• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

ASR reccomended external crossovers?

Neddy

Addicted to Fun and Learning
Joined
Mar 22, 2019
Messages
754
Likes
1,019
Location
Wisconsin
Thanks! (I'm a huge Sellers/Goonshow fan, of course, and just enjoy Being There, so to speak.)
I've spent some quality time with a pair of DIY M2s, and hope someday I get a chance to compare them to 4367s in person.
The horn/waveguide costs for either HF components are beyond my budget reach, otherwise I'd add one of those as well.
(Your thread here got me looking up pricing on them - Again.)
FWIW, even the owner of the DIY M2s agrees that there's 'just something' a tiny bit more 'musical' about the old JBL horn combos than the new ones, which are dramatically cleaner and much broader/smoother horizontal/vertical dispersion.
RoomEQ is Your Friend.
Thanks again, and Enjoy!!!
 

CDMC

Major Contributor
Forum Donor
Joined
Sep 4, 2019
Messages
1,172
Likes
2,320
Right.
Biwiring -- still puzzles me. ;)

The idea behind it is that back EMF from the woofer doesn’t affect the tweeter and midrange. Some manufacturers insist it makes a major difference (Vandersteen), others none (Salk).
 

mhardy6647

Grand Contributor
Joined
Dec 12, 2019
Messages
11,211
Likes
24,170
The idea behind it is that back EMF from the woofer doesn’t affect the tweeter and midrange. Some manufacturers insist it makes a major difference (Vandersteen), others none (Salk).
yeah... but... I mean...
everything is still connected at the amplifier...

:)
 

Doodski

Grand Contributor
Forum Donor
Joined
Dec 9, 2019
Messages
20,745
Likes
20,756
Location
Canada
I assume I invalidate my warranty attempting something like this?
Leave the passive crossovers inside the speaker cabinets just like they are ready to be reconnected at any time. Do not cut off any wire ends and try to do it that way. If you need wire terminals then go buy the proper crimp on connectors and a crimper and do it properly. You may be able to use the present wiring from the terminals at the back of the speaker to the crossover boards. Keep pen and paper handy for notes and take pictures so you remember how to reassemble stuff. It is possible to do the wiring change over and not void any warranty coverage. I've never voided a speaker warranty unless it was a repetitive issue that was blatantly warranty void. Running a active crossover system is initially a time and labor sink but once it's done and operating it's very rewarding.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
What would be useful is a list of crossovers on the market and some explanation of methods. Would anyone be up for putting that together?

Making speakers like this full active is imo realy techi. . DSP imo needed and a lot of knowledge, measuring and tweeking. Thats not plug&play. So you have to be a real techi person. And they know what to do if they want to have a real benefit. I'am at least a little techi but making them better? This is for many already a dream speaker including me. If you like to do a engine tuning on a formula1 car you have to know what to do.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
Making speakers like this full active is imo realy techi. . DSP imo needed and a lot of knowledge, measuring and tweeking. Thats not plug&play. So you have to be a real techi person. And they know what to do if they want to have a real benefit. I'am at least a little techi but making them better? This is for many already a dream speaker including me. If you like to do a engine tuning on a formula1 car you have to know what to do.
I understand your point, but if you're converting an existing passive loudspeaker, you have that design as a starting point. The manufacturer's choice of crossover frequencies is a very good starting point, and the sensitivity of the individual drivers can be checked with the respective driver manufacturers if OEM, or from looking at the crossover design. In most cases, the bass unit is the most sensitive, and that's left unattenuated, whilst the MF and HF drivers tend to be more sensitive and have attenuation built into the crossover, so that gives you numbers to put into the active crossover.

At that point you can start listening/measuring and adjust the crossover parameters either to improve the measurements or just by ear, in the knowledge that you can always go back to the manufacturer's recommended starting point.

Yes, it's a bit techy, and it's not plug and play, but it's very satisfying, and I suggest pretty much always results in an improvement, confirmed by measurements. If for no better reason, DSP-based crossovers can easily do 48dB/octave L-R filters, whilst with passive designs, 24dB/octave is about the practical limit and many work on less.

If you really want to get into it, then you can apply EQ, and get the overall response as flat as you have the patience for. I achieved +-1dB above 200Hz with my 801s, I don't know below that as I can't measure it, and anyway, it's largely academic given how much the low end if affected by the room, but it won't be worse than the passive design.

S.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
I understand your point, but if you're converting an existing passive loudspeaker, you have that design as a starting point. The manufacturer's choice of crossover frequencies is a very good starting point, and the sensitivity of the individual drivers can be checked with the respective driver manufacturers if OEM, or from looking at the crossover design. In most cases, the bass unit is the most sensitive, and that's left unattenuated, whilst the MF and HF drivers tend to be more sensitive and have attenuation built into the crossover, so that gives you numbers to put into the active crossover.

At that point you can start listening/measuring and adjust the crossover parameters either to improve the measurements or just by ear, in the knowledge that you can always go back to the manufacturer's recommended starting point.

Yes, it's a bit techy, and it's not plug and play, but it's very satisfying, and I suggest pretty much always results in an improvement, confirmed by measurements. If for no better reason, DSP-based crossovers can easily do 48dB/octave L-R filters, whilst with passive designs, 24dB/octave is about the practical limit and many work on less.

If you really want to get into it, then you can apply EQ, and get the overall response as flat as you have the patience for. I achieved +-1dB above 200Hz with my 801s, I don't know below that as I can't measure it, and anyway, it's largely academic given how much the low end if affected by the room, but it won't be worse than the passive design.

S.

I absolutly agree. I just wanted to make clear that a action like this needs some knowledge. And passion to do it. It's more than just connect a new amp.
 

Juhazi

Major Contributor
Joined
Sep 15, 2018
Messages
1,717
Likes
2,897
Location
Finland
Passive bi-amping, i.e. using two (or more) power amps whilst retaining the loudspeaker's passive crossover is totally pointless. It's just the more expensive version of bi-wiring which is pointless enough in itself.

For a reason why passive bi-amping doesn't and can't work, look up the Superposition Principle. As amplifiers are linear to an extraordinary extent, there can't be any benefit to passive bi-amping except to the dealer's profit.

Active bi-amping, however, using an external electronic crossover before the power amps has many benefits, all quantifiable.

For very good explanations, look at
https://sound-au.com/bi-amp.htm
and
https://sound-au.com/biamp-vs-passive.htm

S.
Yep, and that was in 2004.

Seriously, dsp-xo and eq is not easy, you must learn to take valid measurements and understand basics of loudspeaker and crossover design. There are ready-made dsp-xos for only some specific commercial and diy speakers.

ps. Real bi-amping audiophiles also know how to select best cables for each amp, not forgetting internal wirings... But since I am using the poor little minidsp, even good cables couldn't help!
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
If you want to learn cool it's fun.

But if you do it with the intention do make a speaker on this level realy better.. good luck!
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
If you want to learn cool it's fun.

But if you do it with the intention do make a speaker on this level realy better.. good luck!

My first attempt to use an external DSP crossover was on a far simpler speaker.

I took an old Dynaudio Audience 42 (15 cm mid woofer, 28 mm tweeter) with 1st order crossovers, and made an external DSP crossover.

I underestimated the other corrective elements to phase and response the passive crossover was applying.

The exercise was enough to convince me that it's very, very hard to improve the passive crossover of a well-engineered (not 2 guys in a garage) speaker from the likes of one of the Big Boys who make their own custom drivers (JBL/Revel, Focal, Dynaudio, TAD, etc.) without a serious amount of time and measuring equipment.
 

tomtoo

Major Contributor
Joined
Nov 20, 2019
Messages
3,607
Likes
4,514
Location
Germany
My first attempt to use an external DSP crossover was on a far simpler speaker.

I took an old Dynaudio Audience 42 (15 cm mid woofer, 28 mm tweeter) with 1st order crossovers, and made an external DSP crossover.

I underestimated the other corrective elements to phase and response the passive crossover was applying.

The exercise was enough to convince me that it's very, very hard to improve the passive crossover of a well-engineered (not 2 guys in a garage) speaker from the likes of one of the Big Boys who make their own custom drivers (JBL/Revel, Focal, Dynaudio, TAD, etc.) without a serious amount of time and measuring equipment.

Exactly what i think. This are well engineered. They use horns that need EQ. They are engineered to not only have good on axys response.To make them realy better thats not just a walk in the park.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
Exactly what i think. This are well engineered. They use horns that need EQ. They are engineered to not only have good on axys response.To make them realy better thats not just a walk in the park.

On top of all of that, the 4367 uses a pretty unusual charge-coupled crossover with a diode charging circuit.
 

pozz

Слава Україні
Forum Donor
Editor
Joined
May 21, 2019
Messages
4,036
Likes
6,825
Making speakers like this full active is imo realy techi. . DSP imo needed and a lot of knowledge, measuring and tweeking. Thats not plug&play. So you have to be a real techi person. And they know what to do if they want to have a real benefit. I'am at least a little techi but making them better? This is for many already a dream speaker including me. If you like to do a engine tuning on a formula1 car you have to know what to do.
A list of devices and a short intro in the form of an article would help anyone I think, even if only as a starting point.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
On top of all of that, the 4367 uses a pretty unusual charge-coupled crossover with a diode charging circuit.
Presumably this is used for overload protection, and isn't a part of the actual crossover. If so, then that can be very easily applied in DSP as a limiter. I've done this to protect my tweeters, set the limiter to come on at around 40 watts equivalent. Never actually activated, as I don't play that loud, but it's nice to know it's there.

If by any chance, the diode charging circuit isn't the overload protection, then I can't think what it could be used for.

S.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
Presumably this is used for overload protection, and isn't a part of the actual crossover. If so, then that can be very easily applied in DSP as a limiter. I've done this to protect my tweeters, set the limiter to come on at around 40 watts equivalent. Never actually activated, as I don't play that loud, but it's nice to know it's there.

If by any chance, the diode charging circuit isn't the overload protection, then I can't think what it could be used for.

S.

"The networks also employ DC voltage bias to operate the capacitors effectively in a Class A mode. DC Bias for the 4367 Studio Monitor is accomplished using a diode charging circuit internal to the crossover network and driven by the applied signal, so no additional external power is needed for this feature."

"Each capacitor position is actually made up of two capacitors connected in series. The voltage generated from the resistor and diode combination is applied to the center connection of the two capacitors. This produces a voltage potential between the two plates within the capacitor. When the two parts are taken as a whole, there is no DC voltage that appears across them, but individually they are each biased. The sonic result of the biasing yields an increase in detail, increased smoothness and considerably more natural decay of sounds within the music"

https://www.jblsynthesis.com/on/demandware.static/-/Sites-masterCatalog_Harman/default/dwcea263d6/pdfs/JBL Synthesis 4367 White Paper.pdf

So they say it makes it sound better.

They don't mention overload protection. And it exists on both the woofer and tweeter circuits.
 

sergeauckland

Major Contributor
Forum Donor
Joined
Mar 16, 2016
Messages
3,440
Likes
9,100
Location
Suffolk UK
"The networks also employ DC voltage bias to operate the capacitors effectively in a Class A mode. DC Bias for the 4367 Studio Monitor is accomplished using a diode charging circuit internal to the crossover network and driven by the applied signal, so no additional external power is needed for this feature."

"Each capacitor position is actually made up of two capacitors connected in series. The voltage generated from the resistor and diode combination is applied to the center connection of the two capacitors. This produces a voltage potential between the two plates within the capacitor. When the two parts are taken as a whole, there is no DC voltage that appears across them, but individually they are each biased. The sonic result of the biasing yields an increase in detail, increased smoothness and considerably more natural decay of sounds within the music"

https://www.jblsynthesis.com/on/demandware.static/-/Sites-masterCatalog_Harman/default/dwcea263d6/pdfs/JBL Synthesis 4367 White Paper.pdf

So they say it makes it sound better.

They don't mention overload protection. And it exists on both the woofer and tweeter circuits.
Ah well, yet another bit of manufacturer's nonsense. If they used non-polar capacitors, they wouldn't need any bias, but then it would cost a bit more and they wouldn't have a story to tell.

However, this is nothing to do with the crossover function, it's a way of polarising back-to-back electrolytics. Non-polar electrolytics are actually two normal electrolytics back to back in the same can so each one blocks the 'wrong' polarity for the other. It's the standard way of creating a non-polar electrolytic from what are essentially polarised components. It's cheaper and physically much smaller than using non polar plastic film capacitors.

There is some evidence that long-term, the lack of a polarising voltage across an electrolytic reduces the total life of the item, so it's an idea not totally without merit, but this does seem to me more a marketing exercise than a genuine technical one. The technical solution is to use non-polar capacitors. As to sound quality improvements, I don't think so.

Nevertheless, if replacing the passive crossover with an active solution, the whole problem of polarised electrolytics goes away.

S.
 
Top Bottom