Linkwitz-Riley slopes and crossovers

[Andres Silva]

Please help:

I am a little confused.

I read in Zaph audio he only "use a couple of component to combine with the driver's natural rolloff and reach 4th order target slopes" in his 4th order L-R crossovers. As you may see I'm new on this, and in my knowledge the 4th order crossover have 2 coils and 2 caps, to get 4th order slopes. But I see there is other way to get 4th order slopes with less components. Were I can read how to get a Linkwitz-Riley crossover slope with less components?

Regards,

Andres

 

[Radugazon]

ah ah...for starting look at this fresh and great post, btw you can read the thread too

 

[Andres Silva]

Thanks to you Radugazon! I understand the topic, and I had read the thread. But now. How can I, or where can I read about LR24 slope crossovers? Because is easy read about LR4 (2 caps 2 coils), but I havent find how to calculate how to made LR24 crossovers.

Thanks your you help...

 

[Davey]

Andres,

LR4 and LR24 are the same thing. 4th-order or 24db/octave are just different ways of describing the roll-off rates.

Speaker system crossovers are defined by the acoustic responses created. The electrical response of the component network PLUS the raw acoustic response of the drivers themselves yield the result. So, an electrical crossover constructed with 2 coils and 2 caps doesn't necessarily yield a 4th-order L-R (acoustic) crossover. In fact, it almost never would.

A speaker designer starts with measured acoustic responses for the drivers mounted on the appropriate baffle, then decides what type of crossover is the objective, and then assembles the components to try and meet that objective.
As an example.....many times you'll see a 4th-order acoustic crossover created with just 1 coil and 1 capacitor in each filter network. But there are many variations.

Hope that helps.

Cheers,

Dave.

 

[sreten]

Hi,

Zaphs site is the perfect place to understand the difference
between acoustical target functions and the nonsense
peddled by crossover calculators.

rgds, sreten.

http://sites.google.com/site/undefinition/diy (see FAQs)
Zaph|Audio
FRD Consortium tools guide
RJB Audio Projects
Speaker Design Works
HTGuide Forum - A Guide to HTguide.com Completed Speaker Designs.
DIY Loudspeaker Projects Troels Gravesen
Humble Homemade Hifi
Quarter Wavelength Loudspeaker Design
The Frugal-Horns Site -- High Performance, Low Cost DIY Horn Designs
Linkwitz Lab - Loudspeaker Design
Music and Design

 

[wintermute]

Hi Andres, As Dave says measurements on the final baffle (both FR and impedance) are what you really need. Once you have those, get a crossover simulation program and experiment. Jeff Bagby's PCD or Speaker Workshop are both good. PCD is a good starting place.

Sim some simple crossovers, and then implement them and do the measurements, check that they match what the sim predicted. If your measurements are good then the actual vs sim should be pretty close.

Trying different combinations of cap and coil will give you an idea of how the various components interact. Pay attention to what happens not only to the FR but also the phase.

Both the mentioned tools allow you to put in a target response (say 4th order L/R at 3Khz and 89db) you then try and get your measured response to fit that curve, by varying the components in the crossover. You can start with say a stock 2nd order LR filter and then add, remove, vary components, experimenting to see what has what effect until you get something approaching your goal.

Not everything will work as expected, and sometimes strange results can occur. as a sanity check implement a simple crossover and measure the results, compare to your simulation of the same circuit.

Tony.

 

[Andres Silva]

Thanks sreten, Dave and Tony. In fact I am in the lerning curve (climbing hymalaya actualy!) of the Speaker Workshop, a did several measurements but I not sure if the acoustic ones are fine. My disign is a modified version of the Roman Bednarek's Asterion louspeaker and hibridized with Tony Gee's cup-a-sup scaled up size box. I use my gated in axis tweeter and woofer in my box curves and impedance curves. However when Mr. Bednarek say he uses a 4 order LR crossover and I see only 2 caps and one coil I think "this is not a 3th ordre crossover????". Now I see he is talking about a 24db slope. I use the crossover of mr Bednarek and "play" in the Speaker Workshop modifying the crossover to get a nice flat frequency response, but until now I am learning. I am using right now an Asterion's modified crossover with with my boxes and drivers and I get very good results to my ears. But I'm not sure these are good crossovers.
Don't confuse, I don't do the boxes, I ask to do the boxes to a very good carpenter.

Thanks for your help.

PS Now is late, in soon I will post my gated curves and let me know your thoughts. I will apreciate rour help.

An externally hosted image should be here but it was not working when we last tested it.