Just a question.  We've got some JBL SRX series 15" boxes and double 12" boxes.  These can sound really nice bi-amped however when run through the passive crossover inside them they spew out way to much high end from the compression driver/horn.

 

I've been dealing with it by eq'ing the sh*te out of the boxes, pulling everything after the 1.2Khz crossover point back by 6-12db most times.

 

With the EQ applied they sound really nice again.

 

I did start thinking about better solutions and was wondering if there was something I could put in series with the compression driver inside the box that would lower the signal going to it?  A resistor or something?  I'm no expert about things like this but would be interested if anyone had suggestions.

 

I can deal with it as is with the EQ but would be nice not to have to think about it.

 

rez.

 

http://www.lalena.com/Audio/Calculator/LPad/ - http://www.lalena.com/Audio/Calculator/LPad/

- Many say that JBLs sound bright.

Im not sure if this is due to the program material,or an edgy titchy compression driver with strange resonances on it. DJK might have some input.

 

 I dont think JBL would purposely emphasise anything - HIFI speakers do this enough.

 

SRX715 and SRX722.

 

I'm sure JBL are doing everything ok... this is probably a matter of my personal taste I don't know but for me the highs are too loud.

Loudspeaker enclosure 'diffraction loss' occurs in the low frequency range of loudspeakers in enclosures that are located in the open, away from walls or other surfaces. The essence of it is this: a speaker radiating into half space plays 6 dB louder than the same speaker radiating into full space. A full range speaker finds itself radiating into half space that the upper frequencies but radiating into full space at lower frequencies. This results in a gradual shift of -6dB from the highs to the lows. This is called the '6 dB baffle step' or the enclosure's 'diffraction loss'. The center frequency of the transition is dependent on the dimensions of the baffle. The smaller the baffle the higher the transition frequency. 

All enclosure shapes exhibit a basic 6 dB transition or step in the response with the bass ending up 6 dB below the treble.

 

The -3dB point for the 722 is about 300hz, you will need about 3dB of boost there and about 6dB of boost in the 80hz region, and about 1dB of boost around 600hz. The step points for the 715 will be roughly 250hz, 63hz, and 500hz.

 

The JBLs are flat to very high frequencies. Close mic techniques sound un-naturally bright on flat systems. Sound at a distance does not have as much HF content as sound up close does. It is recommended you follow a 'house' curve and do a 3dB/oct roll-off starting about 3Khz.

 

These two combined will give a shift of almost 12dB between 100hz and 10Khz and will probably sound close to what your ear thinks is 'right'.

 

I'll give what you suggest a try... although the purpose of this was to run them without any EQ but hey.

 

Thanks again!!!

 

rez.

 

and p.s. jokes aside HansA it's not just my hearing trust me stand in front of these bad boys you'll be reaching for that EQ in no time!!!

Look in the eminence book for l pad resistor values.

hope that helps

chris

 

Didn't look at the schematic for the networks, did you?

http://www.jblproservice.com/pdf/SRX700%20Series/SRX715F.pdf - http://www.jblproservice.com/pdf/SRX700%20Series/SRX715F.pdf

 

Im certainly no expert but I dont understand why its not possible to put an L pad at j2, 1+2, maintaining the same impedence seen by the crossover but diverting some of the power from the driver

 

Id be grateful for an explanation djk

It will however have the following effects:

1) it will increase the apparent power handling of the comp driver as some power is burnt in the l pad

2) It will effect the spectral balance of the speaker as a whole.

 

Now, the increase in power handling is an advantage, but the effect on spectral balance ( sound)  is debatable. If your preference is for less treble, then i suggest you go for it - at worst it will cost you a few quid in resistors and a bit of time.

 

You will do no damage to the speaker providing you do the job properly.

 

I have designed many many passive filter networks over the years and can confirm that every passive network can be modified in this way as the nature of the l pad, keeping the load impedence the same from the viewpoint of the crossover network in essence makes the l pad invisible to the crossover. Electrically, the crossover simply doesent know it is there. Anybody who says that you cant add an l pads to this network is incorrect.

 

let me know how you get on

 

chris

 

I just have seen the post above mine.

 

There is no such thing as a constant resistance driver, therefore the crossover can not be designed that way.

 

Drivers only have a constant resistance at DC ( 0Hz), which also happens to be very effective at cooking voice coils. Drivers are rated with a nominal impedence ( impedence being dependant on frequency and effected by the inductance of the voice coil)

 

Of course the connection dot drawn linking L+ and IN- is an error.

 

The notch formed by the 16µF and 0.20mH is at about 2.8Khz, and the notch formed by the 3µF and 0.3mH is at about 5.3Khz

 

A horn woks by adding radiation resistance to the resistance of the driver by itself. With a real horn (finite mouth) we will see a series of cyclical peaks in the impedance due to reflections from the mouth. Corresponding to these impedance peaks we will also see peaks in the acoustic output of the horn, most noticable above the 1/2W point (the finite mouth generally masks the lower frequency peaks).

 

JBL does not show the impedance of the horns and drivers used in the speakers in question here, but they do show the 2451H (used in the earlier version of the SRX722), but on a bigger horn. The bigger horn will have its peaks at different frequencies, but will serve as an example. The second peak is in the middle of the operating point of the horn (2352 in this example), and the impedance rises from about 12R either side of the peak to about 30R right at the peak (the plot is actually of a J, not an H as specified in the graph). If you were to add an L-pad with some attenuation the impedance peak will be swamped by the parallel resistor in the L-pad. This will now cause excess attenuation in the notch filters and interfere with the crossover point too.

 

At a bare minimum we will have to adjust the values of the notch attenuation resistors (this will change with the setting of the L-pad), and we will also have to increase the value of the 2.2µF in the 18dB filter (this will also change with the setting of the L-pad). With the L-pad set to about 6dB of loss "pulling everything after the 1.2Khz crossover point back by 6-12db" you will probably need to increase the 2.2µF to 5.6µF (or more). Keep in mind that the textbook 18dB crossover for an 8R resistor would have 10µF, 0.75mH, and 30µF for 1.3Khz (the SRX722 nominal crossover frequency is 1.2Khz).

 

 

Also consider the 0.3mH in series with the 5.6µF, what is it actually doing? It has a pole zero at about 3.9Khz and about 30R of reactance at 16Khz. It's difficult to calculate because the reactive component of the load (driver + horn) is not known. If it was about 30R as shown on the data sheet for the 2451H (which is not correct, and not our horn), and purely resistive (which it is not), then this inductor would only cause about a 3dB loss at 16Khz. We do know that if we use an 8R L-pad set to 6dB of attenuation that the driver which is about 6R mid-band and around 15R in the high end will drop to 9.5R, so the 0.3mH inductor will cause the high end to roll off even more than the design calls for.

 

 

ok so will just any sock in the horn do the trick or what???

 

(interesting thread though thanks for everyones input learning a little although it is above my head...)