• 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!

Crossover, passive to active, not that easy?

quorzar

Member
Joined
Dec 27, 2018
Messages
61
Likes
67
Hi,

I have a DIY set of speakers with a Tangband 1772 and 2 Eminence Alpha 15 like woofers for the lows. The speaker has a simple 500Hz 18 dB/octave HP, 12 dB/octave LP crossover. Because I bought a Hifiberry Beocreate I was thinking of making my speaker pair active and then use the 4 channel output to make it an active crossover.

My question is now, is it that easy to only copy paste the 18 dB/octave HP, 12 dB/octave LP and the 500Hz into the config of the Beocreate and then everything works or are there some rules to follow that make it harder or even impossible? And yes, MiniDSP and Hypex vs Beocreate is a bad change, but thats not the discussion here.

TY
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa
Hi,

I have tried several dozen dsp crossovers (on minidsp2x4 hd), and I would judge it to be trivial to recreate any cross you need.

it also opens up some asymmetric crossover topologies that would be pragmatically impossible to do with a post amp crossover.

After dozens of experiments I settled on the Harsch topology as best, I think it sounds best because of its inherent low phase shift through crossover frequency.

I suggest a 300-350 hz Harsch cross. The low pass is 24 dB butterworth, the high pass is 12 dB Bessel and the fullrange is delayed* by 1/2 cycle at your chosen cross frequency.

I suspect you would also appreciate the ability to add some bass boost under 100 hz with the dsp.

* Delays don’t really function properly on open baffles. If you delay an element for the front lobe, you are advancing that element in the rear lobe.

Minidsp also has the ability to create FIR filters, but in my experience there are not enough taps available to cope with anything below about 2 khz, so I don’t use it.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa

Speedskater

Major Contributor
Joined
Mar 5, 2016
Messages
1,610
Likes
1,333
Location
Cleveland, Ohio USA
David Rich technical editor at "The Audio Critic" magazine, writes:

Crossovers
“In the old days, there was a lot of trial and error in designing a crossover. So, adding more than a few components was futile, since it wasn’t practically possible to optimize them. Simplicity yielded better optimizations”.
“The individual crossover components usually don’t have functions in a way that can be isolated. A crossover is a filter network that implements a transfer function as a result of all of its parts. Sometimes it is possible to generalize and say, Oh, this resistor adds damping. Or, This network compensates for a resonance. But, really, the better a crossover design is, the more the parts work together, symbiotically”.
Dave Rich


So sometime making an active copy of a passive crossover is not simple.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa
I’d say with electronic cross you can emulate anything (if dsp based, easily...opamp much more work to get there, but better overall). You need to measure the drivers with and without cross to know what to do in the electronic version. That can be simple with 1st/2nd order, but yours is asymmetric.

I found using minidsp I could try until I got it right (measured same as post amp cross)...and once that’s done, it’s easy to implement in an opamp design, because the building blocks are similar.
 

gene_stl

Addicted to Fun and Learning
Forum Donor
Joined
Jun 14, 2019
Messages
867
Likes
1,198
Location
St.Louis , Missouri , U.S.A.
I have been told by several(three) audiophiles , all of whom should know better, that there is "magic" in some high level crossover networks.
The main thing they try to convince me of is that there is (what must be special) equalization for individual drivers that a generic line level active crossover cannot add. My reply is that if your driver needs "magic" EQ you shouldn't be using it. If it's frequency response is wrong then you have a built in non linearity. And anything you can do with a high level network can be much more easily done or experimented with with a DSP.

This is typical audiophile nonsense. The use cases where passive might be better than active are all rare or unusual or even non existent.

But I am a multi amp evangelist.

A lot of things that are spoken of and argued about , would not even be audible if you could manage a decent blind test. Although not blinded I used to test a lot of different crossover characteristics in my system and usually could not hear much difference and especially could not usually decide on a preference. They were not blind tests but I did not have much prejudice for one setting over the other. What I came away with was how little I could discern. (and please spare me the, "Oh well you can't hear! :);):p)

I think if you took ANY four way DIY system , wherein the drivers were selected(and mounted and baffled) with even the slightest care, and quad amped it, it would sound better than a huge percentage of the overpriced systems on the market. I have seen it done multiple times.

I get it that bi tri and quad amping may be more than many people want to commit to.
 
Last edited:

Kal Rubinson

Master Contributor
Industry Insider
Forum Donor
Joined
Mar 23, 2016
Messages
5,273
Likes
9,790
Location
NYC
I have been told by several(three) audiophiles , all of whom should know better, that there is "magic" in some high level crossover networks.
fv00255[1].gif
 
Last edited:
D

Deleted member 2944

Guest
Yeah, there's no magic in high-level crossovers. Any standard, passive crossover can be duplicated by simply making a differential measurement across the network of each section and then matching that transfer function with a DSP (or analog) line-level equivalent.
As an example, this is what such a measurement of the JBL 530 yielded:
JBL 530 electrical drive.png


But that's not say you can't improve upon an existing design when you venture into the line-level/multi-amp world. That is 'definitely' possible.

Dave.
 
Last edited by a moderator:

gene_stl

Addicted to Fun and Learning
Forum Donor
Joined
Jun 14, 2019
Messages
867
Likes
1,198
Location
St.Louis , Missouri , U.S.A.
I have even been willing to concede that some of the advantages of multiamping are slightly reduced "nowadays", since the wide wide availability of very powerful amps with sub audible distortion and less IM than back in the "olden days." But most advantages remain and flexibility, ease of experimentation, and wonderful adjustability are just a few. I also have strongly believed for decades that it is the most economical way to make your system more wonderful than your audiophile friends.:p;):cool:
 
Last edited:
D

Deleted member 2944

Guest
There's much entrenched audiophileness on this topic. It's been that way for many years, and I don't see anything changing it.
There are some audio pundits (like Kal, I suspect) who can grasp the concept and don't get side-tracked on some silly tangent when discussing it, but those are the minority, unfortunately.

Usually when I'm discussing the advantages of active setups I comment a multi-amp system with line-level crossover is simpler in concept and execution than a traditional setup. The audiophile eyes start rolling right about then. :)

Dave.
 
Last edited by a moderator:

gene_stl

Addicted to Fun and Learning
Forum Donor
Joined
Jun 14, 2019
Messages
867
Likes
1,198
Location
St.Louis , Missouri , U.S.A.
I am all ears
I have not had the pleasure of the JBL M2. I have heard the 4367. The weren't biamped.

I have not read a JBL explanation of the "dual voice coil " arrangements of both M2 (and I think 4367 ) drivers, nor have a seen the EQ that is programmed into the matching Crown DriveCore amps. But they do recommend a multi amp set up for the best results.

I am a long time JBL fan. The BFF who helped me build my speakers built his own and made them all top of the line JBL (077, 375, LE10 and LE15/PR15. I liked them alot and did not particularly find them to be objectionably different from mine with Be drivers. I think they were probably able to play louder than mine though mine are able to play at pretty ear shattering levels.

It has been a geologic age since I heard those JBLs. I liked the 4367 but did not love them. They seemed a little bright and harsh to me. They were driven with very expensive gear but full range with a high level crossover.

I hope to hear M2 s one day.
 

watchnerd

Grand Contributor
Joined
Dec 8, 2016
Messages
12,449
Likes
10,408
Location
Seattle Area, USA
I am all ears
I have not had the pleasure of the JBL M2. I have heard the 4367. The weren't biamped.

I have not read a JBL explanation of the "dual voice coil " arrangements of both M2 (and I think 4367 ) drivers, nor have a seen the EQ that is programmed into the matching Crown DriveCore amps. But they do recommend a multi amp set up for the best results.

I am a long time JBL fan. The BFF who helped me build my speakers built his own and made them all top of the line JBL (077, 375, LE10 and LE15/PR15. I liked them alot and did not particularly find them to be objectionably different from mine with Be drivers. I think they were probably able to play louder than mine though mine are able to play at pretty ear shattering levels.

It has been a geologic age since I heard those JBLs. I liked the 4367 but did not love them. They seemed a little bright and harsh to me. They were driven with very expensive gear but full range with a high level crossover.

I hope to hear M2 s one day.

The 4367s you heard were heavily EQed passively in the crossover.

So it points out a flaw in your idea that if you need to EQ a driver, you shouldn't be using it.
 

Plcamp

Addicted to Fun and Learning
Joined
Jul 6, 2020
Messages
860
Likes
1,315
Location
Ottawa
The 4367s you heard were heavily EQed passively in the crossover.

So it points out a flaw in your idea that if you need to EQ a driver, you shouldn't be using it.

I disagree with that, because post amplifier equalization equates to loss of efficiency...the thing realism requires the most.

I don’t have any yet, but I think I have become a (closet) horn fan.
 
D

Deleted member 2944

Guest
Nearly every crossover in every speaker system ever built uses some sort of shaping to flatten the curves of individual drivers and also to integrate them with each other in the crossover region(s). To think that "equalization" of this sort is not necessary or not being used is naive, at best.

Dave.
 

AnalogSteph

Major Contributor
Joined
Nov 6, 2018
Messages
3,334
Likes
3,278
Location
.de
True, a certain amount of flexitivity in the XO is definitely required. You won't get very far if you've only got textbook level LP/HP functions available.

For the sake of completeness, another potential snag worth mentioning is that the gain structure / level distribution throughout the system changes. On the one hand, not having any losses in the passive XO is a good thing. However, this can also come back to bite you in the rear end if you have a high sensitivity tweeter that requires a lot of attenuation - it'll do its best to present any hiss at the amplifier output (from the amplifier itself or preceding stages) in all its glory. Plus any potential advantages in distortion performance the driver might gain from approximate current driving.
Clearly you can always add a dropper resistor, you just need to take its interaction with driver impedance into account as well. Not a super major complication, it just changes how you have to calculate the response deltas a bit.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL

gene_stl

Addicted to Fun and Learning
Forum Donor
Joined
Jun 14, 2019
Messages
867
Likes
1,198
Location
St.Louis , Missouri , U.S.A.
The fact that JBL does this or that in a particular model proves nothing whatsoever. The 4367 is 45 years newer than my speakers and supposedly has lots of new technology in it. It was in a very high end system in an acoustically treated environment and demoed by it's proud owner with outrageously overpriced electronics (Several pieces were recent and current Mark Levinson).

Going into the audition I was concerned that I would discover that technology had left me behind and that I would become dis satisfied with my own system. However no such thing occurred. I listened to a number of files of music which I was familiar with and while I would not have characterized the sound as bad, it certainly did not make me want to start reaching for my wallet. Not in the least. I felt that it had a lot of the problems which JBLs got accused of back in the day, namely "too bright and forward" I found the sound harsh.(please don't tell me I am not used to flat frequency response. I am. I tune my system with a one third octave RTA to get my initial settings and have done so for decades) Being JBLs I was hoping to like them but I really didn't. ymmv I still think if a driver requires lots of EQ before you even use it you have selected or designed wrongly.

In all of our audio gear we strive for low distortion , flat frequency response and linearity. Why should speaker design be any different. Putting EQ into a high level passive crossover makes worse the reasons why you want to multi amp in the first place. Such EQ can and is inserted into multiamped systems too (ie the M2 biamp system using two DriveCore amps with everything done on the DSP input of the two power amps).

Perhaps we shall have to agree to disagree on this issue.
 
Last edited:

levimax

Major Contributor
Joined
Dec 28, 2018
Messages
2,348
Likes
3,462
Location
San Diego
Hi,

I have a DIY set of speakers with a Tangband 1772 and 2 Eminence Alpha 15 like woofers for the lows. The speaker has a simple 500Hz 18 dB/octave HP, 12 dB/octave LP crossover. Because I bought a Hifiberry Beocreate I was thinking of making my speaker pair active and then use the 4 channel output to make it an active crossover.

My question is now, is it that easy to only copy paste the 18 dB/octave HP, 12 dB/octave LP and the 500Hz into the config of the Beocreate and then everything works or are there some rules to follow that make it harder or even impossible? And yes, MiniDSP and Hypex vs Beocreate is a bad change, but thats not the discussion here.

TY
I have have been playing around with converting passive to active crossovers for several years. Recreating the original crossovers in DSP (including level matching drivers) is not difficult although you do need to check your work with REW or similar to make sure things are working as expected. Recently I "checked" what I have accomplished by going back to passive and I have to say I was disappointed. While the actives have a lot of advantages on paper, quiet and powerful amps and more importantly tools like REW and rephase / convolution, which allow "fine tuning" of passive crossovers, makes any real improvements hard to realize. While I was originally convinced active was the best "bang for back" improvement you could make I am not so sure anymore. If I EQ passive and active with REW / Rephase / convolution I really can't tell them apart. To me it seems like the real difference between speakers is the mechanical part.... how the drivers and enclosure work together. If you get that part right everything else falls in place and heroic efforts of crossover design are not required. I would still suggest you try it as it is fun and you will learn a lot.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,201
Likes
16,983
Location
Riverview FL
While I was originally convinced active was the best "bang for back" improvement you could make I am not so sure anymore. If I EQ passive and active with REW / Rephase / convolution I really can't tell them apart.

I EQ my passives too, with what seems to be good results.

I even EQ my black-box actives, since what is built into them doesn't know my room or preferences.
 
Top Bottom