bi-amping phase

[onplane]

Last week I bi-amped my AR-3a's and this week I bi-amped a set of JBL's. Bi-amping improved the sound coming from both the 3a's and the JBL's. The improvement on the JBL’s, however, was not nearly as dramatic as on the 3a's

Bi-amping the modern JBL’s was easier than the 3a’s, because the speaker is designed with terminals for bi-amping.

In both cases, I used “horizontal” style bi-amping, because I’m cheap and had “spare” receivers/amps lying around.

Today I ran an experiment on the JBL’s. I “reversed” the phase of the tweeters relative to what it was and I think the sound is a little “brighter”. I say, “I think” because there is no way I can do an A/B comparison. I have to physically switch the speaker contacts and then listen to the same track again.

In any event, the difference is so slight as to be almost insignificant.

Now, let me be clear, the tweeters are in phase with each other. I just reversed the leads on both tweeters, so the phase relative to the woofer changed. Notice I didn’t say that the tweeters and woofers were in phase initially, because I have no way to know whether they were or weren’t. Since I’m using amps from different manufacturers, I have no way to judge the phase on the output signals from the different amps.

That is, it’s quite possible that when I “reversed” phase on the tweeters, I actually brought them into phase with the woofers. Further complicating this is that the phase between the two amps most likely varies over the audio spectrum.

Anybody have any comments on how phase impacts bi-amping?

It might prove that phase is NOT terribly significant when speakers are operating independently in different frequency bands. I have to admit, this just doesn’t seem correct. My ears, on the other hand, appear to be telling me this is the case.

If I use the analogy of a rock band full of electronic instruments, I can’t imagine that any of those amps are “in phased” with each other.

[rrcrain]

The major impact of having the tweeter/woofers out of phase will be the loss of a soundstage. Rather than listening to the tweeter versus the woofer when you revers the phasing, listen to instrument placement and how well te bass and treble merge or blend together. Very seldom does an instrument such as a giutar or violin produce a single, pure tone but rather they produce a spectrum of frequencies that meld together properly, giving the instrument its signature sound.

There are also test CDs available which contain recordings of the left and right channels in and out of phase. You might want to find one.

[onplane]

Richard, thanks for your response. The more I think about this the more confused I get.

If you take the example of a simple two way speaker mono-amped with passive xover, won't there be a different phase shift seen by the tweeter as opposed to the woofer due to the xover? I mean, Richard, are they ever actually "in phase".

Seems to me the only way to get these two speakers in phase is by means of vertical bi-amping with the xovers completely removed.

Then when you get to three way speakers with more complex xovers, I just get a headache trying to figure out what's in phase with what.

In my horizontal bi-amping approach, I suspect is not as simple as "in phase" or "out of phase". Those two descriptors assume phase shifts of zero or 180 degrees over the entire audio spectrum. Supposing the two amps are out of phase by 90 degrees at 1000 hz , 45 degrees at 100 hz and 125 degrees at 10K Hz. Then since the xovers are still intact, the xovers will introduce additional phase shifts that are frequency dependent.

Richard, I did see somewhere on the net where someone had designed an xover network where the two speakers where stacked (bottom of one speaker directly feed the top of the other). So, the speakers were in series, but it’s far more complicated than that. There were a bunch of passive elements (coils and caps) that allowed frequencies a “shortcut” to bypass the speaker those frequencies where NOT intended to drive. For example, a cap across the woofer would allow a shortcut around the woofer, but the cap would be in series with the tweeter. Something like ….

http://www.classicspeakerpages.net/dc/user_files/1464.jpg

[rrcrain]

Don't worry about the drivers being in our out of phase. Your main, and only concern as an end used is making sure the two amps driving the speakers in a biamp configuration are in phase. The engineers who designed the speakers and crossover may have designed the drivers to be in phase, or out depending on driver orientation and design.

An example I'm familiar with: Side firing woofers in speakers such as the AR90 are 180 degrees out of phase from the other three drivers, but if you closely study the schematic, you soon discover the negative terminal for the lower mid, upper mid and tweeter are connected to the positiver speaker terminal. Technically speaking, one could argue that the three drivers are 180 from the music and the bass is "in phase".

The bass crossover was designed this way to compensate for the side firing woofers and it's behoves us not to mess with the original intent unless we really know what we are doing.

In the end, don't worry about phase alignment except from right to left and that the two amps are in phase. Generally, as long as both amps are either solid state or tube, they will match.

Caps and inductors in a crossover have two simple functions(and several comples ones); either to block or pass certain frequencies. Since I'm not a design engineer nor do I own a crossover simulation program, I refuse to mess around with my crossovers design.