PDA

View Full Version : JBL L300 / 4333 3 way analog electronic crossover question



gwho
12-27-2020, 09:59 AM
There is another active thread on using a JBL L300 with an electronic crossover but I want to confirm a couple of things that are unclear to me.
I've finally scrounged the parts to build my non-original 4333 with some utility cabs I purchased a while ago.

I will be using:
2405
2311/2441/2445 diaphragms
2235
Marchand 3 way XM9 crossover 24db/octave
DC blocking cap / 20 ohm shunt resistor across 2405 and 2441
No L-Pads (will control volume in each speaker from either crossover or amplifiers

Just want to verify before hooking up
1. All the above drivers have black as their + terminal
2. The 2441 signal should be inverted compared to the 2405 and 2235 (as shown in the Nelson Pass schematic).

I appreciate any help.
Best Regards,
gwho

Mr. Widget
12-27-2020, 10:20 AM
Sounds like a cool project!

Question: What are the 20 ohm resistors for?

A thought about polarity: Some amps are inverting and others are noninverting. Recently I was helping a friend dial in his tri-amped system and I noticed a massive suck out at one of the crossover frequencies. This was because one of his amps was inverting the phase. Finding this kind of thing as well as balancing your system are best done with some form of measurement system.


Widget

Riley Casey
12-27-2020, 10:36 AM
The original crossover topology in the 4333 was not 24db per octave 4th order filters . The Nelson Pass crossover schematic shows 2nd order for the low and mid and 3rd order for the high. The summing will be substantially different with 4th order filters. You will likely have to determine the best polarity empirically by listening or ideally using test software like Smaart, Foo or REW.




...

Just want to verify before hooking up
1. All the above drivers have black as their + terminal
2. The 2441 signal should be inverted compared to the 2405 and 2235 (as shown in the Nelson Pass schematic).

I appreciate any help.
Best Regards,
gwho

gwho
12-27-2020, 11:27 AM
Riley / Widget,

Thank you for the responses and encouragement. I will definitely try to take some measurements after putting them together. Just want to get an idea of a starting point.
I will now assume that I will try it both ways and determine what works best. I've attached a couple pictures of the boxes. They are not perfect but good enough.

8791987920

The 20 ohms resistors discussed in one of the threads I read for electronic crossovers where they recommended a DC blocking cap and a shunt resistor across the CD inputs (is this wrong?).

Regards,
Gary

Mr. Widget
12-27-2020, 11:33 AM
The 20 ohms resistors discussed in one of the threads I read for electronic crossovers where they recommended a DC blocking cap and a shunt resistor across the CD inputs (is this wrong?).Not sure why you would need the resistors, I have never used them. I use a very large value cap in series with my compression drivers as protection, but thats it. I am a fan of less is more.

The cabinets look fine. You can even veneer them at some point if you want.


Widget

gwho
12-27-2020, 11:46 AM
I won't put in the resistors. I agree less is better.

Regards,
gwho

macaroonie
12-27-2020, 04:55 PM
In my experince it's fairly apparent which way is correct. It might take a little while as you get levels adjusted etc but it only takes a minute or so to switch phase on an amp. you will have a eureka moment.:)

Mr. Widget
12-27-2020, 06:05 PM
In my experince it's fairly apparent which way is correct. It might take a little while as you get levels adjusted etc but it only takes a minute or so to switch phase on an amp. you will have a eureka moment.:)Maybe for a trained/experienced listener like yourself. I have heard so many DIY setups that were so very wrong. Basic measurements make it a lot easier to get in the ballpark. Once there, subtle tweaking to taste can be very rewarding.


Widget

gwho
12-28-2020, 08:32 AM
At a minimum I can take a measurement with a DEQ2496 and pink noise to determine what the response of the speaker is. I will likely also try to learn how to use REW if time permits and take some measurements.

Best Regards,
gwho

Riley Casey
12-28-2020, 01:34 PM
REW is a pretty remarkable piece of software particularly for free. I've used measurement software since the original Crown TEF machines and I have SpectraFoo and Smaart licenses and I still use REW for some things. Well worth the time and effort to learn. The RTA in the DEQ ( well any RTA really ) isn't worth much in assessing speaker performance although it is handy for ear training.


At a minimum I can take a measurement with a DEQ2496 and pink noise to determine what the response of the speaker is. I will likely also try to learn how to use REW if time permits and take some measurements.

Best Regards,
gwho

gwho
12-28-2020, 05:38 PM
Riley,

I will try to do some reading on REW. Using the Deq2496 relates to my original question about inverting the signal to the mid-horn. The answer was try it both ways. I assume the Deq2496 could show me which connection to the mid-horn provides the flattest response if it was not obvious.

Regards
Gary

Riley Casey
12-28-2020, 07:33 PM
The thing about polarity is that neither could be right. What you want to know is the phase relationship between the drivers. The voice coils in a 4333 are on three different horizontal planes. Something that JBL got 'fairly' close with the phase relationships inherent in the crossover filters they used. Each 6db per octave pole of a crossover filter adds 90 degrees of phase delay to the signal. Your DC blocking caps ( possibly not needed. Depends on how much you trust the output relays in your amps ) each add that same 90 degrees of phase shift. You're using different filters and thus have to come up with a different solution. Perhaps not one available in an analog crossover fixed at 4th order filters. REW will show you the phase ( essentially when the sound arrives at any given frequency ) as well as the frequency response of the speaker system. You can almost certainly get your speakers to sound OK, even pleasant but without time aware measurement it's a crap shoot.

Mr. Widget
12-28-2020, 09:16 PM
The thing about polarity is that neither could be right.True.

Depending on the crossover type, (number of poles and characteristics at the knee) the polarity may make a larger or smaller impact on the overall frequency response. The phase will most certainly be wrong regardless, but we try to make it less wrong. :D

Basically, you need to measure, listen, tweak, and listen again until you have the best compromise with the system you are working on. Kind of like making a nice soup. You add a bit of this, you add a bit of that, and with luck it is a masterpiece and not just salty water.


Widget

macaroonie
12-29-2020, 08:13 AM
Might be an idea to move up a notch in the Behringer stable and get the DCX , then you get the helpful option of delay , parametric EQ and variable filters.
I understand completely your desire to use the Marchand BTW.

One other point , my system is very similar and it didn't take long to fathom out that the 2235 is not a happy bunny up at 800Hz.
I eventually landed on just over 600Hz , I'm using 2380 horn and thats about as low as that goes.
The mass ring in the 2235 smudges the mids making the sound dark , call it what you will.
I got a very worthwhile improvement in clarity , defenition and imaging.

Just my 0.2c

Mr. Widget
12-29-2020, 08:37 AM
One other point , my system is very similar and it didn't take long to fathom out that the 2235 is not a happy bunny up at 800Hz.
I eventually landed on just over 600Hz , I'm using 2380 horn and thats about as low as that goes.
The mass ring in the 2235 smudges the mids making the sound dark , call it what you will.
I got a very worthwhile improvement in clarity , defenition and imaging.

Just my 0.2cI agree with your analysis. I would suggest moving to the 2216 woofer. I think it is better than the 2235H in every way. With the 2216, you can keep the 800Hz crossover and stay with the 2312. In an active system, making the swap should be fairly straightforward.

While the digital crossover has a lot of powerful and beneficial features, I would still stay analog, but then I’m an audio dinosaur.


Widget

gwho
12-29-2020, 08:47 AM
Riley / Widget / Macaroonie

Thank you for the concise explanation. I now understand that there is more to using a crossover than meets the eye.
I may look into digital in the future for a crossover but I really like the DAC in my preamp now (ATC CDA2 Preamp/CD/Dac) so I will try my best to get it dialed in with the Marchand.

Riley - The tools you disclosed (spectrafoo or smaart). Is either very easy and intuitive to use for dialing in an electronic crossover (digital or analog). I would entertain getting a license
for one if it was easy to use as I am never going to become an expert in this but may use electronic crossovers in other speakers in the future.

Regards,
gwho

Riley Casey
12-29-2020, 10:48 AM
On the digital vs analog crossover question I am a big proponent of using DSP crossovers for discovering the best crossover characteristics for any speaker system regardless of the eventual solution being passive networks after the amplifier or analog crossovers before the amplifiers. Very flexible digital crossovers ( in the sense of having multiple filter slope selections ) are very cheap on Ebay now that the virus has killed the live sound industry top to bottom. Hate to be ghoulish about that but buying someones DBX or BSS crossover used may put breakfast on some poor sound engineers table.

I'm not familiar with the Marchand crossover but I'm a big fan of the old analog BSS FDS360 electronic crossovers. Most were used in live sound applications and would need a thorough rebuild at this point but they offer much of the flexibility of DSPs while being analog.

On acoustic analysis software Spectrafoo was originally designed for use in recording studios for signal analysis and leans heavily in that direction. It does have the core acoustic anaylsis tool for transfer function but is a bit less malleable in that area. Smaart is the industry standard for sound system measurement and has a large user base of live sound and installed sound practitioners. It actually allows measurement with program material ( if you like heavy metal guitar anyway ) Its fairly pricey for software to be used only a few times a year and often requires for fee training for effective use. As I mentioned I'm pretty impressed with REW for home use and it's certainly effective for evaluating crossover characteristics. REW is free and I'm fairly sure Foo and Smaart have free demo downloads. All of them are easier to use than the TEF machine was.:blink:


Riley / Widget / Macaroonie

Thank you for the concise explanation. I now understand that there is more to using a crossover than meets the eye.
I may look into digital in the future for a crossover but I really like the DAC in my preamp now (ATC CDA2 Preamp/CD/Dac) so I will try my best to get it dialed in with the Marchand.

Riley - The tools you disclosed (spectrafoo or smaart). Is either very easy and intuitive to use for dialing in an electronic crossover (digital or analog). I would entertain getting a license
for one if it was easy to use as I am never going to become an expert in this but may use electronic crossovers in other speakers in the future.

Regards,
gwho

Mr. Widget
12-29-2020, 11:36 AM
Riley - The tools you disclosed (spectrafoo or smaart). Is either very easy and intuitive to use for dialing in an electronic crossover (digital or analog). I would entertain getting a license
for one if it was easy to use as I am never going to become an expert in this but may use electronic crossovers in other speakers in the future.
These are all professional tools and are designed to be used by someone with extensive knowledge of sound system design/operation. As Riley points out the different systems are aimed at slightly different markets and applications. MLSSA, CLIO, and the now defunct LEAP are more for loudspeaker design and may be more appropriate for your application, but they are/were also costly and not intuitive to someone who isn't deeply involved in audio.

REW is likely your best bet... and a simple RTA while ultimately not that useful is not a terrible starting point. Most of us probably got their start with one. I know I did.


Widget

macaroonie
12-29-2020, 12:59 PM
Don't the current DBX DSP's have a set up wizard in them ?

Riley Casey
12-29-2020, 01:17 PM
If it does that's a good reason not to buy one right there.


Don't the current DBX DSP's have a set up wizard in them ?

gwho
12-30-2020, 10:34 AM
Is it possible to get the 2216 woofer? I assume they are probably very expensive.

Regards,
gwho

Mr. Widget
12-30-2020, 11:07 AM
Is it possible to get the 2216 woofer? I assume they are probably very expensive.

Regards,
gwhoThere are two versions of this driver. I have not played with them myself so I can not say whether the 2216ND or the 2216ND-1 will be better in your application, but they are available and while on the expensive side, they are not refi expensive.

https://reconingspeakers.com/product/jbl-2216nd-woofer-for-jbl-m2-5041785/

https://reconingspeakers.com/product/jbl-2216nd-1-woofer-for-4367-320-0045-002/


Widget

gwho
12-31-2020, 08:20 AM
Widget,

Thank you for that information. I will consider it in the future. I was contemplating putting a 10" woofer on top of the box if I did not like the sound of the 4333 or just to try it (I know another crossover). I was able to pick up a couple of 2251J very inexpensively so they were my back up plan.

Happy New Year
gwho

Riley Casey
12-31-2020, 11:46 AM
The whole 15" woofer crossing directly into a compression driver type of speaker design is a very 1950s way of approaching a world of 15 watt tube amps. As transistor amps became more powerful three way designs with cone driver mids crossing into compression drivers above 1k hz made for much better sounding speakers. Putting a ten between your 15 and your one inch will likely make for a much nicer sounding box. The ten can be in a small sealed box and crossed into at 400 hz. This among other things means that 40 hz movement of the woofer cone aren't modulating the 1200 hz tones from the same driver.


Widget,

Thank you for that information. I will consider it in the future. I was contemplating putting a 10" woofer on top of the box if I did not like the sound of the 4333 or just to try it (I know another crossover). I was able to pick up a couple of 2251J very inexpensively so they were my back up plan.

Happy New Year
gwho

gwho
12-31-2020, 03:31 PM
Riley,

Thank you for that information. I will definitely try that at some point. From reading in the forum it appeared that there were varying opinions on where you could cross (800hz - 1.2Khz) the 2311/2441 to the 2235. Some said you can cross the 2441 at a lower frequency because of the larger diaphragm. Other said you couldn't because of the 2311 horn. I never could find a definitive answer which is why I purchased the JBL 10" speakers just in case.

Regards,
Gary