Linkwitz transform and a passive crossover?

Today Linkwitz transform is pretty well known but I've not seen anyone discussing its use with a passive crossover in the speaker.

Am I insaine to even consider this?
Consider a regular 2-wayspeaker w. 900Hz passive 2nd order crossover.
Linkwitz transform could be kicking in at 100Hz or so.
If I would implement this without regard to safety or any deeper concearn for the system in general, I could have something like a 20dB gain in the lowest octave. Let's say +10dB @ 40Hz.

The HP crossover will attenuate the signal a little over 48dB and that's not counting the HF driver's natural roll-off.

Why is it a bad idea to implement a LT on a standalone speaker with a passive crossover? Doing the numbers it would seem pretty harmless?

Yes, I'm talking about a regular passivly crossed 2-way speaker with a single power amp and an active or digital implementation of the LT between the preamp and the power amp.
 
I'm not a crossover wizzard I'll happily admit being a n0ob but let's just run with that notion.
Using 11V input instead of 2.83V.
A 30uF cap will will be some 133'ish ohm at 40Hz.
A 3uF cap will be 1330'ish ohm.

Now let's crank it up another 10dB...
Input now at 28.3V.

I'm having some serious trouble seeing how a voltage like this would pose any problem but I am a n0ob and I'm here to learn. :)

Oh, yeah... forgot to mention the sensitivity of the speaker, 96-100dB/2.83V*1m.
No need for huge voltages, a headphone amp can drive the pair, lol.
I don't have the speakers yet since I'm still in the planning stage. I am however examining different solutions to known challenges. :)
 
Last edited:
I still don't see it? He's running a active system.
He set's the monitor up as a HF unit with a HP filter.
The woofer is activly filtered with a LP + LT and you still need two amps to drive it?
It's a little unclear to me but he never uses anything but an active set-up with the LT as far as I know.
 

ezavalla

Member
2008-12-08 11:53 pm
Every speaker is a HP filter by its own transfer function, as is a closed box monitor. So, SL is shaping the frequency response from the entire box instead of shaping each loudspeaker.

Yes, I'm talking about a regular passivly crossed 2-way speaker with a single power amp and an active or digital implementation of the LT between the preamp and the power amp.
In this way, you use a single amplifier to drive your 2-way system, but now you add a LT to make it behave a little (or a lot) different in the low frequency end.

I guess it is what you try to do... :confused:
 
Yes, that's exactly it. LT to shape the low frequencies but only use one amp in total for the speaker.

Linkwitz always use an active configuration with multiple amplifiers afaik.
FAQ#15 tells us that he proposes the monitor to recieve the regular unaltered signal and the sub gets the lower frequencies with the LT.
With some skill and a little luck the monitor doesn't need any additional filtering but in most cases it will need additional filters.

And yes a CB is a 2nd order HP in itself and a BR a 4th order HP.

And we're still not adressing the issue, if it's possible to use LT on a 2-way speaker with passive crossover with no additional subwoofer and only one amplifier. :h_ache:

If I'm insaine it's ok to tell me so but I refuse to believe I'm the only one thinking outside the box?
 
Last edited:

ezavalla

Member
2008-12-08 11:53 pm
And we're still not adressing the issue, if it's possible to use LT on a 2-way speaker with passive crossover with no additional subwoofer and only one amplifier. :h_ache:
Oh... we are addressing it, but you mustn't look at the subwoofer....just forget it ;)
SL is using the LT to shape the low frequency response of the closed box as if it was a 2º order Butterworth (Qtc=0.707). Then, he adds an electrical 2º order Butterworth at the same frequency to build a 4º order acustical Linkwitz-Riley xover in order to cross the box with a subwoofer.
But you can do all the same thing without any subwoofer at all, and you also can avoid the electrical Butterworth and just let the LT to do its own thing by shaping the frequency response as you like (just a 2º order HP).
 
Ok, I see what you're getting at. He's shaping the FR for the monitor to fit the crossovercurve for the sub.
I sort of assumed there would be a difference since his example is dealing with 100Hz and I was thinking <40Hz.

But ok, it would appear to work and I'm happy. I'll just check the component ratings before doing anything stupid. :)

Being able to run the whole thing with one amp and still using a LT seems like a nice option. It gives the closed box cabinet a definate appeal. :)
 

ezavalla

Member
2008-12-08 11:53 pm
It works fine ;).
I have done it a couple of years ago in order to shape a couple of not-good-at-all speakers which have a Qtc=1.46 and fb=157Hz. I was able to put them down at Qtc'=0.71 and fb'=76Hz with an overall gain of about 10dB...but it's no safe at all to try to go down in frequency because a 6" extended range speaker is unable to handle it :smash:.. at least not without a lot of THD.
 

Jay

Banned
2003-02-11 9:02 am
Jakarta
If I'm insaine it's ok to tell me so but I refuse to believe I'm the only one thinking outside the box?

No you're not. I've done this several times. First time was before Linkwitz Transform was known, but I forgot the schematic but the idea must be similar. Later was after Linkwitz Transform, but I implemented it to match real-time conditions, as the basic LT circuit is actually flexible.

Read ESP site (I think it is a pdf or something). I've done just that based on Linkwitz. But Rod Elliot has his own name for the method. Basically it is similar to Linkwitz Transform.

One precaution is that the woofer must have high Xmax. It is difficult to control this (I mean accident can happen anytime). I have broke a 6" woofer.
 

Jay

Banned
2003-02-11 9:02 am
Jakarta
Why is it a bad idea to implement a LT on a standalone speaker with a passive crossover? Doing the numbers it would seem pretty harmless?

From my experience, the only issue is the distortion added by the circuit (in some implementation where ONE amp is used, the signal must pass this circuit before going to the tweeter). If OPAMP is used (as for the LT), why not just go build active crossover entirely??
 
Awsome. :)
I wanted to keep the thread a little more general in it's nature but if I choose to do this I'll use a DSP not a traditional active filter.
The only reason for doing it this way would be to keep things simple and economical.
One amp only cost half of two, I can keep my dac...
It's just a much smoother solution all around. :)