Word 4 word: Dr Bullock's 2nd-order APC (Linkwitz-Riley) 3way Xover formulas. The formula for the Gain parameter (H) completely depends on the spread of (S) of the chosen crossoverfrequencies. It is wise to place Xcover freq. no closer than 3 octaves apart which occurs when S less 8. The Gain parameter"H" is given by the formual: H=S +'a' squared + 3/S -4
Ah yes. Filters use a property called resonance. You'll especially notice this with second order filters since you can control more than just the rolloff, but the level at the 'knee' of the rolloff. This is because of the way the components interact with each other.
Not only do the components of a filter interact, but different filters interact with each other. A low pass and a high pass sometimes do more than simply their own tasks, especially when they are close in frequency. They can be designed to give a perfectly predictable result, a flat response, a reduced or increased response, or it may be peaked at the edges if you're not careful.
I've made this example, where the combined filters draw more current into the circuit than the driver would alone, and they elevate its level (here compared to having no filters) it appears to have what's sometimes called 'midrange gain' ...
Not only do the components of a filter interact, but different filters interact with each other. A low pass and a high pass sometimes do more than simply their own tasks, especially when they are close in frequency. They can be designed to give a perfectly predictable result, a flat response, a reduced or increased response, or it may be peaked at the edges if you're not careful.
I've made this example, where the combined filters draw more current into the circuit than the driver would alone, and they elevate its level (here compared to having no filters) it appears to have what's sometimes called 'midrange gain' ...
So in essence, the spacing (minimum 3 octaves) to the most extent prevents this gain? The woofer is going to be the JBL 2205B. My midrange (http://www.lansingheritage.org/images/jbl/specs/home-comp/le8t/page2.jpg) should come into play about 390Hz and pass on to tweeter about 5khz The tweeter is the LCY 103 a true foil ribbon. I've been saving for this project since 2009 when I acquired the 2205B 15" and my wheels started turning! I was undecided on how to fill the mid-section... Horn driver as a 2way or Fiata midrange, till the LE8 came available on Fbook.
I've also come to the conclusion that: though I am using a broadband woofer or broadband midrange using the filter design/crossover I can shape the response to suit my needs. This forum says Xover without measurements, would it be too much to ask about box design? I was intrigued by Onken style Box with the symmetrical ports. Your thoughts?
I've also come to the conclusion that: though I am using a broadband woofer or broadband midrange using the filter design/crossover I can shape the response to suit my needs. This forum says Xover without measurements, would it be too much to ask about box design? I was intrigued by Onken style Box with the symmetrical ports. Your thoughts?
The midrange crossover with the close frequencies does want to take that shape and that gain, but it doesn't have to. The problem is it can get a little difficult to predict. So to answer the question does it have to be 3 octaves apart? No, it doesn't, but you can be even more sure if you do some experimenting. One good way to do that is with a simulator.
You can see I used a simulator above without taking measurements. While this is not going to tell you everything, you'll find that even the band-pass style filters alone are worth examining.
Often 8" drivers can be crossed around 2kHz.
What questions do you have about the onken box? I like that it has a good baffle frontage. The distributed ports are a nice feature.
You can see I used a simulator above without taking measurements. While this is not going to tell you everything, you'll find that even the band-pass style filters alone are worth examining.
Often 8" drivers can be crossed around 2kHz.
What questions do you have about the onken box? I like that it has a good baffle frontage. The distributed ports are a nice feature.
I've been running Onken bass cabinets since 2006 with Iconic 165-8G in them. 11 cu ft + vent volume of around 1 cu ft.
This Onken calculator aligns well with the calculator I used to design mine: http://www.mh-audio.nl/Calculators/Onken.html
I am upgrading to Yuichi A-290 midrange horns which will be driven by my Radian Neo 950PB-16 drivers, I'll post some pictures at that point if anyone is interested. (It's a mess right now as I am experimenting)
This Onken calculator aligns well with the calculator I used to design mine: http://www.mh-audio.nl/Calculators/Onken.html
I am upgrading to Yuichi A-290 midrange horns which will be driven by my Radian Neo 950PB-16 drivers, I'll post some pictures at that point if anyone is interested. (It's a mess right now as I am experimenting)
Well AllenB, I have been forced out of the Onken design due to the perameters of my woofer. Even the box design would be a huge enclosure, but I am not done yet. So, after extensive trial and error then trial and success, I have realized fighting phisics is never a winning strategy! The woofer, though huge for a midrange has been designed for large venues and public address and lacks the low bass capabilities. It bested a 70Hz in my test box. With this discovory and a tweeter that can play into the lower ranges of 1.3k Hz I think this woofer is better suited for open baffle. The midrange clarity is very good and by avoiding the point where most tweeters cross (2k-Hz) I find the expected enomolies people claim to hear at crossover (2k-Hz) is not present, thus, to these ears, very pleasing. Im using BassBox Pro and Crossover Pro to simulate possible response, but have several times done the math and found it closer to what I prefer over the simulation. And still no measurments! I have settled with a 2way second order Xover system and using a powered subwoofer to fill the bottom octave. Butterworth didn't sound quite right and prefered the Linkwitz-Riley Math.... Thnk you for reintroducing me back into the Hobby. Keep it fun!
Now on to the Box, no Box challenge!
kingfisher
Now on to the Box, no Box challenge!
kingfisher
I also use pro woofers (closed box) and the fact that one might cut off at 70Hz isn't necessarily a bad thing. The room is taking over down there and using subs is good as it allows you to make choices about where they are placed. (Also the higher resonance allows higher efficiency)
For example.. don't worry about crossing the woofer, just add the sub. Experiment with its position, its level, its polarity and its crossover until your bass fills out.
For example.. don't worry about crossing the woofer, just add the sub. Experiment with its position, its level, its polarity and its crossover until your bass fills out.
I read your reply and want to thank you for your insights. I have to apologies for my miss information on my design for the 2225 JBL woofer. As I have another pair of JBL 15's they are not the same model. and I input the wrong spec to my online Onken box design. After I clarified, I see that the 2225 will work well and cut off is nearer to the low 40 Hz than previously thought.
In most of these designs 2000 Hz. crosspoint is asking a lot from this woofer even as the spec. look favorable, the graph may be smoothed. In most cases using this woofer, others are implementing a horn mid-range. Some are going without a tweeter also. For my taste I would prefer to use a tweeter.
Lets say I was to use a mid-range but not a horn, in fact it would be another model JBL, but a full range woofer. I would use this as my mid-range and roll it off at 200Hz and 5k-Hz respectively. However this is where my question lies. The 15" Woofer has an open frame as you would imagine this gives it the need for an enclosure, also something you know. The mid-range also has an open frame and it was designed to be mounted in an enclosure too! If I were to house them in the same enclosure as part of a 3 way system would I have ti isolate the Mid-range woofer in a sub-enclosure also? and how would this affect the crossover calculations?
kingfisher
In most of these designs 2000 Hz. crosspoint is asking a lot from this woofer even as the spec. look favorable, the graph may be smoothed. In most cases using this woofer, others are implementing a horn mid-range. Some are going without a tweeter also. For my taste I would prefer to use a tweeter.
Lets say I was to use a mid-range but not a horn, in fact it would be another model JBL, but a full range woofer. I would use this as my mid-range and roll it off at 200Hz and 5k-Hz respectively. However this is where my question lies. The 15" Woofer has an open frame as you would imagine this gives it the need for an enclosure, also something you know. The mid-range also has an open frame and it was designed to be mounted in an enclosure too! If I were to house them in the same enclosure as part of a 3 way system would I have ti isolate the Mid-range woofer in a sub-enclosure also? and how would this affect the crossover calculations?
kingfisher
I just installed XSim on my computer trying to figure out how to design a crossover for a 3 way system based on Dayton Audio Reference series drivers attached
So my question is has designing a rather complex crossover became this simple using software? What am I missing and what am I doing wrong?
Just to make some things clear, I'm a complete novice in this area, have basically no knowledge about electronics and acoustics and the only thing I've built so far is a soundbar based on 2 full range drivers with a simple bsc.
The whole idea behind this post is building a pair of passive floorstanders for my living room that will be driven by a Yamaha AX-870 and the audio source is my linux based PC with Audigy FX sound card.
So my question is has designing a rather complex crossover became this simple using software? What am I missing and what am I doing wrong?
Just to make some things clear, I'm a complete novice in this area, have basically no knowledge about electronics and acoustics and the only thing I've built so far is a soundbar based on 2 full range drivers with a simple bsc.
The whole idea behind this post is building a pair of passive floorstanders for my living room that will be driven by a Yamaha AX-870 and the audio source is my linux based PC with Audigy FX sound card.
Simulation's big limitation is that the driver responses from the datasheets don't take the placement onto a given baffle and cabinet into account (diffraction and baffle step loss being two big acoustic phenomena). Therefore, simulation alone is incomplete if you want real world results. The next step would be to model the expected diffraction (can also be approximated in software) or dive into measurements (needing calibrated microphone and soundcard). Measurements will also let you analyze phase and off-axis responses due to acoustic offset between the drivers. Simulation also doesn't tell you how different crossover points and slopes will sound in your room, to your ears (equivalent to the tweaking by ear sections in the thread).
Basically, the usefulness of simulation is only fully realized with measurements.
Basically, the usefulness of simulation is only fully realized with measurements.
Last edited:
Found a design thread using your drivers on htguide - as shown, I think he starts with the datasheet but ends up needing to take several measurements and iterations to crossover based on the effects of the baffle and cabinet used.
Listening and tweaking is a way to compensate for the things that are missed when designing without measurements. In three dimensions each cabinet shape sounds different. A box is just one option.
You are not asking Xsim to show phase for your drivers. Turn off system phase too. Don't ask your woofers to play into their breakup region.
