JBL 4732 triamp - crossovers?

[Magnus Eriksson]

Just wondering, is anyone sitting on more specific crossover details for this speaker? Frequency, slope, delay, gain etc.
These speakers were inherited for a charity project (saving an old cinema) together with a QSC DCM-30 and the existing crossover configuration is, well, I guess they needed A LOT of EQ in the sound processor somewhere else

/M.E.

[Steve Guttag]

The 4632 and 4732 do indeed have response issues. Beware however, I've found the canned parameters in the DCM line to not always be good. They used what they were "supplied" or used the crossover points via the datasheets...which didn't include any other voicing.

I've tried using numerous devices trying to get them to sound, what I consider, good and never did. I used Crown's PIPs (back in the 4632 days), analog crossovers, DSP crossovers and it was always a fight and the net result was a nasily sounding speaker.

So,

Here is what JBL thinks is the proper DSP configurations are:

For the LF section:

• HP at 30Hz with a 24dB/Octave Butterworth filter, LP at 200Hz with a 24dB/Oct Butterworth filter.
• PEQ Filters at:
  • 42Hz, 5 dB, Q=3.5
  • 60Hz, 2.5 dB, Q=2
  • 120Hz, 1 dB, Q=4
  • 150Hz, -1.5 dB, Q=4
• Delay: 1.4ms

MF Section:

• HPF 275Hz with a 24dB/Oct Butterworth filter, LPF 1.2KHz with a 48dB/Oct L-R filter, gain = -4dB
• Delay = 550us
• Shelving filter at 900Hz with 7dB of boost with a Q of 9 (.16Oct).
• PEQs at:
  • 325Hz, 1.5 dB, Q=3
  • 500Hz, -1.5 dB, Q=5
  • 615Hz, -1.5 dB, Q=4
  • 900Hz, -2.0 dB, Q=4
  • 1.1KHz, 1.0 dB, Q=4

HF Section:

• HPF 1.1KHz with a 48dB/Oct L-R filter, No LPF, gain = -11.5dB
• No Delay
• Shelving filter at 5.5KHz, 13dB, Q=6 (.24octave).
• PEQs at:
  • 1.6KHz, -0.5 dB, Q=2
  • 3.75KHz, -2 dB, Q=2
  • 5.75KHz, 1.5 dB, Q=4
  • 8.5KHz, -0.5dB, Q=2.5
  • 11.75KHz, 4.0 dB, Q=7 (the driver/horn combination just can reproduce this, I've found so one boosts it and it doesn't sound good)
  • 17KHz, 4.5dB, Q=3 (really? Who is listening to this through a vinyl screen? They're trying to get it to have a response out to 16KHz).

Remember, DSP processors do not all function alike so what one manufacturer has to do to get a response may not match another but this is what JBL came up with using their in-house Crown products. So, before you put your mics out...how many filters did they have to throw at this thing? 14 PEQs, plus shelving filters. You don't have that in a DCM-30 so choices will need to be made!

Here is the tuning for the 4632T (its predecessor...primarily, it is the LF drivers that changed, as I recall). It has filters that are more akin to the DCM series:

DSC260_4632_Tuning.png

To a degree, you can put the crossover points in...look at the response on the three speakers. Look for commonalities and go the worst aspects that are common to all three within the crossover and leave the little details to final EQ.

Good luck.

[Magnus Eriksson]

Thanks Steve.
I first tried a clean start with the "canned" 4632 parameters, but that was even worse than what was used before.
Then I got some other parameters from a colleague in Norway and got closer to something to work on.
But they are indeed difficult speakers.
/M.E.

[Ioannis Syrogiannis]

I suppose that your Norwegian friend would prefer the other parameters to be kept from public?
If not, maybe they can help others in the future, who will find themselves in the same situation...

[Magnus Eriksson]

They are more or less identical to what Steve posted when I looked closer and compared.