Guys,
Looking for some advice from past experiences using either passive Bessel or Butterworth nertworks. I'm looking for something to go into a new set of 3-ways, and I've heard many people praise the Bessel. As usual, I get the perfect -3dB cutoff and flat response of a BW when plotting it in WinISD, but by some mysterious force I've heard Bessel is the new name in the game.
Can anyone confirm/advise on this build?
Thanks,
Marc.
Looking for some advice from past experiences using either passive Bessel or Butterworth nertworks. I'm looking for something to go into a new set of 3-ways, and I've heard many people praise the Bessel. As usual, I get the perfect -3dB cutoff and flat response of a BW when plotting it in WinISD, but by some mysterious force I've heard Bessel is the new name in the game.
Can anyone confirm/advise on this build?
Thanks,
Marc.
I was reading about the difference between Butterworth and a Sub-Bessel alignment earlier over at ESP:
http://sound.westhost.com/project81.htm
col.
http://sound.westhost.com/project81.htm
col.
Hello,
While having excellent time response when only one filter is used, Bessel filters when used in group (by example low-pass + high-pass) don't retain the excellent time response properties of the isolated filters.
You may simulate the result of a complete 3 ways system, even introducing delays between loudspeakers or phase compensation, with my spreadsheet:
http://ndaviden.club.fr/outils/filtre_simul.zip
This spreadsheet gives also the group delay curve of the complete 3 ways system.
With another version of that spreadsheet you can study the square wave response of the complete 3ways system:
http://ndaviden.club.fr/outils/filtre_carre.zip
Best regards from Paris, France
Jean-Michel Le Cléac'h
While having excellent time response when only one filter is used, Bessel filters when used in group (by example low-pass + high-pass) don't retain the excellent time response properties of the isolated filters.
You may simulate the result of a complete 3 ways system, even introducing delays between loudspeakers or phase compensation, with my spreadsheet:
http://ndaviden.club.fr/outils/filtre_simul.zip
This spreadsheet gives also the group delay curve of the complete 3 ways system.
With another version of that spreadsheet you can study the square wave response of the complete 3ways system:
http://ndaviden.club.fr/outils/filtre_carre.zip
Best regards from Paris, France
Jean-Michel Le Cléac'h
MarcMTL said:
Hey,
The problem is, that you will never get near the target response.
You cant calculate crossovers. Nor the bas or tweeter are imp, frq og fase linear to start with.
You have to get you construction meassured, and then find the right falloff. It might be 16½db/oct for the bas and something different for the tweeter, in order to be in fase and sum the frq.
The problem is, that you will never get near the target response.
You cant calculate crossovers. Nor the bas or tweeter are imp, frq og fase linear to start with.
You have to get you construction meassured, and then find the right falloff. It might be 16½db/oct for the bas and something different for the tweeter, in order to be in fase and sum the frq.
Hello,
Many people are using my spreadsheet and are reporting excellent sonic results.
Here one example:
http://freerider.dyndns.org/anlage/LeCleach.htm
The goal of such spreadsheet as the one I wrote is to see what are the best results one can obtain if the loudspeakers are ideal.
If in a first time, by this kind of modelisation, you succeed in improving the response (level + group delay) of a 3 ways system based on theorically perfect loudspeakers you should surely improve the results with real loudspeakers.
Also, before to built you have to know what is the attainable ideal. Then it is up to you yo linearize the less than ideal loudspeakers you plan to use.
The initial question was about Bessel / Butterworth. Using that spreadsheet the answer to this question is more obvious than with long sentences.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Many people are using my spreadsheet and are reporting excellent sonic results.
Here one example:
http://freerider.dyndns.org/anlage/LeCleach.htm
The goal of such spreadsheet as the one I wrote is to see what are the best results one can obtain if the loudspeakers are ideal.
If in a first time, by this kind of modelisation, you succeed in improving the response (level + group delay) of a 3 ways system based on theorically perfect loudspeakers you should surely improve the results with real loudspeakers.
Also, before to built you have to know what is the attainable ideal. Then it is up to you yo linearize the less than ideal loudspeakers you plan to use.
The initial question was about Bessel / Butterworth. Using that spreadsheet the answer to this question is more obvious than with long sentences.
Best regards from Paris, France
Jean-Michel Le Cléac'h
Syncroniq said:
Hi,
I have Clio 7.3, and have build speakers for a while now. On my website you can see some of my projects, with meassurements.
http://www.speakerbuilder.dk/
Pick a project, and simulate a filter. Then i can show you exactly how it meassures, as i have lspcad and the raw meassuremens of the drivers used.
So, if you give me a crossover layout for this project
http://www.speakerbuilder.dk/content/getPage.asp?id=37
I will show you how it meassures, if build.
I have Clio 7.3, and have build speakers for a while now. On my website you can see some of my projects, with meassurements.
http://www.speakerbuilder.dk/
Pick a project, and simulate a filter. Then i can show you exactly how it meassures, as i have lspcad and the raw meassuremens of the drivers used.
So, if you give me a crossover layout for this project
http://www.speakerbuilder.dk/content/getPage.asp?id=37
I will show you how it meassures, if build.
MarcMTL,
I am a bit confused. You mention winISD which is simply for calculating how the woofer performs in it's enclosure, but has nothing to do with the frequency response needed for the crossover. So are you talking about the woofer high pass response in the enclosure or the acoustic slopes for the woofer, mid and tweeter crossovers?
If you are talking about the crossover slopes, then the real question is do you have the ability to measure the frequency response and phase of your drivers in you enclosure? If not then, the rest is academic as without measurements or some pretty heavy simulation using others measurements, you won't know what your final acoustic slopes are and how they will sum anyway and that is what matters.
Kind regards,
Dennis
I am a bit confused. You mention winISD which is simply for calculating how the woofer performs in it's enclosure, but has nothing to do with the frequency response needed for the crossover. So are you talking about the woofer high pass response in the enclosure or the acoustic slopes for the woofer, mid and tweeter crossovers?
If you are talking about the crossover slopes, then the real question is do you have the ability to measure the frequency response and phase of your drivers in you enclosure? If not then, the rest is academic as without measurements or some pretty heavy simulation using others measurements, you won't know what your final acoustic slopes are and how they will sum anyway and that is what matters.
Kind regards,
Dennis
richie00boy said:
A crossover formed from a Bessel pair (HP + LP) never sums flat perfectly. If the -3dB pointis used as the corner frequency then the phase will be +/- 74 degrees at the crossover point for a 2nd order Bessel pair and there will be a peak or dip at the crossover frequency, depending on polarity. It is possible to shift the corner frequencies of the HP and LP sections (stagger them) so that the response is 0dB at the crossover point, but there will then be ripples (or dips or peaks) to either side of the x-o point. The magnitude of the ripple will depend on the order of the Bessel pair and polariity of the connecton, but a Bessel pair never sums flat.
Re: Re: Crossover Debate: Bessel vs. Butterworth
This is not entirely correct. High pass filters cannot have flat time response. So even an isolated HP Bessel does not have a flat time delay like its low pass counterpart.
http://users.ece.gatech.edu/~mleach/papers/phasedist.pdf
Jmmlc said:
This is not entirely correct. High pass filters cannot have flat time response. So even an isolated HP Bessel does not have a flat time delay like its low pass counterpart.
http://users.ece.gatech.edu/~mleach/papers/phasedist.pdf
To talk about a crossover as if it were a simple electrical problem is simply absurd. The crossover is a complex system of electro-acoustic summations of multiple non-coincident and non-ideal sources that occurs in a full three dimensional field. The correct crossover to yield the best impulse response, both on and off axis, is never going to have a simple electrical specifcation. Its not going to look anything like an ideal "Bessel" or Butterworth" or anything else you want to call it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.