I third Kal's comment.
As far as amplifier loading is concerned, here is the impedance curve for the LS50. It reaches a minimum of 4 ohms at 200 Hz, although at an undemanding phase angle, and also reaches 5.3 ohms at 135 Hz in combination with a current hungry phase angle of -41 degrees. Both of those frequencies, of course, are in areas that for a lot of music require lots of energy.
If the speakers are paralleled, the amplifier will therefore have to drive 2 ohms at 200 Hz, and 2.65 ohms at a severe phase angle at 135 Hz. Few amplifiers in the price range that is likely to be used with these speakers, or anywhere close to it, will sound their best working into loads like that. And as Matt indicated, some will even shut down after a while.
If the speakers are connected in series, the amplifier will see twice the impedance of a single speaker, which will avoid those loading problems. However the TOTAL amount of power the amplifier will be able to deliver to both speakers will be cut approximately in half, relative to what it could deliver into a single speaker. And each speaker will receive only 1/4 as much power as it would if it were the only speaker connected on that channel.
In either case, Matt is correct that the speakers could end up being damaged. Depending on volume level, either of those scenarios could result in the amplifier being driven into clipping, especially given the relatively low sensitivity of the speakers. Severely clipped waveforms tend to destroy tweeters, because the abrupt transitions of their waveforms between musical content and flat tops and bottoms correspond to much greater high frequency content than the original musical waveform may have contained.
Tony's observations make perfect sense. Unless the listening position is perfectly optimized vertically, AND unless the speakers are inverted relative to each other (as he was clever enough to do), a given note will have at least two equally loud arrival times at the listener's ears. That will result in comb filtering, which will tend to be perceived as a general loss of clarity. Exactly as Tony described.
As far as aiming the second speaker backwards is concerned, it might make for some interesting experiments but I suspect that there are likely to be more productive ways of investing the $ that the second set of speakers would cost.
Regards,
-- Al
As far as amplifier loading is concerned, here is the impedance curve for the LS50. It reaches a minimum of 4 ohms at 200 Hz, although at an undemanding phase angle, and also reaches 5.3 ohms at 135 Hz in combination with a current hungry phase angle of -41 degrees. Both of those frequencies, of course, are in areas that for a lot of music require lots of energy.
If the speakers are paralleled, the amplifier will therefore have to drive 2 ohms at 200 Hz, and 2.65 ohms at a severe phase angle at 135 Hz. Few amplifiers in the price range that is likely to be used with these speakers, or anywhere close to it, will sound their best working into loads like that. And as Matt indicated, some will even shut down after a while.
If the speakers are connected in series, the amplifier will see twice the impedance of a single speaker, which will avoid those loading problems. However the TOTAL amount of power the amplifier will be able to deliver to both speakers will be cut approximately in half, relative to what it could deliver into a single speaker. And each speaker will receive only 1/4 as much power as it would if it were the only speaker connected on that channel.
In either case, Matt is correct that the speakers could end up being damaged. Depending on volume level, either of those scenarios could result in the amplifier being driven into clipping, especially given the relatively low sensitivity of the speakers. Severely clipped waveforms tend to destroy tweeters, because the abrupt transitions of their waveforms between musical content and flat tops and bottoms correspond to much greater high frequency content than the original musical waveform may have contained.
Tony's observations make perfect sense. Unless the listening position is perfectly optimized vertically, AND unless the speakers are inverted relative to each other (as he was clever enough to do), a given note will have at least two equally loud arrival times at the listener's ears. That will result in comb filtering, which will tend to be perceived as a general loss of clarity. Exactly as Tony described.
As far as aiming the second speaker backwards is concerned, it might make for some interesting experiments but I suspect that there are likely to be more productive ways of investing the $ that the second set of speakers would cost.
Regards,
-- Al