bi amp imrpoving?

"Keep in mind that it is not REALLY necessary to bypass, or delete, the existing crossovers in your speaker system, to experience most of the benefits of active biamping."

For MOST of the benefits of bi-amping, yes it is necessary to go to an active crossover. For one thing, by connecting the drivers directly to the amplifiers the speakers present a nearly perfect resistive load, which is much easier for an amp to drive. Bi-amping via passive crossovers has some benefit, but using an active crossover puts you in another world as far as engineering benefits.

Active crossovers are seldom used though, because an active crossover costs A LOT more to build than a passive crossover, you need multiple channels of amplification (perhaps as many as eight for two speakers), and it's better if the amps are engineered to the speaker system, with more watts in the lower frequency connections, etc. All of this conspires to relegate 99% of commercial products to passive crossovers, which are inferior.

07-10-10: Mhedges
Kal - why do you need a custom crossover? What's wrong with a of the shelf active crossover like the NHT X2?

That is because a crossover from a textbook (as must be all off-the-shelf designs) cannot compensate for the parameters of each and every set of drivers and cabinets. Even a choice of crossover slopes is inadequate to match the inherent variables of every hardware set.

For example, from the X2 page:

The X-2 is a general purpose active, electronic crossover. It mates with the NHT A-1, but can be used with any stereo or monaural amplifier. The X-2 is stereo/mono switchable and has a feature set designed to shape bass frequencies for optimizing room response. Note: The X-2 will not work properly with the NHT U2, iWS, or any Evolution series product,....

Kal

Most of the biamping that I've done, for myself and others, has been with planar mains(either Acoustat or Maggie), and transmission line woofers. To me; the greatest benefit to realize with biamping, is the cleaning up of the signal passed to the mains amp, via removal of bass freqs. That and the other benes mentioned in my first post, none of which would be minimized by the original system's crossover. If one were separating the drivers within a three way system, I'd have to agree that the woofer's crossover network would best be removed. The TS didn't specify the system he intends to employ, outside of mentioning, "bass, highs and mids".

Rodman, that is confusing two meanings for biamping. The usual meaning in most audio arenas is independently powering the drivers (or groups of drivers) within a commercial multiway system or by design of an active multiway system. In those cases, the bypass or removal of the passive high-level network (or its omission) is mandatory for the use of a custom external active low-level network. The reason is that you do not want multiple networks operating in the same bandwidth because they will interact in, often, unpredictable and unfortunate ways.

What you are directly referring to is the addition of (sub)woofer to an existing system. In that case, the built in crossovers are not operating in the frequency range where the external crossover to the (sub)woofer is operating. Thus, removal of the inbuilt one is not only not necessary, it is incorrect.

While the use of the term "bi-amping" for this is technically correct, it is not what is most often meant. The OP is terse to the point of obscurity but my interpretation is that he is asking about the use of additional amp(s) with an already existing system and not the addition of speakers.

Kal

Some of the above responses seem to imply that when "passively" biamping (i.e., using two amps on a biwirable speaker with no external line-level crossover added, be that active or passive -- "passive" biamping is actually a misleading/confusing term in that sense!), you don't get the benefits of relieving the upper-frequency amp from having to deal with the bass frequencies. Although passive attenuation (by the speaker's crossover) after the power output isn't exactly the same as filtering out those bass frequencies before feeding the signal to the treble-range power amp in the first place, there is still worthwhile benefit to be gained (at least in theory) by not requiring that amp to drive a woofer. The amp inputs may "see" the full frequency range, but the output section of the HF amp is not being called upon to deliver much current in the LF. This is of course in addition to the other theoretical benefits, like doubling the available amp power (assuming the top and bottom amps are identical).

I've only ever done passive biamping, but my feeling is that the main benefits of "active" biamping (again in theory - in practice, as has been pointed out above, all bets may be off under the wrong circumstances) are mostly in implementing the frequency-divider network at the low-power line level instead of the high-power speaker level. (Which, as was also pointed out above, relieves the power amps from having to deal with the more reactive load of a speaker's crossover -- in addition to eliminating the high power-handling requirement so the crossover can be miniturized, and removing it from the highly vibrational environment inside the speaker cabinet.) My guess is that this factor is much more significant than any benefit gained from restricting the frequency range seen at the power amp inputs per se. After all, to my knowledge no one has postulated a benefit in relieving the system preamp from handling the full frequency range by dividing the signal after the source and using two preamps...

BTW, I also have to disagree about the wisdom of using different and possibly dissimilar amps for the different frequency ranges (subwoofer amps excluded) -- in my (albeit limited ) experience, using identical or very similar amps (and cables) top and bottom is necessary for the sound not to risk becoming discontinuous, at least with speakers that are coherent/consistent top-to-bottom to begin with.