DIY biamp 6-24 crossover

Hello members,

I've got a request from member PC997. He wants to drive his two-way-speaker (with passive crossover)
actively. Why not try to build a 6-24 active crossover (6-24 AXO)?
The drivers are:
SEAS W16NX-001/8 (MidBass), frequency range around 45Hz - 2500Hz
SEAS T25CF-001 (dome-tweeter), frequency range around 2000Hz - 25000Hz.
He sent me informations about the passive crossover and some SPL-graphs. I checked on the SEAS-internet-
site to get some infos about the drivers. I know the tweeter, but never used the Midbass W16NX-001/8.
I will add some pics below.
I asked PC997 if we shouldn't discuss this here in the public. Why? Because I think there are a few members out
there, who are thinking about this. Whether experienced or 'beginner level'.
And I am also not Mr. I-know-everything... No, I am learning every day.
Cheers
Dirk 😉
 

Attachments

  • seas_excel_loudspeaker_woofer__w16nx001.jpg
    seas_excel_loudspeaker_woofer__w16nx001.jpg
    69.2 KB · Views: 101
  • seas_excel_loudspeaker_tweeter_e0006_t25cf001.jpg
    seas_excel_loudspeaker_tweeter_e0006_t25cf001.jpg
    61.9 KB · Views: 100
  • SEAS-2_way-speaker_measurements_SPL-curves.jpg
    SEAS-2_way-speaker_measurements_SPL-curves.jpg
    376.5 KB · Views: 111
  • Like
Reactions: Russellc
On the third pic you can see the summed SPL-curve with the passive crossover.
Crossover frequency is clearly at 2000 Hz.
When I checked the schematic of the passive crossover, I immediately saw, that there is
some equalization in the passive crossover.
Especially the MidBass (W16NX-001/() has an 'interesting' behaviour in the upper frequency
range. This will be difficult, if not impossible to cure inside the 6-24AXO.
Cheers
Dirk
 

Attachments

  • 2-way-passive-crossover_seas_drivers.jpg
    2-way-passive-crossover_seas_drivers.jpg
    53.6 KB · Views: 144
Then I started to simulate the values for Cs and Rs for a two-way-active-crossover.
I would use a 24dB/oct.- slope. Crossover-frequency at 2000 Hz.
And I had always the 'problem' in mind how to solve that roll-off in the upper frequency range
of the W16NX-001/8. Hmmm? 🤔
I made the following simulation with Mike Rothachers-calculator - ignoring the roll-off of the W16NX-001. Check pics.
Cheers
Dirk
 

Attachments

  • SEAS_6-24AXO_2000Hz-sim.jpg
    SEAS_6-24AXO_2000Hz-sim.jpg
    70.9 KB · Views: 162
The 6-24AXO-board for the 2000Hz crossover with 24dB-slope could look like this:
 

Attachments

  • 6-24AXO_2000Hz_24dB_LowPass_values_board_move_to_other_channel.jpg
    6-24AXO_2000Hz_24dB_LowPass_values_board_move_to_other_channel.jpg
    268.5 KB · Views: 247
  • 6-24AXO_2000Hz_24dB_HighPass_values_board_move_to_other_channel.jpg
    6-24AXO_2000Hz_24dB_HighPass_values_board_move_to_other_channel.jpg
    268.2 KB · Views: 242
But how could we adress the problem with the roll-off of the MidBass?

Perhaps lowering the crossoverpoint of the lowpass in the 6-24AXO?

Any hints/tips are welcome! I am awaiting a good discussion. Let's have fun...
Cheers
Dirk 🤔
 

Attachments

  • seas_excel_loudspeaker_woofer__w16nx001.jpg
    seas_excel_loudspeaker_woofer__w16nx001.jpg
    69.2 KB · Views: 88
  • seas_excel_loudspeaker_woofer__w16nx001_roll-off.jpg
    seas_excel_loudspeaker_woofer__w16nx001_roll-off.jpg
    54.6 KB · Views: 83
Wow. Dirk, thank you so much for putting all this into a direct perspective for me. I am digesting all this information and will complete the board as described above. Speakers sound very good as they are now but I bet there are more to discover. I will post some pictures of the speakers.
Appreciate all your knowledge and enthusiasm for good audio.
 
Hello PC997,

you are welcome! But you know - there will be some homework left... 😉
A lot of listening and readjusting the filters ( I think especially the upper crossover frequency of the lowpass will have to be lowered?)
If you have questions - ask. I / We will try to help.
Adjust your trimpots before you solder them in. It is difficult / impossible to do this later.
Check every resistor with your DMM before you solder it in.
I hope you will be successful!
Have fun listening.
Cheers
Dirk 🙂
 
  • Like
Reactions: Russellc
Hello PC997,

you are welcome! But you know - there will be some homework left... 😉
A lot of listening and readjusting the filters ( I think especially the upper crossover frequency of the lowpass will have to be lowered?)
If you have questions - ask. I / We will try to help.
Adjust your trimpots before you solder them in. It is difficult / impossible to do this later.
Check every resistor with your DMM before you solder it in.
I hope you will be successful!
Have fun listening.
Cheers
Dirk 🙂
Yes.. Definitely Dirk..
Appreciate your help..
 
I am actually stuffing my 6/24 board, but I have three questions. Look at the image:

Unklarheiten auf PCB.jpg


1. The value 221K is printed, but it should be a 1MEG resistor. Correct?
2. I cannot find these caps in the schematics. Whats the purpose? /Edit: I just see in the diyaudio Guide from 6L6: they are part of the PSU
3. These are the input caps, isn't it?
3a. In the schematics 0.1uF, Here some 10uF.
3b. is the polarisation on the PCB correct? + is connected to the 1MEG resistor to ground, minus to the potentiometers wiper connection?

Thanks for your support!
Franz
 
Last edited:
I am looking through this thread and I could find some more answers:

1. Post #180 mentionned 1MEG is correct
2. is already answered by the build guide from 6L6
3a. Post @171 Input caps
3b. not found up to now. I am quite sure the polarization is not correct when using electrolytics in this place

Franz
 
Hello Franz Gysi,

the input caps are 10uF (if you use polarized caps). The polarity of the input caps is correct on the pcb.
The 6-24-AXO is single ended. The music - signals '0 V - level' is shifted midway between ground and the +24 V rail (slightly below 12 V).
Cheers
Dirk
 
When using the Toshiba 2SK170's in the buffers, is a bias resistor used?
The bias resistor on Q2 sets the current to a value lower than Idss of the fet. It is optional. The bias resistor for Q1 is optional and only trims the DC setting. If I revise the board in the future I will probably drop the Q1 resistor and also replace the Q2 with the much less expensive J113 or similar.
 
  • Like
Reactions: A Jedi