[1st Order] Butterworth vs Solen Split.

[presscot]

Hi everyone,

I'm trying to do a low pass filter for my SEAS H602 P17REX/P. I plan to use 1st order filter with the crossover point at 4kHz.

When I've tried to search for an information about the 1st order network, I saw there are two types of 1st order, Butterworth type and Solen Split type.

So I want to know what is the different of usage for these two types ?


Thank you.

 

[presscot]

Nobody knows ? ToT

 

[john k...]

I believe it is just shifting the HP and LP corner frequencies so that the crossover point is at -6dB. It relies of offset to align the phase correctly at the crossover point but will never sum perfectly flat mathematically.

 

[presscot]

I believe it is just shifting the HP and LP corner frequencies so that the crossover point is at -6dB. It relies of offset to align the phase correctly at the crossover point but will never sum perfectly flat mathematically.

And how different of usage between them ?

My high-pass filter is 2nd order linkwitz-riley. So what type between two of them (Butterworth or Solen Split) should I use for my Lo-pass network ?

 

[john k...]

None of this matters. The bottom line is that if you are just going to put a cap on the tweeter and an inductor on the woofer you mess around with the values of each to achieve the response you want, assume you can. By the time you are done with it you probably won't have any values close to a Butterworth or the so called split Solen values. It's all nonsense. Drivers are not resistive loads the native response of the driver will always be combined with that of the filter to determine the system response. Even if you eliminate effect of the non-resistive load by going active, you still have to address the native driver responses. These low order text book and theoretical filters seldom, if ever work in practice.

 

[presscot]

Seems like it won't be too difficult anymore. 😱 haha (just kidding)

So, I want your help to decide. As I'm looking for 1st order filter to be using in my lo-pass filter at 4khz.

From the calculation, The "Butterworth" equation tells me to use L = 0.32 mH while the "Solen Split" suggests me to 0.47 mH.

And that when compared to each other, the value of 0.32 mH from "Butterworth" will provide a cross point at about 5kHz in Solen Split's table and the value of 0.47 mH from "Solen Split" will give a cross point at about 3kHz in Butterwoth's table as these link below.

Butterworth : Solen Electronique Inc.
Solen Split : Solen Electronique Inc.

So, Could anyone please tell me the expectation of the different of sound between the two values ? I mean, for example, the first one will provide more mid-frequencies and the latter will give more bass.

Thanks for insight 🙂

 

[wintermute]

I think you have completely missed John's point 🙂

What matters is the acoustic slope that results from the combination of the filter with the driver. That is not something you can get from a theoretical calculation.

It's like asking "If I meet a complete stranger and offer them an apple or an orange which one will they prefer?" 🙂

It depends on the person! and in the case of your filters, it depends on the drivers 🙂

Tony.

 

[richie00boy]

These low order text book and theoretical filters seldom, if ever work in practice.

This is the key statement. Have a look at the sticky thread in loudspeakers/multi-way about crossover design without measurement. Even that is just ball park though.

Go with the bigger inductor as it's likely to have less of a bump at the crossover frequency.

You get different crossover points with the two tables because they align at different points on the phase curve.

 

[Jay]

So, Could anyone please tell me the expectation of the different of sound between the two values ? I mean, for example, the first one will provide more mid-frequencies and the latter will give more bass.

Theoretically, the Butterworth will have more overlap around crossover frequency, so there will be "bump" at 4kHz, may be 3dB. OTOH, the other one has less bump, could be 0dB or -3dB at 4kHz.

As 4kHz is usually the location of breakups, and our ears are very sensitive to this frequency, and the filter is too simple, it is better to lower the response around this frequency, so I will prefer the "Solen Split", or even much lesser bump.

Butterworth: 0.33mH - 5.0uF
Solen Split: 0.47mH - 3.6uF
Lesser Bump: 0.50mH - 3.3uF
More Lesser Bump: 0.56mH - 3.0uF

So you get the picture from looking at the above values (to minimize bump and keep the XO frequency unaltered you want to more or less increase the L and reduce the C). They are not accurate calculation anyhow (as your drivers are not pure 8 Ohm drivers), but that's what you have. Which one that will sound best? The one that has better phase match (you wont know if you don't measure, but trained ears can hear it). It is probably "saver" to choose lesser bump as it is more forgiving.

Again, you rely on LUCK here. Luckily your woofer doesn't have too many breakup issues.

 

[presscot]

Theoretically, the Butterworth will have more overlap around crossover frequency, so there will be "bump" at 4kHz, may be 3dB. OTOH, the other one has less bump, could be 0dB or -3dB at 4kHz.

As 4kHz is usually the location of breakups, and our ears are very sensitive to this frequency, and the filter is too simple, it is better to lower the response around this frequency, so I will prefer the "Solen Split", or even much lesser bump.

Butterworth: 0.33mH - 5.0uF
Solen Split: 0.47mH - 3.6uF
Lesser Bump: 0.50mH - 3.3uF
More Lesser Bump: 0.56mH - 3.0uF

So you get the picture from looking at the above values (to minimize bump and keep the XO frequency unaltered you want to more or less increase the L and reduce the C). They are not accurate calculation anyhow (as your drivers are not pure 8 Ohm drivers), but that's what you have. Which one that will sound best? The one that has better phase match (you wont know if you don't measure, but trained ears can hear it). It is probably "saver" to choose lesser bump as it is more forgiving.

Again, you rely on LUCK here. Luckily your woofer doesn't have too many breakup issues.

Thanks for more information, Jay 🙂

But I still have a little doubt. If I plan to use a 2nd order "Linkwitz-Riley" filter on my tweeter and 1st order "Solen Split" on my woofer.

Are they compatible ? I mean, the phase of these different types and orders will cause any problems ? and Do I need to do anything else to fix the problems ? such as "Reverse polarity" of the tweeter, etc.

Thank you

P.S. My speaker is currently wired like this
Tweeter : 2nd order Linkwitz-Riley @4kHz (C = 3.3uF, L=0.47mH) with "reversed" polarity.
Woofer : no any filters or networks. (Directly wired to the amp)
So, the problems now are too much "lack of bass" and a little "rise up" at mid-frequencies. And I hope making a crossover for the woofer will help relieve these problems.

 

[richie00boy]

Just try the tweeter with both polarity and see which sounds best. As we have already said, this design is so unoptimised anyway that applying theory is useless.

 

[wintermute]

I think it would be a very good idea to have a read through the sticky on crossover deisgn without measurements http://www.diyaudio.com/forums/mult...designing-crossovers-without-measurement.html 🙂

Tony.

 

[Jay]

But I still have a little doubt. If I plan to use a 2nd order "Linkwitz-Riley" filter on my tweeter and 1st order "Solen Split" on my woofer.

Are they compatible ?

It depends on the characteristics of your drivers. You cannot be sure if you don't measure and/or simulate.

A simple way to do it is by using software. The software needs data of your drivers. To have the data you can measure your drivers, or get data from measurement of others.

So search for the FRD file and ZMA file of your woofer and tweeter, post it and someone may help you build the crossover. Better, you can download the software yourself and somebody will help you to use it.

Other than the easiest way above, you're just having fun, without any technical achievement possible.

 

[sreten]

Hi,

Design the cross over properly :

http://audio.claub.net/Simple Loudspeaker Design ver2.pdf
http://www.rjbaudio.com/Audiofiles/FRDtools.html
Designing Crossovers with Software Only

rgds, sreten.

http://sites.google.com/site/undefinition/diy
(see if nothing else, the excellent FAQs)
The Speaker Building Bible
Zaph|Audio
Zaph|Audio - ZA5 Speaker Designs with ZA14W08 woofer and Vifa DQ25SC16-04 tweeter
http://audio.claub.net/Simple Loudspeaker Design ver2.pdf
http://www.rjbaudio.com/Audiofiles/FRDtools.html
Designing Crossovers with Software Only
http://www.rjbaudio.com/
Jay's DIY Loudspeaker Projects
Speaker Design Works
http://www.htguide.com/forum/showthread.php4?t=28655
A Speaker project
DIY Loudspeakers
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

Great free SPICE Emulator : SPICE-Based Analog Simulation Program - TINA-TI - TI Software Folder

 

[PeteMcK]

re:"If I plan to use a 2nd order "Linkwitz-Riley" filter on my tweeter and 1st order "Solen Split" on my woofer" - 1st order on woofer & 2nd order LR on tweeter is a combo I've used successfully a lot, should work well with your mid-woofer, although I would cross a bit lower than 4KHz; Tweeter needs to be reverse polarity; what is your tweeter?
You'd need to work out component values using the impedance plots of the drivers, & account for BSC if needed

re:""lack of bass" and a little "rise up" at mid-frequencies" - The inductor can tame the second, the first would be fixed by correct box size & tuning