: 138
Joined: 26-February 11
Member No.: 88533



QUOTE (DVDdoug @ Mar 22 2011, 20:27) *
QUOTE
A few times people have mentioned the benefits of bi/tri-amping as far as load and control, but are there audible benefits?
Well, you can't say one approach is always better... You can build a good sounding passive-crossover system or a good sounding bi/tri-amped sytem. Just as there may be benefits to a 3-way system over a 2-way system, but you can't always say one sounds better.



QUOTE
Is this why a lot of studio speakers are active and have one amp per driver? What's the relationship to that?
wink.gif Marketing might be involved... It might actually be cheaper to build a bi-amped active monitor. Electronic parts are fairly cheap, and if you put two amps on the same circuit board, powered by the same power supply, two small amps might be about the same price as one big amp. And you save the passive-crossover cost. Good quality coils & and large-value (non-electrolytic) capacitors for passive crossovers are relatively expensive. And, you can make a more precise (or more advanced) active crossover than passive crossover.

Cost is always a major factor. You want the best sound for a given cost, and/or low cost compared to the selling-price for maximum profit.

QUOTE
Bi-amping refers to driving a passive speaker having 2 x 2 binding post with 2 amps.
The passive crossover remains in place.
Actually, with speakers that are designed only for bi/tri amping, there is no passive crossover. Large-scale P.A. systems usually have only active crossovers, although there might be a "protection" capacitor on the horns.


Here are a few reasons why active x-over and consequently separated amps for loudspeaker-units quality-wise is a good solution:

- the designers have absolute control over the whole system ("Total Q" of the system), which is a mayor design advantage. For regular passive-filtered loudspeakers, manufacturers need to take an average Q-value for components (the amp and cabling used by the costumer) that are not within their control, which basically is a compromise.
- amplifiers used in an active setup are serving a limited frequency-range. This leads to a lower inter-modulation distortion.
- an active filter does not have to deal with high-power signals and will, in general, have less "destructive" influence on signal quality compared to passive filters.

Regards,
Peter
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 28th August 2014 - 01:10