Can someone please explain to me how this crossover works?

B

bonchef

Audiophyte
I have this mystery with me since 2007.

Back then my Celestion A1's (from the 90s) tweeter was damaged. I ordered a replacement from Celestion and was pissed to see a cheapo replacement for the price I paid for. So I decided to re-build speakers with new tweeters and crossovers.

I came across the following design (see my crappy drawing) where the author did not respond to my inquiry. Armed with virtually no analog circuit knowledge (besides some basics of C/L/R), I put together a pair pf XOs based on the design anyway, and it worked beautifully that to-date, I have not heard any speakers in the thousands USD range that would make me want to upgrade.

I always wanted to learn exactly how it works and maybe fine tune the components' values. But I lost the schematic and reluctant to re-open the speakers, until this week, where one of the solder joint snapped off (due to my poor worksmanship I am sure). So I opened it up and traced the circuit. I hope my drawing is correct. Let me know if something does not make sense, and I will triple check the connections...

So here you go - can someone please explain to me how this crossover works?

The drivers:
- Woofers - original A1's. I assume the sensitivity is around 88db because the speakers were rated as such
- Tweeters - Morel MDT40
 

Attachments

carlthess40

carlthess40

Audioholic
Is that the crossover that the speaker came with ? If so. Those are some nice parts
It may be a 2nd order filter. I’m 90% sure that I’m wrong. Lol


Sent from my iPhone using Tapatalk
 
D

D Murphy

Full Audioholic
I have this mystery with me since 2007.

Back then my Celestion A1's (from the 90s) tweeter was damaged. I ordered a replacement from Celestion and was pissed to see a cheapo replacement for the price I paid for. So I decided to re-build speakers with new tweeters and crossovers.

I came across the following design (see my crappy drawing) where the author did not respond to my inquiry. Armed with virtually no analog circuit knowledge (besides some basics of C/L/R), I put together a pair pf XOs based on the design anyway, and it worked beautifully that to-date, I have not heard any speakers in the thousands USD range that would make me want to upgrade.

I always wanted to learn exactly how it works and maybe fine tune the components' values. But I lost the schematic and reluctant to re-open the speakers, until this week, where one of the solder joint snapped off (due to my poor worksmanship I am sure). So I opened it up and traced the circuit. I hope my drawing is correct. Let me know if something does not make sense, and I will triple check the connections...

So here you go - can someone please explain to me how this crossover works?

The drivers:
- Woofers - original A1's. I assume the sensitivity is around 88db because the speakers were rated as such
- Tweeters - Morel MDT40
Are you sure you have that traced out correctly? The woofer circuit by itself is a simple 2nd-order electrical with a damping resistor on the 15 uF Capacitor to control the shape of the knee of the roll-off. That could translate into a 3rd or 4th order acoustic slope depending on the natural roll-off of the woofer. The tweeter circuit would make more sense if the 8 uF cap followed the 2.5 Ohm resistor and then connected to the positive terminal of the tweeter. There shouldn't be anything in the negative tweeter path, assuming this is a standard parallel crossover and not some kind of weird series topology. I don't suppose you still have the link to the dude who cooked this up?
 
TLS Guy

TLS Guy

Seriously, I have no life.
Are you sure you have that traced out correctly? The woofer circuit by itself is a simple 2nd-order electrical with a damping resistor on the 15 uF Capacitor to control the shape of the knee of the roll-off. That could translate into a 3rd or 4th order acoustic slope depending on the natural roll-off of the woofer. The tweeter circuit would make more sense if the 8 uF cap followed the 2.5 Ohm resistor and then connected to the positive terminal of the tweeter. There shouldn't be anything in the negative tweeter path, assuming this is a standard parallel crossover and not some kind of weird series topology. I don't suppose you still have the link to the dude who cooked this up?
To me that looks like a series crossover circuit with modification. Very unusual. Series circuits are not often used but, I have one in my rear backs. They can work well, but I not often the ideal choice.
 
B

bonchef

Audiophyte
I should add that the time I picked up the design I already knew it was an unusual/atypical design, hence the question.

By the way, I thought the guy (Lars) took off the page - but actually it is still there after all these years! Maybe this time around he will answer my question due to Covid-stay-home LOL. Now you can check if my drawing is correct. I never quite get the hang on topolgy beyond simple serial and parallel circuits... am I right that this is a network?

http://lcaudio.com/index.php?page=26 (detail)

http://lcaudio.com/index.php?page=15 (intro)
 
TLS Guy

TLS Guy

Seriously, I have no life.
I should add that the time I picked up the design I already knew it was an unusual/atypical design, hence the question.

By the way, I thought the guy (Lars) took off the page - but actually it is still there after all these years! Maybe this time around he will answer my question due to Covid-stay-home LOL. Now you can check if my drawing is correct. I never quite get the hang on topolgy beyond simple serial and parallel circuits... am I right that this is a network?

http://lcaudio.com/index.php?page=26 (detail)

http://lcaudio.com/index.php?page=15 (intro)
You did not draw the circuit correctly. It is a classic second order series crossover. These are not often used as they are actually at a disadvantage with the usual parallel circuit.

For first order they can have an advantage. Series crossovers are hard to implement. When I built my state of the art studio monitors in the early eighties, I was interested in having speakers as phase and therefore as time linear as possible. These speakers are my rear backs now. They are phase and time aligned from 180 Hz to 20 KHz.

Since the bass drivers are KEF B 139s in a TL they needed a far steeper slope than a first order crossover can provide. However with the cooperation and interest of Dynaudio a phase coherent crossover was developed over 10 years, yes 10 years, to incorporate a Dynaudio cone midrange and their smaller midrange dome the D52 AF and their phenomenal D21 AF 3/4" dome. Crossovers were at 180 Hz third order active, 900 Hz first order series passive and 5 KHz first order parallel.
Although seemingly simple, it was far from it in achieving a really superb first class speaker. These will till stand the comparison with anything currently available.

The reason for all this was that I became in addition to my responsibilities as an ICU physician, I became the outside broadcast engineer for the local public radio station for classical programs. I had a keen interest in making time coherent recordings and therefore required highly time coherent speakers for their evaluation.

After many years of experience and experiment and can vouch for the fact that a phase coherent recording played back through essentially phase coherent speakers does in fact produce the closets approach to the original sound and set a significantly superior standard.

Now your circuit is interesting, but second order and therefore not time coherent.

You can find more about all of this here. But personally I would only use a series crossover network in a first order network. So in essence I'm not overly impressed with that network.
 
D

D Murphy

Full Audioholic
You did not draw the circuit correctly. It is a classic second order series crossover. These are not often used as they are actually at a disadvantage with the usual parallel circuit.

For first order they can have an advantage. Series crossovers are hard to implement. When I built my state of the art studio monitors in the early eighties, I was interested in having speakers as phase and therefore as time linear as possible. These speakers are my rear backs now. They are phase and time aligned from 180 Hz to 20 KHz.

Since the bass drivers are KEF B 139s in a TL they needed a far steeper slope than a first order crossover can provide. However with the cooperation and interest of Dynaudio a phase coherent crossover was developed over 10 years, yes 10 years, to incorporate a Dynaudio cone midrange and their smaller midrange dome the D52 AF and their phenomenal D21 AF 3/4" dome. Crossovers were at 180 Hz third order active, 900 Hz first order series passive and 5 KHz first order parallel.
Although seemingly simple, it was far from it in achieving a really superb first class speaker. These will till stand the comparison with anything currently available.

The reason for all this was that I became in addition to my responsibilities as an ICU physician, I became the outside broadcast engineer for the local public radio station for classical programs. I had a keen interest in making time coherent recordings and therefore required highly time coherent speakers for their evaluation.

After many years of experience and experiment and can vouch for the fact that a phase coherent recording played back through essentially phase coherent speakers does in fact produce the closets approach to the original sound and set a significantly superior standard.

Now your circuit is interesting, but second order and therefore not time coherent.

You can find more about all of this here. But personally I would only use a series crossover network in a first order network. So in essence I'm not overly impressed with that network.
I'm a little confused. Did you mean to say you did not draw the circuit incorrectly?
 
D

D Murphy

Full Audioholic
I'm saying that the hand drawn circuit in his original post is not correct.
the drawing looks like a weird combination of a series and parallel circuit. I wasn't sure what you were referring to when you said "This is a classic series crossover.........."
 
S

stalag2005

Full Audioholic
Capacitors band pass high frequency, the inductive coils band pass low frequencies. Depending on how many stages, one can fine tune the crossover to each speaker.
 
newsletter

  • RBHsound.com
  • BlueJeansCable.com
  • SVS Sound Subwoofers
  • Experience the Martin Logan Montis
Top