Crossover Components Suggestions for SEAS Coax

[lucadelcarlo]

I'm a product designer first and foremost and have little experience with crossover design, so I'm looking to the expertise/knowledge of others. Using a SEAS L12 coaxial, I'm intending to put together an x-over which meets the following criteria:
1. Is rather compact, as it will mount within a small enclosure, a satellite speaker
2. Remains affordable (below 20 Euro/30 USD, preferably half that)

The prototype x-over we're using now is far too expensive, particularly the inductors and the larger caps.

Attached is a schematic from SEAS of their recommended design.
Tweeter
C1 4.700uF
C2 10.000uF
R1 2.700ohm (in series)
R2 10.000ohm (parallel)
L1 0.470mH

Woofer
L2 1.200mH
C3 33.000uF (parallel)

Any recommendations on where how to to cut costs without too great a sacrafice in performance? Hoping to source something soon. Thank you in advance.

[AllenB]

The woofer capacitor can be electrolytic, the tweeter inductor can be made with small gauge wire, and the resistors can be sand cast. If you want even cheaper, the woofer inductor doesn't have to have ultra low DCR and the tweeter caps can be polyethylene.

[lucadelcarlo]

Thanks Allen, that points me in the right direction. I may ring up Madisound, or a more local supplier in the UK, and ask for assistance in piecing together all parts for a given budget, but at the moment, it seems clearer where to make compromises first or where to do it last.

[PeteMcK]

perhaps (depending on driver characteristics & desired output levels), you could redesign the crossover as 1st order on the woofer, 2nd order on the tweeter, using only a series resistor for padding - result, 3 fewer parts.

[lucadelcarlo]

Pete,
At present, what order(s) is the design supplied by the manufacturer, SEAS? Reducing the complexity is of course sounds appealing and worth testing against the recommended x-over. What could the potential drawbacks to such a design be? Lower output from the woofer, and damage to the tweeter?

This fabric dome tweeter from SEAS seems to handle very high loads: the low and high ends of the x-over were crossed once and the tweeter subjected to the wrong source at high levels for more than a brief transient. Surprisingly, no damage occurred, so a less protective crossover design might be tolerated.

[PeteMcK]

after looking at the data sheet, I think this driver isn't a good choice for you, it demands a more complex crossover rather than a simpler one (Al cone). I think you'd get more affordable results designing a conventional 2 way with drivers that require less work in the crossover (paper or poly cones).

[5th element]

I wouldn't go changing the SEAS crossover at all, providing the phase integration between the two drivers is good (which it should be) then there's very little wrong with the design at all.

It looks like it has 3-4dB of baffle step compensation built in which is a useful amount. Both drivers need 4th order (acoustic) slopes to work well together. The metal mid-bass will have a peak in it's third harmonic @ 2khz, so a steep slope @ 1800hz is right on the money. This works the tweeter about as low as it can go too, hence again the requirement for the steep slope.

The crossover is about as simple as they get whilst still doing a really good job (a necessary job considering the drivers limitations). All we're looking at is a 2nd order electrical filter on the mid/bass and a 3rd order electrical filter + Lpad on the tweeter.

As has been said before, use a non polar electrolytic for the woofer cap and the cheapest coil you can get for the tweeter. You can get decent film caps for little money and if you really want to keep the cost down, use a non polar electrolytic in parallel with a 0.1-1uf film cap for the 10uf cap in the tweeter network. The resistors can be anything cheap that can handle a bit of power. The only component I would advocate spending more then a couple of pounds on would be the 1.2mH inductor on the mid/bass, get something with as low a DC resistance as your budget will allow.

Europe-audio has a good range of xover components at a wide range of prices. The non polar electrolytics they sell work well too. You could buy everything from Europe-Audio and save a packet on p&p from the states. I would recommend you contact EA an ask them if they are looking at getting the L12 in stock, or if they could get them in and how much they would cost.

[lucadelcarlo]

Pete, there may not be much to do in terms of simplifying the x-over design for this driver then, as the driver is fixed in this equation, and can't change. The enclosure is spherical and really designed around the driver. The x-overs have all been external to up to this stage. Now it's a design issue to come up with an elegant solution. Somehow.
Thank you for looking over the data sheet. What is it about the alu cone that requires a complex x-over? Odd peaks? (Ah, this was just answered in the post subsequent to yours).

[lucadelcarlo]

@5th Element:

I'll pursue this as you've suggested. The thread has been quite useful so far, so many thanks all.