First Order Crossovers: Pros and Cons

I hear what you are saying about xovers, Roy.
But the fact remains that in there own pass bands tweeter and woofer are
working 90deg apart. If I now want to replay say a large cymbal with a fundamental at 440Hz and strong harmonics all the way beyond our hearing range (in our case past the xover point) it still means the fundamental will be 90deg out of phase with (some of) its harmonics.
I don't really care what vectors do as I don't hear vectors but I hear phasing.
We all do since, with the exception stereo recordings, all our spatial information derives from phase differences. You can test that next time you have a bad headcold that cloggs up one ear: Listen to your stereo and its like mono, go outside and you can easily tell where a noise comes from. This also works with a small ball of cotton wool, if you haven't got a cold handy.
But anyway, what I understand as phase coherent means that the entire output is in phase ideally independent of listening position.
The only speakers capable of this are full-range, single driver designs.
But Tannoy makes an acceptable(to me) compromise. Seperate drivers on a vertical line are just one step too far for me.

Still can't accept your time/phase 'explanations' it goes against everything I have ever learned and would directly contradict my two relevant college lecturers, my Professor at the Technical University Berlin ,
Guy R. Fountain ( Founder Tannoy)
Peter Walker ( Founder, Quad)
Peter Voigt ( Lowther )
and pretty much everybody else I know who's worked with AC current and/or acoustics. I don't think your lone voice is enough for me to budge on that one.Again lets look at sinewaves and lets only regard 3 points (max., null point and minimum)of it curve for the moment: to be time coherent 2 sine waves need only to start at the same moment, they could start at any of our 3 points: max and falling; min and rising; nullpoint either rising or falling.
Thus we have 4 ways in which our two sine waves can be in time.
To be in phase our 2 waves have not only start at the same moment but also the same point. There are now 3 ways in which our 2 waves are in time but not in phase.

You mention some distortion regarding my Tannoys, fair enough they distort. So do all speakers, but of course total distortion is very easily measured and mine measure up thus: for 90dB SPL, 50Hz-20kHz less than 1%;for 110dB less than 3%. How do yours do?

The thing with your test tones is quite amusing since you should be using pink noise or white noise to measure for phase coherence. Its not difficult to find a driver thats in phase with itself and one single tone from another but that does not make it phase coherent.

I'm sure you could hear the comb-filtering going on if you'd honestly compare to speakers which do not exhibit this particular problem.
I can and, compared to some people, my hearing isn't that good.
Its the comb filter effect thats (partially) responsible for the sweet spot ie the sweet spot is the area where the comb filtering is at its minimum. With speakers that emulate the point-source ideal (planars,Tannoy DC's and full-range drivers) this is much less pronounced although fr-drivers teend to produce their own version of the sweet spot due to beaming.

So by "phase coherent" you really mean that a first order filter remains entirely true to its 90deg phase separeation between tweeter and woofer?
Thats a new definition to me, maybe we have to start with defining every term properly first.
Also a speaker which has the drivers on a vertical axis can only be phase coherent at one point in space. This of course is a purely geometrical problem independent of any electrical feature.

Last but not least a time coherent speaker can easily be not phase coherent if the crossover separates the phase between drivers (as they all do). On the other hand a phase coherent speaker has got to be time coherent as well as a slight time mis alignment MUST result in a phase shift. Indeed a phaser works by splitting the signal in two and then time delays one with respect to the other before merging the signal again. If you modulate the delay you get a phaser/flanger sound , if you don't its a chorus. I hope you get the idea: all phasing is done by playing in the time domain.
Basically a phase coherent speaker is one that is not only in time but also in phase; a time coherent speaker is one thats in time but not in phase.
And most speakers are neither.

Tannoy manages to get theirs phase coherent by fitting the treble coil exactly half a wavelenght (at crossover) behind the bass coil and then using crossover characteristics and a very simple delay circuit. Thus there is no phase separation between tweeter and woofer at all (except some small aberration at crossover). They are thus in phase acoustically and electrically AND independent from the listening position.
And no I am not trying to sell Tannoy, I don't actually like the new ones with their plastic cones.
Personally, I think the aforementioned geometrical phase problem is the reason that conventional speakers have a sweet spot where everything sounds better and the stereo image snaps into place. Tannoy don't really have that: they sound the same where ever you are although the imaging shifts if you stand to the side, like looking through a window at an angle.

Having had a look at your website I see you claim a phase error of 2deg!
A bold statement, care to back that up with some measurments?
I am sure you've got access to an anechoic chamber and the necessary equipment.

Cheers Golix

Every step up in xover order shifts the phase between tweeter and woofer by an additional 90deg. This means for maximum phase-coherence the obvious choices are second order (180deg shift so you just invert the connections to one driver to get it all back in phase) or fourth order (360deg shift: back in phase but not in time, a simple delay circuit will fix this). The ones to avoid are first and third order as a 90deg (270 for the third order) is very dificult to fix.
One advantage of a first order xover is a small number of parts which results in a very good transient response. The other is their immunity to electrical resonance ("ringing"). The higher the order the xover the easier it is to induce resonance with a high enough power input. When such a filter resonates it actually turns into a sinewave generator and produces an output at the xover frequency! Almost all analog synthsizers use this effect very succesfully but you don't really want that happening in your speakers, to this end most higher order xovers contain a damping circuit to avoid this.

The only truly phase-coherent dynamic multiway speaker in existence (Tannoy DualConcentric) uses a second order xover and inverts the tweeter.
It is phase-coherent to within 18deg. The only way to get better than that is a full-range driver, be it dynamic or planar. The result of this is a stereo image
that is far more precise than any other dynamic speaker I have ever heard.
And, being a true point source, you are always in the sweet spot as long as you are somewhere between the speakers and, where ever you are in the room, the sound never changes! After having spent a fair bit of time in recording studios I can say I have never heard a speaker as close to the real thing as a Tannoy! None, at all. Electrostats have better micro resolution and a Klipshorn better dynamics but, apart from those and overall, theres just nothing that comes even close. I put this down to phase-coherence and the point source characteristic. There is also a marked difference if I invert the connection (on both speakers at the same time): the stereo image pretty much collapses ie it goes flat in depth and height and the bass goes soft.
With other speakers I've never heard a difference.