Converting 3rd order Butterworth to aB

Status
Not open for further replies.
I think ought to hit the hay in the UK. Way past my bedtime. 😀

Taking crossover low is a general solution full of pitholes as Lynn Olson wisely says:
http://www.diyaudio.com/forums/multi-way/190663-can-you-have-sparkling-treble-but-without-sibilance-2.html#post2604196

I don't mind metal tweeters as long as they have a Zobel to tame high frequency nasties.

We do know that a three way works much better than a two way though. I find an 8" bass works better tonally with two ways. A couple of models of mine below. Both are musically satisfying IMO.

I was reassured that Troels Gravesen shares my opinion. TQWT-

To be continued another day. :drink:
 

Attachments

  • Visaton Drive Units system7 lashup.jpg
    Visaton Drive Units system7 lashup.jpg
    81.9 KB · Views: 136
  • Modified_MA-R300MD_HT22-8.jpg
    Modified_MA-R300MD_HT22-8.jpg
    79 KB · Views: 147
re: "little 10L speaker I usually feel like kicking round the room" - Steve, what drivers are they?
It's some sort of Elac 6" bass from the eighties fitted in the Mordaunt Short MS15. Rubber dustcap being an Elac characteristic. I use a Visaton TW70 with it since it fits.

Originally the bass just had a 0.4mH coil, and second order 3.3uF/0.3mH Audax mylar tweeter with a 3.3R resistor. Sounded harsh and lots of cone breakup around 5kHz I guess.

I believe I have 2.5mH and 6.8uF bass with a tank notch of 33R and 0.33 uF around 6kHz, not too sure, and a third order 3kHz tweeter now. Cone tweeters need steeper filters IME. Cabinet is too small fundamentally.
 

Attachments

  • MS15_Final.jpg
    MS15_Final.jpg
    73.5 KB · Views: 129
Last edited:
Hi keilau,

Thought this link might be of interest to you.

http://www.diyaudio.com/forums/multi-way/189530-kef-104-kef-104ab-mkii-spice.html

Peter

Peter, thank you for the link.

I am very interested in the discussion on Vifa D27TG-35. Both Falcon and Willslow recommend it as a potential KEF T27 substitute. But they did not give the reason nor offer any information on crossover mod from the T27. The Vifa D27TG-35 was mentioned in the link above, but no follow-up discussion.

I have a pair of the Vifa D27TG-35 sitting on my shelf waiting for application.
 
I'm not sure that a third order KEF AB circuit is really helping you here, keilau. I have had some goodish results with the classic KEF 8" bass third order butterworth 3kHz BW3 filters, but these rely on the negative polarity on tweeter trick that works because of the approx 5cm distance between acoustic centres with this sort of BBC speaker.

I think you should be looking at a time aligned LR2 circuit. Because 6" bass just doesn't rolloff very fast. Troels gets a bit vague on detail with his preferred 2.5kHz v3 LR2 crossover, but not hard to model if you get the basic ideas:
18W-8434G00

This is not a journey I will share with you, because I have my own interests in musical speakers that are cheap to build, but I think you'd end up with something good. FWIW, Troels is still doing a Fs tweeter notch which works much like a KEF AB in practise. It all seems to make sense at a theoretical level, and is really pushing the envelope IMO. 😎

Here's an online calculator I use frequently to work out where notches act and with what Q:
Alan Yates' Laboratory - VK2ZAY's Engineering Calculators - LCR Resonance
Amazing what good stuff is available online through the goodness of the author's heart. 🙂
 
Last edited:
I'm not sure that a third order KEF AB circuit is really helping you here, keilau. I have had some goodish results with the classic KEF 8" bass third order butterworth 3kHz BW3 filters, but these rely on the negative polarity on tweeter trick that works because of the approx 5cm distance between acoustic centres with this sort of BBC speaker.

I think you should be looking at a time aligned LR2 circuit. Because 6" bass just doesn't rolloff very fast. Troels gets a bit vague on detail with his preferred 2.5kHz v3 LR2 crossover, but not hard to model if you get the basic ideas:
18W-8434G00

This is not a journey I will share with you, because I have my own interests in musical speakers that are cheap to build, but I think you'd end up with something good. FWIW, Troels is still doing a Fs tweeter notch which works much like a KEF AB in practise. It all seems to make sense at a theoretical level, and is really pushing the envelope IMO. 😎

Here's an online calculator I use frequently to work out where notches act and with what Q:
Alan Yates' Laboratory - VK2ZAY's Engineering Calculators - LCR Resonance
Amazing what good stuff is available online through the goodness of the author's heart. 🙂

Steve, you are throwing stuff at me faster than I can digest! LOL😉

I will finish my current Peerless-Dayton_Audio bookshelf with the commercial tricks #1 and #3 that Gravesen quoted.

#1. Use an asymmetrical crossover usually consisting of 2nd order to the bass and 3rd order to the tweeter to make "proper" phase integration.
#3. Tilt the front panel so that with a listening height somewhere between the bass and tweeter we more or less compensate for the time difference.

I will re-examine the crossover by investigating the benefit of notch filters at both the woofer and the tweeter side.

I am giving up the acoustic Butterworth for the Peerless-Audax bookshelf. I could not figure out how to get the Martin Colloms' transformation equations to work right.

The time aligned LR2 will be for a future project.

I found this LRC calculator and like it.

CalcTool: RLC or LC circuit calculator
 
Sloped baffles seem to work well enough. Below is one that somebody built for a series crossover I rustled up with Visaton drivers that play nicely together. The tweeter is set back about 2 cm.

The series filter is quite neat, and includes a hidden tweeter Fs of 1.8kHz correction along with a notch on the bass around 6kHz. Apparently it sounded good. The 1 ohm resistor is optional, and just takes down a little 2dB peak at crossover to get nearer a pure 24dB/octave LR4. Each driver gets a Zobel that matches its inductance and DCR.
Impedance Equalization (L-Pad) Circuit Designer / Calculator

For me, this was more a theoretical exercise than anything. I had become interested in the 3:1 ratio of components in classic filters. That phase alignment is quite impressive IMO. Should sound very integrated. But as I said, this sort of design is not my primary interest. 🙂

It would be much harder to do this with notches on a parallel circuit.
 

Attachments

  • system7_TimeAligned_SerialXO.jpg
    system7_TimeAligned_SerialXO.jpg
    66.3 KB · Views: 117
  • S7 Series Crossover Time Aligned.PNG
    S7 Series Crossover Time Aligned.PNG
    8.6 KB · Views: 117
  • S7 Series Crossover Freq Response.PNG
    S7 Series Crossover Freq Response.PNG
    19.5 KB · Views: 78
  • S7 Series Crossover Phase.PNG
    S7 Series Crossover Phase.PNG
    18.5 KB · Views: 71
  • S7 Series Crossover Electrical Response.PNG
    S7 Series Crossover Electrical Response.PNG
    17.2 KB · Views: 68
  • Aequal_root2_Steen_Duelund_filter.JPG
    Aequal_root2_Steen_Duelund_filter.JPG
    34.6 KB · Views: 66
Last edited:
Status
Not open for further replies.