Horn LF cut-off question

[stevem]

Does the slope of the crossover have anything to do with the useable lower frequency for a CD horn? For instance, I have the 2332 horns, which are always used in systems with a 1 KHz crossover, I think usually at 12 db/octave. Can the crossover frequency be lowered if 48 or 52 db/octave slopes are used? If so, how low would be OK? Assume that the compression driver would not have any problems down to 500Hz.

[4313B]

Good questions! Let's see what kinds of answers we get.

Hint - let's see some of those Horn/CD frequency response and distortion curves.

[Earl K]

Hi

For SR Work : ( for low-powered home HiFi ,,,, ??? )

- I certainly wouldn't crossover any lower ( no matter what slope is used ) than the lowest load limit of the horn under consideration .

The Limit ?
(a) Look at the drivers plane wave response ( to get an idea of the drivers' extension & performance in the lower octaves when fully loaded ) .

(b) Run some response studies of the horn/driver combo with milliwatts of input ( & gently crossed over 1.5 to 2 octaves below what you think is the normal crossover point ). The lower frequency load limit of the horn should be quite apparent by the 3 db down point of the combos' low end respone .

(c) Run an AC impedance study of the horn/driver combo . Compare it to the first significant null ( or trough, > high side of drivers' resonance ). This first null should closely track the 3 db down point found in the FR study . ( Zilch has inadvertently published most of the necessary graphs needed to make this correlation. ) Also look at Widgets' impedance studies for his radial-type H4003 . .

[Zilch]

1) Published FR and distortion curves on a different horn. No PWT data available.
2) Driver only impedance.
3) Driver and throat adapter impedance.
4) Driver, throat adapter and subject horn impedance.
5) FR on subject horn w/CD compensation (2.5 dB per dotted line).

And the verdict is???

"Zilchster, you can successfully play this combo with a HP crossover frequency of _____ Hz."

[As long as it sounds good down there.... ]

[Earl K]

Hi Zilch

- Thanks for the visual aids !

- The 3rd harmonic distortion spec. really underscores why this (?) horn/driver combo can't be crossed over any lower than 1000 hz ( whether or not a 100 db per octave slope is present ).



EDIT: The sustained weakness in electromotive force that manifests itself below @ 1.8K should be audible enough to warrant a move to a higher crossover point (such as 2 K ).

[Zilch]

Yup, manufacturer's recommended crossover frequency is 1 kHz.

While I have SOME hope (albeit minimal) CLIO will reveal that the SPL deficit below 1.8 kHz is room cancellation, the published response curve above says it's real.

The distortion specs indicate that the driver could tolerate some EQ down there, and I'll try that.

An interesting Geddes paper in the current AES Journal postulates there's more latitude with distortion available than we traditionally suppose, at least insofar as listening performance is concerned.

Not sure I like the experimental design, tho. Talking Heads was the test program.

Geddes notwithstanding, I don't think I'd be boostin' at 800 Hz....