Two speaker enclosures in series or parallel??

Status
Not open for further replies.
I have two **identical** pairs of Infinity RS7s speakers whose nominal impedance is between 4 and 8 ohms. I want to "stack" these speakers side-by-side or one on top of the other with one upside-down--sortof a large W-M-T-T-M-W arrangement and driving them with one amp, maybe two, but prefer one.

Q1 If I run these boxes in parallel giving a nominal load of 2-4 ohms, providing my amp can cope with the 2 ohm load impedance, there should be no problems (right?).

Q2. What happens if I run the two *boxes* in series (using their outside terminals) to provide a nominal 8 to 16 ohm load to the amp? Will each box's XO still work OK? I guess I will end up with a sortof series-parallel XO of the six drivers on each channel. Can anyone see anything seriously wrong with this scheme.

The alternative is to disassemble the XOs and re-build them (so Cs are in seies to give 1/2C or // to give 2xC value, Ls are put in series to produce 2x L etcetera), wire each pair of drivers in series and put the whole lot back together as an 8ohm system. Questions above relate to seeing how it sounds "as-is" and possibly using existing XOs unmodified. I appreciate that inductors in series is not ideal, but caps in parallel or series should be OK, resistors in series or parallel also OK. 5 ohm L-pads not ideal either, but could easily be replaced.

If no-one has guessed by now I am considering converting one speaker design (2 pairs of Infinity RS7s--3-way sealed box) into one pair of pseudo RSIIs (3-way with sealed box woofers and dipole mids and tweeters. main difference is that RS7 has 8" woofer, RSII 10" woofers. Bi-wiring and bi-amping may be necessary to solve difference between bass & mid sensitivities. Mids and tweeters are identical in two models. NB rear tweeter is wired reverse-phase to give dipole effect to match mids. I am also thinking of isobarik BR design for bass box to save space--comments? BTW original RS7 enclosures are pretty basic vinyl veneered chipboard boxes that are hardly worth saving.
see:
http://www.infinity-classics.de/inf...ries-1+b-2-10+A+B-1981-86/index-Reference.htm

for various models of Infinity Reference series enclosures.

The driver compliment of the two designs are very similar and there are many common components in XOs. Any suggestions, warnings will be carefully considered and appreciated.
An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.

An externally hosted image should be here but it was not working when we last tested it.
 
Well, you gave the critical parameter yourself -

.. providing my amp can cope with the 2 ohm load impedance...

There is some flexibility in Amp output ratings, though certain amps are more sensitive than others to the heavy current draws required by excessively low impedance. For a common, off-the-shelf consumer brand-name amp, I wouldn't trust it below 3 ohms, and with some brands 3 ohms might be too low for long term at high output.

Pro amp, like PA amps, guitar amps, or other amp specifically meant to source high currents, this is probably not a problem, and the specs for those amp, if you can get them, will probably bare that out.

As to the Crossovers in a Series configuration, just one man's opinion, but I think the speakers will operate independently. They will see a certain voltage applied to them, see certain frequencies in the incoming signal and respond accordingly. In my opinion, the crossover is not something to be concerned about.

The website you referenced doesn't give an impedance response curve or graph, so we don't know if the speaker briefly dips down to 4 ohms at certain extreme frequencies, or if it lingers there for any length of time.

As a generalization, the impedance tends to dip at the very lowest of the low frequencies and in the upper mid-range. You can see a variety of sample impedance charts at this website

Part Express Project Show Case-
http://www.partsexpress.com/projectshowcase/homeaudio.cfm

Select a few speakers at random, then on the speaker page, select 'Crossover Design', and at the bottom will be a frequency response and impedance graph.

One thing to keep in mind when comparing Series and Parallel arrangements is that in Parallel, the speakers get the same Voltage, in other words they receive the same signal, but the current is divided between them. When they are in Series, they share the signal, that is, it is divided between them with each getting half the drive signal or voltage, but they share the same current.

Usually, the reduced signal is offset by the fact that you are moving twice as much air. The exact measure of the trade-off and gains, I don't have available to me.

In the end, it is as simple as this, your Amp absolutely WILL handle the Series load, it may or may not handle the parallel load. The final question is, how much risk are you willing to take.

That probably didn't help much.

Steve/BlueWizard
 
Series connection is perfectly ok. Since the impedances of the two speakers are identical, half the voltage will end up over each speaker, for any frequency. In other words, it is equivalent to driving each speaker with a voltage source with half the voltage.
 
Q2. What happens if I run the two *boxes* in series (using their outside terminals) to provide a nominal 8 to 16 ohm load to the amp? Will each box's XO still work OK? I guess I will end up with a sortof series-parallel XO of the six drivers on each channel. Can anyone see anything seriously wrong with this scheme.

If you put the 2 speakers in series, it will be fine.

The XO doesn't make the unwanted frequencies disappear, it just makes it go somewhere else. For example, I think a low pass filter lets low frequency voltage go through to the driver, while higher frequency content shorts over to ground. So by using a crossover, the signal is "split" to the individual drivers, and then joined back together on ground. By wiring them in series, the positive terminal of the second speaker is connected to ground from the other speaker, thus getting all of the content for it's XO to filter out.


...I think....

It probably depends on what kind of XO it is.

Either way, try it out; it shouldn't damage anything if you wire it correctly.
 
Thanks guys, I thought what you all say may be the case, but it never hurts to ask.

Did anyone have further opinions on putting pairs of drivers in series and re-wiring XO as 8 ohms nominal? Or should I just put all the drivers in the layout arrangement I suggested (RSII pic) and just use the original XOs as is ie: going to their respective 4 ohm drvers? The existing L-Pads should cope with different output from running midranges open baffle rather than in their normally sealed enclosures. I will probably swap components for better quality ones after initial listening.

Regards
Ozziozzi
 
Status
Not open for further replies.