Filter for bi amping

[saevox]

I have a set of JBL 3678 speakers and wanted to run them bi amped, what slope/filter do I apply and what is the cross over point? I beleive for this model it is ~1200..???...Also how does one find this information for various JBl applications that can be passive or active like the 3678? Will be using a Mini DSP for the crossover which can apply:

Linkwitz Riley (LR): Available in 12, 24, 48dB/oct
Butterworth (BW): Available in 6, 12, 18, 24, 30, 36, 42, 48dB/octBessel: Available in 12dB/oct


Thanks in advance for any information :)

[Fort Knox]

I got a pair ...they have passive nets built-in ,....which I prefer
bi-amp stuff is noisy...for home..... but..suitable for noisy gigs

[hjames]

I got a pair ...they have passive nets built-in ,....which I prefer
bi-amp stuff is noisy...for home..... but..suitable for noisy gigs


Depends on your gear, but I know of 3 folks who've been running big 4 way JBLs in biamp mode with no problems,
and certainly with no noise issues.

I ran my biamped 4341s for a few years with no real problems,
I used Ashly XR1001 electronic xover and a pair of Adcom GFA-555s.

BMWCCA is running his 4345 system in biamp mode ...
also using Ashly crossover but I believe he is using crown amps.

Dkalsi is running his 4345s in biamp mode with some high end amps and, I think, a Marchand xover ...

I know there are others, thats just a few I can think of quick off the top of my head ...

[saevox]

Yea, the noise issue struck me as off, I would think it was an equipment issue. But again I am looking to find out what the factory JBL spec is on the cross over db/oct and what filter is used. All the threads I find on this are more about preference and opions on how each setup their own speakers. I was looking for how to find what JBL specs and them maybe play around from there. I

In case anyone is curious or does not want to look the 3678 is a 2226h woofer with a 2425h compression driver on a 2342 horn, the newer models have the 2426h compression driver...but basically the same thing.

Thanks again.

[Earl K]

From Google ! :dont-know:

3678LF with network ! (http://www.jblproservice.com/pdf/Theatre Series/3678LF.pdf)

Network schematic reads as a 3/3 ( 18db/octave over 18db/octave ) with a nominal 1K Crossover Point.

A Butterworth filter-type looks to be your only option ( don't forget to add a 1200hz, notch filter to simulate C4,L2 & R3 / first try a Q setting of 1.5 octave wide & that is 3db deep ) . Measure & then re-tweak the notch !

:p

[saevox]

From Google ! :dont-know:

3678LF with network ! (http://www.jblproservice.com/pdf/Theatre Series/3678LF.pdf)

Network schematic reads as a 3/3 ( 18db/octave over 18db/octave ) with a nominal 1K Crossover Point.

A Butterworth filter-type looks to be your only option ( don't forget to add a 1200hz, notch filter to simulate C4,L2 & R3 / first try a Q setting of 1.5 octave wide & that is 3db deep ) . Measure & then re-tweak the notch !

:p

Wow Earl! thank you so much, I do not totally understand all that you said, but it is exactly what I was looking for. My technical expertise on these thing is well....Lacking.
:)

[4313B]

I got a pair ...they have passive nets built-in ,....which I prefer
bi-amp stuff is noisy...for home..... but..suitable for noisy gigsWithout the padding afforded by passive filters they can be a bit noisy. They're quite efficient.

[Earl K]

Wow Earl! thank you so much, I do not totally understand all that you said, but it is exactly what I was looking for. My technical expertise on these thing is well....Lacking.
:)

You're welcome saevox / the problem now is that if you build a straight 18db hipass filter ( within your minidsp thingy ) the results will really sound like crap .

You need to copy the voltage drive ( of the horn circuit schematic posted above ) to get something listenable and HiFi-ish .
- Since you've already tipped your hand ( that this is mostly beyond your present capabilities ) , I'm left wondering what is best way for you to proceed .

- One thing would be to still bi-amp, but build the passive portion of the network for the horn and use it bi-amped in series with the horn driver . You can still drive it with the minidsp with some sort of unity gain circuit .
- IME, you'll want to add a ( 3 to 6 db ) PEQ boost filter centered around 14-15K, to pop up the UHF ( which will unfortunately be already sagging ( after @ 10 or 12K ) / this opinion is based on having had another look at the posted schematic / it's a classic middle-years SR-PA circuit that I use in a similar way ) .

- Another idea is to build the passive HF portion of the 4425 / since it's a more HiFi circuit than the 3678s ( using the same driver & horn I think ) .
- Once again biamp the horn with the passive HF portion inline with the driver .
- Here's the network ( N3124 ) (http://www.jblproservice.com/pdf/Network Schematics/3124 Network.pdf)

:)

And again ;

[4313B]

3678 voltage drive ( 8 ohm dummy loads )

The notch filter across the 2425 is centered at 1.2 kHz and operates from 1.1 kHz to 1.3 kHz.

[4313B]

With the minidsp you might want to set up a 12 dB/octave LR at 4 kHz to get the curve shown below ( high pass red curve). I don't have the minidsp software on this machine otherwise I'd just build the transform for you to cut and paste into your software.

This might get you in the ball park but it is derived from an 8 ohm dummy load rather than the actual 2425/2344 load. Some tweaking might be required. And there might still be an issue with noise.

[Earl K]

That's some nice work there 4313B. :)

I'm sure saevox will appreciate this .

<> EarlK

[saevox]

All of the information has been very helpful. This is my first active attempt and it seems it is a bit more involved than I thought. The 3678 that I have do have the passive networks in them with the option to run them active so I thought I would experiment with that. The other reason being that I only have 2 of the speakers and was able to get them for a steal. I plan to eventually build a third for LCR and was hoping to run all 3 active in order to save having to build the xover for a passive one....but that is working out to be a bit more involved than I thought, but I will still try it out as the mini dsp is only 99- and I still have the passive network if need be.

I did look into the 4425, the compression driver on that on is the 2416 for the HF with the same horn. I am not too sure how similar that is to the 2425 or if that would even be a possibility to use that HF network.

Also on a side note I did want to test my compression drivers to make sure they are still in good shape, is there any simple test that can be done to make sure the driver and the diaphragm are in good shape without disassembling as I have read the tolerances are quite tight and the driver can be damaged by someone with not so much experience.

Also what is the deal with the voltage drive piece and how does that get worked into this?

[saevox]

Thanks 4313B, you posted while I was typing up my long winded peice there....

[4313B]

Another idea is to build the passive HF portion of the 4425 / since it's a more HiFi circuit than the 3678s ( using the same driver & horn I think ) .
- Once again biamp the horn with the passive HF portion inline with the driver .
- Here's the network ( N3124 ) (http://www.jblproservice.com/pdf/Network Schematics/3124 Network.pdf)

:)

And again ;That could be very interesting as well.

[Mr. Widget]

All of the information has been very helpful. This is my first active attempt and it seems it is a bit more involved than I thought.

Also on a side note I did want to test my compression drivers to make sure they are still in good shape, is there any simple test that can be done to make sure the driver and the diaphragm are in good shape without disassembling as I have read the tolerances are quite tight and the driver can be damaged by someone with not so much experience.

Also what is the deal with the voltage drive piece and how does that get worked into this?

Yes, to do this right it is far more complicated than you might think.

You probably should not open your drivers, the likelihood of them being out of spec is high, but looking at them won't tell you much. The only practical tests require knowledge and specialized equipment... you can measure gross problems with an ohm meter, but even that really tells you very little.

Probably your best test is listening to them.


Widget

[wrager]

I just converted to active bi-amping this weekend with my 4622's. I use three Crown DSi-1000's with the pre-set network info. They are dead silent in operation. The dynamics are incerdible.

[saevox]

Great, thanks for the info on the testing, looks like I will just have to try listening to the compression drivers.

That is good to hear about the 4622's, what active crossover are you using? Also is the mini dsp a good product for this or is there another route I should go if I attempt active? (keeping a budget in mind)

I will be ordering my 2342 horns this week in order to get the top ends on these 3678 back together...going to be 200 from Jammin unless anyone knows another source, also looking for 2425h or 2426h compression drivers as mine are "j" if anyone has some lying around PM me and we can work something out. But I can not complain, 2 3678 bottom ends and 6 8330's for 300 beans is a great deal :).