Greenhorn question: Beware!

"Avoid using a series connection! Damping of both speakers by the amp is degraded!"

This is a misconception. Unfortunately it was printed in a JBL white paper on car audio, so the myth has persisted. And on the surface it seems to make sense.

If you were connecting a resistor in series with a speaker, the effective electrical damping (Qes) WOULD be degraded. The BL would stay the same while Re (effective resistance) would increase, and the net result would be a corresponding decrease in Qes. This is the same thing as saying the damping is decreased.

But when using series connection of two woofers, you are connecting a powered voice coil in series with a powered voice coil. The motor strengths (BL) add, just as the voice coil resistances and inductances add. If the two woofers are identical (not the case when one of the motors is an exciter), then the effective Qes (electrical Q, or electrical damping) is unchanged.

Obviously when the two motors in series are different, the effective Qes will not be the same as either one of them, but the resistor analogy still does not apply.

Consider this thought experiment: The first half of your woofer’s voice coil is in series with the second half of your woofer’s voice coil. Do they ruin one another’s damping factor? Of course not. And the reason is, they are both inside the field of an electric motor (the same one in this case), so both of halves of your voice coil bring a commensurate increase in motor strength (BL). This can be extrapolated to two voice coils within the fields of two separate motors, as when two woofers are wired in series.

I manufacture a subwoofer system that has woofers connected in series. If the damping factors were ruined by this configuration, the system would be unbearably boomy. You can take a look at Robert Greene’s April 2015 review of the Swarm and see whether he found the system to be boomy.

Duke

"Another thing -- the crossovers inside the speakers are engineered based on the expected impedance of the following driver (i.e. low-pass coil circuit for a single woofer). If you tack on a second speaker in series, the impedance in that frequency area will rise and cause the crossover frequency to move drastically. This will cause the frequency response of the first speaker to have all these bumps or holes - making a very imbalanced sound. That being said, the first entire speaker will also affect the crossover in the second speaker, therefore causing the same problems in that second speaker."

It’s not that bad. I’ve done it many times.

Doing a crossover section for identical drivers wired in series is no different from using a single driver. The crossover component values will differ, (inductance and resistance values will be doubled, and capacitance values halved), but that’s pretty much it. You might want to take the modified radiation pattern into account, if you haven’t already, but that’s an acoustic consideration, not an electrical one.

I've received three Golden Ear awards and one Product of the Year award from The Absolute Sound.  In each case, ALL of the drivers were wired in series with another identical driver.  So thus far I've had pretty good luck with drivers wired in series.

Duke

"Is a 4 ohm exciter speaker and a 6 ohm woofer box, wired in series (total 10 ohm) connected to an 8 ohm output amp OK? "

It’s "OK" in the sense that, AS LONG AS as you don’t push your amp into clipping and/or push the speaker or exciter beyond their comfort zones, I don’t think anything will blow up.

That being said, to the extent that the exciter’s impedance curve differs from that of the speaker, its presence in series with the speaker will change the way the speaker sounds. I expect that the exciter’s impedance rises significantly as we go up in frequency, so I expect that your speaker will have a lot less top end when the exciter is connected in series with it.

So I don’t see any reason why you can’t start playing music through them, but it may not sound very good.

"Is there a capacitor or inductor that could be added in my case, to bring the final impedance into a normal range?"

Imo not without measurements could an impedance equalization circuit be designed that results in the speaker sounding good with the exciter in series.

EDIT:  Typo in the second paragraph of my first post above.  Here is how the sentence should read, caps lock on for the corrected word: 

"The BL would stay the same while Re (effective resistance) would increase, and the net result would be a corresponding INCREASE in Qes."

Duke