I'm thinking about building an eight speaker array for a particular project. It's quite easy to get the specifications for individual speakers but less easy for me to find out how they react to the presence of other inductors (ie the other speakers in the circuit). Has anyone any practical experience of speaker arrays? In your opinion do you get better low-pass characteristics from series, parallel or series/parallel configurations? I realise I could just build them and then suck it and see but I wondered if anyone had any thoughts about the matter?
In short: Speakers can be arranged in series and parallel so that the resulting coefficients (+1, -1, +1/2, -1/2) lead to high power, radial sound distribution, and reasonable price.
Worth looking at and even trying, just a bit of soldering....
I have no practical experience but my instinct would be to arrange
*nine* speakers in a series/parallel arrangement that presented the same impedance to the amp as one speaker. If it has to be eight speakers, I'd consider adding a dummy load to make up the number.
In general arranging units in paparellel is considered the best as that preserves low frequency damping and means if one blows they both dont go out.,
As far as crossover design goes, don't be too prissy. a single LC and possibly a senstivity matching R is fine per filter and wont give you excess phase shifts to worry about. You are not trying to produce ideal filter shapes, just getting HF out of bass units and vice versa. Its more inporatant not to drive LF energy through tweeters, and a series cap is enough for that, but you can stick an L across them as well.
Its not clear what the OP is trying to do. A Marshall 4x12 was always series parallel. tow of those in paralell makes an 8x1`2.
BUT a Marshall 4x12 isnt the 'closest approach to the original sound' cos there IS no original sound, apart from an uamplpified guitar string and its as uninteresting as hell.
Let's imagine speaker impedance drops to 1R at resonance (it won't IRL), and the amp's delivering 1v to a single speaker. That speaker gets 1A. Now imagine 2 speakers in series on 2v. Impedance is 1R each, so each speaker gets 1A again. IOW putting 2 identical speakers in series has no effect on their damping. I'm assuming you are familiar with what damping is.
The same happens when impedance rises, each series speaker still sees the same voltage, current & phase - and does so at every frequency. And that dictates damping.
For PA systems there are columns of drivers, all arranged to point in the same direction but different phasing affects the sound distribution (it's called a phased array) - there isn't much need to get hi-fi quality, just cover a lot of area with sound. Is that what you're trying to achieve?
The version known as a "line array" stands vertically but radiates horizontally, with very little vertical radiation. Useful to stop the people at the front from getting deafened while still covering the back of a hall.
Conventional wisdom isn't always right. Look at it from the POV of one speaker in series with an identical speaker both experiencing the same displacement. Do a spice simulation if you like. Google it too.
I tried "damping in series speakers" and the first hit was...
My whole point is they are not experienceing the same displacement.
Do a spice simulation if you like. Google it too.
I'm not wrong on this
He has made exactly the same wrong assumption as you have, and his test
- putting two completely separate loudspeakers in series, is not the same as putting owo driver units in a shared air space.
As I said, if you took a loudspeaker and notionally cut the coil into two pairs, but left them wound on the same former, then you have 'two coils in series' connected to a common cone and of course its trivial; to say that the single loudpseaker coil is in fact two coils in series, notionally.
That is what you and he do. In essence.
MY point is that that stops working the moment the two coils are NOT on the same mechanical former and connected to the same cone.
The possibility of differential movement exists, and the whole assumption you and he use to 'prove' the case is no longer valid.
Connecting te speakers in antiphase in series in an infinite baffle would be the most extreme example of 'these loudspeakers no longer move as a single unit with 'half the voltage' across one or the other at resonsance.
They're motors with back emf. If they're identical with identical loads, then in a series connection both will see the full damping of the amplifier output.
I think you're saying that they're not seeing identical loads, because nothing real can ever be identical. This is backtracking while dancing on the head of a pin.
Look at it this way. The speakers are equal, and working together for the common good, pushing and pulling in harmony. Socialism in action.
What is damping? At an electrical level it is the flow of ac current between amp & speaker. That current is identical in either case, thus the damping is identical in either case.
To say damping will be less because you have a series resistance is oversimplifying the situation, hence reaches the wrong answer.
they're identical, hence their variations from ideality are the same
If the position is symmetrical it won't, if not it may well do
because there's no reason to use parallel or series speakers for hifi or studio monitors.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.