Your opinion on 4333 crossover please

[pauly]

Hi all,

I'd really appreciate some informed comment on the issue of crossovers for 4333 monitors.

I have speakers configured (physically) as 4333 which have. 2234 woofer, 2420 mid with horn and lense, and slots. Currently they have 3110a crossovers.

I want to build new crossovers and have purchased the components to build Nelson pass' version of the l300 crossovers.
-however-
I was browsing this site and saw an equivilent circuit to the 4333 genuine crossovers that didn't use the tapped inductor.

Now I have a quandary.The speakers are used to mix music sometimes, and enjoy music ...let's say 40/60%. I use a set of questeds for critical mixing.

So...I'd really appreciate hearing some informed comment on each of these crossovers, and recommendations on which way to go, particularly from those that have heard or built them.

Thanks
Pauly

[audiomagnate]

Hi all,

I'd really appreciate some informed comment on the issue of crossovers for 4333 monitors.

I have speakers configured (physically) as 4333 which have. 2234 woofer, 2420 mid with horn and lense, and slots. Currently they have 3110a crossovers.

I want to build new crossovers and have purchased the components to build Nelson pass' version of the l300 crossovers.
-however-
I was browsing this site and saw an equivilent circuit to the 4333 genuine crossovers that didn't use the tapped inductor.

Now I have a quandary.The speakers are used to mix music sometimes, and enjoy music ...let's say 40/60%. I use a set of questeds for critical mixing.

So...I'd really appreciate hearing some informed comment on each of these crossovers, and recommendations on which way to go, particularly from those that have heard or built them.

Thanks


Pauly

If you want to go with a passive crossover, you can't go wrong with Mr. Pass' solution for these three drivers. The L300 isn't acoustically identical to the 4333, but it should be close enough to do what you are asking it to do. In Nelson's ears I trust.

[pauly]

Thanks audiomagnate,

does anyone else have an opinion or advice?

thanks
pauly



If you want to go with a passive crossover, you can't go wrong with Mr. Pass' solution for these three drivers. The L300 isn't acoustically identical to the 4333, but it should be close enough to do what you are asking it to do. In Nelson's ears I trust.

[ivica]

Thanks audiomagnate,

does anyone else have an opinion or advice?

thanks
pauly

Hi pauly,

If You can upgrade 2420 & 2312 with 2450 (2441...) & 2311 it would be less 'hhssss' pronounced, and then adjust the network. I do not think that 6dB/oct network for 2405 would be the best solution. May be low-pass for VHF and hi-pass for UHF applied on 3155 would be more applicable solution.
You can find auto-transformer-less design here in the AH Forum.

regards
ivica

[audiomagnate]

Hi pauly,

If You can upgrade 2420 & 2312 with 2450 (2441...) & 2311 it would be less 'hhssss' pronounced, and then adjust the network. I do not think that 6dB/oct network for 2405 would be the best solution. May be low-pass for VHF and hi-pass for UHF applied on 3155 would be more applicable solution.
You can find auto-transformer-less design here in the AH Forum.

regards
ivica

I personally wonder why anyone would even want to go to the effort and expense of building a "high end" passive crossover in 2015. It makes absolutely no sense to me. It's like trying to build a modern racing using carburetors. To each his own I guess.

[4313B]

I personally wonder why anyone would even want to go to the effort and expense of building a "high end" passive crossover in 2015. It makes absolutely no sense to me. It's like trying to build a modern racing using carburetors. To each his own I guess.I guess they go hand in hand with those silly slabs of grooved vinyl... :rolleyes:

[ivica]

I personally wonder why anyone would even want to go to the effort and expense of building a "high end" passive crossover in 2015. It makes absolutely no sense to me. It's like trying to build a modern racing using carburetors. To each his own I guess.

Hi audiomagnate,

But using single (stereo) amplifier with built in preamplifier is much easier then using pre-amp, DSP driven crossover, and several (2 or 3 or 4, or...) stereo amps, if you can get almost the same results..., not to mention that some of them are designed by very experienced persons.

regards
ivica

[martin2395]

It's the cost and hassle that keeps most people from taking the fully active route while you can still get pretty damn amazing results with passive networks. These don't come cheap either but are simple as 1-2-3.

I guess that most of the 3-4 way JBL systems in the EU are in audiophile's hands and these guys don't just buy a Crown + BSS and call it a day. His invoice would probably look something like this:

- 4 stereo amps / 8 monoblocks, most probably 4 different ones for each band :D
- 4 sets of speaker cables from a boutique brand
- 4 sets of interconnects - see above
- 2-in-8out DSP or Quadruple Bryston 10B's in 2-way setting or maybe even something from Marchand/Pass.

Some people like the flexibility of DSP, I don't. Why? It makes you tweak the settings all the time :D

[4313B]

Good post!

I'm working on some passive networks for my M2 clones but I'd never tell anyone on this board that I was doing so.

[pauly]

Why?

Thanks
Pauly


Good post!

I'm working on some passive networks for my M2 clones but I'd never tell anyone on this board that I was doing so.

[jlyons]

A bi amp only variant of the crossover 3133, the 3132, was offered by JBL which uses no tapped inductors, and merely splits the mid and hi and includes a variable L pad. I have used this to good effect, and hope you will try the bi amp option. The 3132 is a 2 way network, and would not be difficult to clone. But you seem committed to the Pass option, which i suppose will offer great results while requiring only one amp.

[pauly]

G'day Jlyons,

Thanks for that - When it's all said and done, the money is an issue. Buying 2 or three stereo power amps, and a quality crossover will cost many thousands of dollars, let alone the complexity, cabling, power, & heat that are all factors. For less that a grand, I'm able to build these passives with quality components and am hoping they will improve the speakers (which still sound good anyway!).

Thanks for your endorsement of the Pass crossover too.

Thanks
Pauly


A bi amp only variant of the crossover 3133, the 3132, was offered by JBL which uses no tapped inductors, and merely splits the mid and hi and includes a variable L pad. I have used this to good effect, and hope you will try the bi amp option. The 3132 is a 2 way network, and would not be difficult to clone. But you seem committed to the Pass option, which i suppose will offer great results while requiring only one amp.

[audiomagnate]

A bi amp only variant of the crossover 3133, the 3132, was offered by JBL which uses no tapped inductors, and merely splits the mid and hi and includes a variable L pad. I have used this to good effect, and hope you will try the bi amp option. The 3132 is a 2 way network, and would not be difficult to clone. But you seem committed to the Pass option, which i suppose will offer great results while requiring only one amp.

I've never heard 4333's but own a pair of L300's and even bone stock I think they're absolutely terrific, a true work of art. My point is that if you're going to go to all the trouble and expense of redoing them, why not go go active? I'm no longer a "believer" in "audiophile" electronics after 40 years in the business; I like pro stuff and think lots of it sounds great. A good pro EC, a handful (five to be precise) of pro amps (like my beloved and dirt cheap BGW 620Bs) and a parametric EQ are easier and cheaper to implement in this application than a custom built high quality passive crossover, and I know damn well it will absolutely crush it performance-wise, and allow some room tuning to boot. Some measurement equipment helps too of course, but that's dirt cheap these days too.

Having said that, the most cost effective upgrade for a pair of L300's/4333's (I know they don't sound exactly the same) is an EC, some big amps a few good subs. Stock L300's with a good sub swarm sound absolutely incredible.

[4313B]

Understood but most people just don't want to go there with the whole multi-amp fully active setup. That is why a $75k system like the Everest II has a passive filter. That is why systems like the 4331, 4333, 4343 and 4345 had the bi-amp option instead of being bi-amp only.

If I had a pair of L300's I'd probably try the Nelson Pass filter if only to see what it sounded like. It can be interesting to hear what other folks come up with.

[audiomagnate]

If I had a pair of L300's I'd probably try the Nelson Pass filter if only to see what it sounded like. It can be interesting to hear what other folks come up with.
I did try it, and it sounded slightly better, but rolling them off at 50, stuffing the cabs and ports and adding a sub swarm took them to a whole new level. Massive difference as opposed to subtle improvement.

[4313B]

Interesting... way back in something like 1980 Mark Gander suggested something similar to me. He suggested blocking off the ports and loading the enclosures up with fiberglass. If I remember correctly, Mark was the transducer engineer on the 124/2203 and 136/2231 and Greg was the systems engineer on the L300/4333.

[martin2395]

I've seen speakers from Stage Accompany that used "15 JBL units (they introduced their own versions later, sold as SA1503/1513) and every time I saw the inside, they were well stuffed with Rockwool-like material.

[audiomagnate]

I did the same thing to my L200's and the improvement is even more dramatic on this older design. IMO, those big empty boxes sound like...big empty boxes. Both designs are attempting to do something they just aren't suited for, produce deep bass, and in the attempt you get a lot of box noise. Relieving them of that task completely and turning it over to modern subs lets them do what they do better than almost any other speaker out there, reproducing music with very low dynamic compression. In addition to damping up the box and ports, rolling off the bass makes them sound better for the well known reason of reduced IM distortion, but you're also not exciting those nasty box resonances in the first place, so it really does clean them up amazingly well. Try it, you'll love it.

[ivica]

Interesting... way back in something like 1980 Mark Gander suggested something similar to me. He suggested blocking off the ports and loading the enclosures up with fiberglass. If I remember correctly, Mark was the transducer engineer on the 124/2203 and 136/2231 and Greg was the systems engineer on the L300/4333.

Hi 4313B,

Interesting, some simulation data can show, but due to Xmax, HP filter around 50Hz would be welcome.

regards
ivica

[pauly]

Gday,

well for those that are interested, I made the crossovers and installed them yesterday. My wife and I spent some time listening last night. There is a substantial improvement in that 'in the room' feeling with voices and instruments able to be pinpointed in the soundstage easier. I think my wife said it best when she said 'it's like I'm hearing this for the first time!' About a track we have listened to for twenty years.

So....my verdict is that the project was a great success, though expensive. The only change I made to the Nelson pass circuit was to use the tweeter circuit from a jbl 4333 crossover that had a 1uf cap in series with a 1.5uf cap, with an inductor between the two to ground.

Ill try try to create a document with photos etc over the coming weeks that describe the build.

thanks
pauly



Hi all,

I'd really appreciate some informed comment on the issue of crossovers for 4333 monitors.

I have speakers configured (physically) as 4333 which have. 2234 woofer, 2420 mid with horn and lense, and slots. Currently they have 3110a crossovers.

I want to build new crossovers and have purchased the components to build Nelson pass' version of the l300 crossovers.
-however-
I was browsing this site and saw an equivilent circuit to the 4333 genuine crossovers that didn't use the tapped inductor.

Now I have a quandary.The speakers are used to mix music sometimes, and enjoy music ...let's say 40/60%. I use a set of questeds for critical mixing.

So...I'd really appreciate hearing some informed comment on each of these crossovers, and recommendations on which way to go, particularly from those that have heard or built them.

Thanks
Pauly

[pauly]

G'day all,

Here's a document running through my build of the Nelson Pass crossovers.

67991

Thanks Pauly


Gday,

<SNIP>

Ill try try to create a document with photos etc over the coming weeks that describe the build.

thanks
pauly

[Mr. Widget]

G'day all,

Here's a document running through my build of the Nelson Pass crossovers.

67991

Thanks PaulyThanks for the write up. I'm not suggesting you do do this, but it would be a very interesting exercise to build a charge coupled version of Nelson's network using more affordable components and compare the two versions.


Widget

[tjm001]

Just what I needed for my L300-like project. Thanks again.

Tom

[pauly]

Gday tom
great , I hope it helps. Looking forward to seeing your results.

pauly,


Just what I needed for my L300-like project. Thanks again.

Tom

[tjm001]

G'day all,

Here's a document running through my build of the Nelson Pass crossovers.

67991

Thanks Pauly

Greetings again Pauly. I was reviewing your modified tweeter circuit and saw a possible typo in the text next to Figure 2. In the text you stated you added a 1.8 mh inductor between the capacitors, but Figure 2 shows a .16 mh inductor. Just wondering which one you used. Thanks.

Tom

[Ian Mackenzie]

Thanks for the great write up.

My own interpretation is that optimising the phase response in the crossover response is the key to getting the best out of this type of two way system.

Obtaining correct phase response in the crossover region is more important than designing for flatness of response alone without checking the phase response.

The phase response in this context includes the physical driver offset of the actual acoustic centres and the electrical phase of both the low and high pass filters.

A test of the correctness of the phase response is to reverse the normal orientation of the woofer and the horn which should yield a steep notch in the acoustic crossover point at the desired listening position.

When the physical locations of the drivers are fixed subtle adjustment by spreading the crossover points can accomplish this.

As Nelson points out sometimes compromises need to be made in a commercial design to protect warranty for high power applications but if this is not a key requirement then some refinement can be considered.

I have always considered the L300 to be one of my favourite JBLs.

Enjoy.

[pauly]

Gday tom,
Yes I noticed a couple of typos and that's one of them! I actually used a 0.2 mh inductor there as that was the closest I could get.

thanks
pauly



Greetings again Pauly. I was reviewing your modified tweeter circuit and saw a possible typo in the text next to Figure 2. In the text you stated you added a 1.8 mh inductor between the capacitors, but Figure 2 shows a .16 mh inductor. Just wondering which one you used. Thanks.

Tom

[pauly]

Gday ian,
Yes I agree. There's only so much you can do when then horn is sitting back there!
A long time ago, I experimented with a long throw jbl horn and jbl 12 in low mids, with subs in separate boxes. I put the 12 inch speakers voice coil at the same plane of the horn driver (they lined up vertically), and built a horn box around them, with a 2405 in a phase plug in front of the 12 inch speaker. They sounded sensational. I attributed that to the physical phase coherence of the low mid and horn driver. No need to run out of phase :-)

thanks
Pauly



Thanks for the great write up.

My own interpretation is that optimising the phase response in the crossover response is the key to getting the best out of this type of two way system.

Obtaining correct phase response in the crossover region is more important than designing for flatness of response alone without checking the phase response.

The phase response in this context includes the physical driver offset of the actual acoustic centres and the electrical phase of both the low and high pass filters.

A test of the correctness of the phase response is to reverse the normal orientation of the woofer and the horn which should yield a steep notch in the acoustic crossover point at the desired listening position.

When the physical locations of the drivers are fixed subtle adjustment by spreading the crossover points can accomplish this.

As Nelson points out sometimes compromises need to be made in a commercial design to protect warranty for high power applications but if this is not a key requirement then some refinement can be considered.

I have always considered the L300 to be one of my favourite JBLs.

Enjoy.