Introduction to designing crossovers without measurement

This is worth looking at, and it's fortunate because I was meaning to ask whether Toaster79 had considered the baffle step. Obviously if it isn't then when it is it would change the crossover.

These components are sometimes used to create baffle step compensation, but they may just be for shaping. There may be other ways to do the same thing. The downside is that if not done carefully they can introduce problems with low impedance, it's important that there is a resistor in there.

I haven't planned a baffle step, also don't know if I will need it. The plan is to have ports on the front side and speakers pretty close to the wall.
l would try removing R3 and C5. I am not sure the purpose of these parts.

Can you outline your driver layout and your baffle dimensions?

R3 and C5 are there to further suppress the woofers cone breakup section so it doesn't interfere with the other two drivers.

The idea is to have a baffle the size of about 250x1100-1200mm, tilted 5° in TMWW arrangement, similar to this

1.jpg
 
Thank you for posting the proposed layout.


One potential issue that you may find is vertical lobbing of the two woofers up towards the crossover point. Secondly you may find ripple in the crossover region due to the difference in the distance of each woofer from mid dome. Unless Xsim can locate your drivers on the baffle these issues won’t be obvious.
 
  • Like
Reactions: 1 user
I had a look at XSim tonight.

lts a cool diy tool to get your basic filter values in the ball park. As discussed listening and then adjustments validated with a measurement tool like REW will keep sanity in check.

What you are attempting is a fairly elaborate loudspeaker design with a few riddles in terms of what goes where.

l’m just wondering was a 95db sensitivity system your plan and a four ohm load? The thing is that level of sensitivity is more beneficial to low power valve amplifier that is going to deliver lower THD and a better damping factor with an 8 or 16 ohm load. Just my thoughts.

We are expecting some bad weather here tomorrow so l will see what l can work out for for minimum system in Leap 5 and a crossover framework that you can use as ready reckoner for adjustments. l have recently installed Leap in a new Mac Pro Laptop within a virtual XP and it needs a run through.

l will see if Jeff Bagley’s excel program can provide insights on diffraction/ baffle step and run some box simulations. For example what you could do it’s load one woofer as a critically damped seal enclosure and the other a custom bass reflex alignment. Tune it all for your room and have a good in room bass responses.

You have some nice drivers and a reasonable investment. I will write walk through on my decisions.

None of this replaces critical listening in terms of assessing what it sounds like. As discussed early listening and intervention can often lead to subjectively successful outcomes.
 
Hi everyone.

Please find below a series of tutorial posts running though the things we like to look at in diy loudspeaker projects such as raw driver response curves, thinking about crossover points, thoughts on how it will all come together and sorting out your crossover design.

This is diy. Which means we get to pick and choose our own drivers which can be in many cases better than those in some consumer designs due to cost limitations in production. However, for many of us what really is the best way to go about using these nice drivers?

Some questions come to mind?
How do you design a crossover network?
What needs to be considered?
How do we weigh up our best options?
Will my loudspeaker be amplifier friendly?
What sort of design is going to work best in room?
How efficient does my loudspeaker project need to be?

Lots of questions starting spinning around in our heads and we surf the www looking for answers, opinons and feedback.

Toaster79 has already chosen his drivers for his project so what we are going to do is look into the nitty gritty and work it out.

Please post questions. This is what this is all about.

Starting with the driver information fortunately Dayton publish driver data for us to use.

Please are the raw published curves for Toaster79 project
 

Attachments

  • Daytonwooferresponsetutorial.JPG
    Daytonwooferresponsetutorial.JPG
    208.1 KB · Views: 74
  • DaytonMidrangeresponsetutorial.JPG
    DaytonMidrangeresponsetutorial.JPG
    206.2 KB · Views: 78
  • Daytontweeterresponsetutorial.JPG
    Daytontweeterresponsetutorial.JPG
    206 KB · Views: 69
Hi everyone.

Please find below a series of tutorial posts running though the things we like to look at in diy loudspeaker projects such as raw driver response curves, thinking about crossover points, thoughts on how it will all come together and sorting out your crossover design.

This is diy. Which means we get to pick and choose our own drivers which can be in many cases better than those in some consumer designs due to cost limitations in production. However, for many of us what really is the best way to go about using these nice drivers?

Some questions come to mind?
How do you design a crossover network?
What needs to be considered?
How do we weigh up our best options?
Will my loudspeaker be amplifier friendly?
What sort of design is going to work best in room?
How efficient does my loudspeaker project need to be?

Lots of questions starting spinning around in our heads and we surf the www looking for answers, opinons and feedback.

Toaster79 has already chosen his drivers for his project so what we are going to do is look into the nitty gritty and work it out.

Please post questions. This is what this is all about.

Starting with the driver information fortunately Dayton publish driver data for us to use.

Please are the raw published curves for Toaster79 project
In the above curves we have one axis response, 15 degrees off axis, 30 degrees off axis and 45 degrees off axis.I have tried to illuminate the various off axis curves with a consistent colour legend for clarity.

If I put them all on the one attachment you would go crazy.

Starting with the woofer you can see things start to go crazy above 800 hertz and around 3 khertz we see cone break up.
Most of us have perhaps heard the term cone break up. What it is is the cone actually flapping and self resonating as opposed to the cone moving in and out as a rigid piston. That's not going to sound very nice. So ideally we need to make sure we can't hear that cone break up.

This means we need to start thinking about the woofer crossover point? Technically it needs to be below the point where the response becomes erratic. The reason is that we want to have a smooth crossover region. That at least an octave either side of the crossover point.

After thinking about I decided to make the crossover point at 600 hertz - just for this exercise. The other important thing to understand is the driver sensitivity which I have referenced to 2.83 volts (1 watt into 8 ohms).
 
In the above curves we have one axis response, 15 degrees off axis, 30 degrees off axis and 45 degrees off axis.I have tried to illuminate the various off axis curves with a consistent colour legend for clarity.

If I put them all on the one attachment you would go crazy.

Starting with the woofer you can see things start to go crazy above 800 hertz and around 3 khertz we see cone break up.
Most of us have perhaps heard the term cone break up. What it is is the cone actually flapping and self resonating as opposed to the cone moving in and out as a rigid piston. That's not going to sound very nice. So ideally we need to make sure we can't hear that cone break up.

This means we need to start thinking about the woofer crossover point? Technically it needs to be below the point where the response becomes erratic. The reason is that we want to have a smooth crossover region. That at least an octave either side of the crossover point.

After thinking about I decided to make the crossover point at 600 hertz - just for this exercise. The other important thing to understand is the driver sensitivity which I have referenced to 2.83 volts (1 watt into 8 ohms).
Next up the mid dome driver frequency response curves on axis and off axis per the woofer.

Lots of interesting things going on here. Firstly this driver has a similar sensitivity to the woofer. That's important because it needs to be able to match the sensitivity of the woofer to provide a natural tonal balance. You can see the on and off axis response is fairly even up to about 3 hertz with some variations. Remember we are looking at graph divisions of 5 db. 1 db is only just discernible, 2 db is discernible. Above 3 khertz here are a few bumps on axis but we are only talking about 2 db variance.

What to do about this information? Firstly lets figure out the crossover point to the woofer. This dome midrange has a resonance of 341 hertz. We don't want to be hearing that. Does power distribution matter in terms of your crossover point. Only where you expect to use the system at or near its rated acoustic output continuously. The X max displacement at low frequencies need to be considered and the thermal rating of the driver. With a 600 hertz crossover point only 25-30% of the overall electrical power expressed in watts as an average is going to be seen by the mid range driver.

Next the high frequency crossover point. 2k hertz is in the middle of the sensitivity midrange frequencies so we have to go higher. Looking at 3.2 Khertz the off axis 30 and 45 degree off axis response curves track so we have a defined off axis response in the horizontal plane.

In the vertical plane things are okay as the on an off axis response of the woofer and mid dome are symmetrical. With the high frequency crossover point to the tweeter the thing that matters is the Design Point for the measurements on axis. The Design Point is normally the Listening Height and is normally on axis with the tweeter. At a listening distance of 2 metres or greater the a variation of 15 degrees is not going to effect the response in the crossover region.

What we can do to minimise and disturbance in the crossover is to look at a steep crossover slopes. We will take about that soon.
 
  • Like
Reactions: 1 user
Next up the mid dome driver frequency response curves on axis and off axis per the woofer.

Lots of interesting things going on here. Firstly this driver has a similar sensitivity to the woofer. That's important because it needs to be able to match the sensitivity of the woofer to provide a natural tonal balance. You can see the on and off axis response is fairly even up to about 3 hertz with some variations. Remember we are looking at graph divisions of 5 db. 1 db is only just discernible, 2 db is discernible. Above 3 khertz here are a few bumps on axis but we are only talking about 2 db variance.

What to do about this information? Firstly lets figure out the crossover point to the woofer. This dome midrange has a resonance of 341 hertz. We don't want to be hearing that. Does power distribution matter in terms of your crossover point. Only where you expect to use the system at or near its rated acoustic output continuously. The X max displacement at low frequencies need to be considered and the thermal rating of the driver. With a 600 hertz crossover point only 25-30% of the overall electrical power expressed in watts as an average is going to be seen by the mid range driver.

Next the high frequency crossover point. 2k hertz is in the middle of the sensitivity midrange frequencies so we have to go higher. Looking at 3.2 Khertz the off axis 30 and 45 degree off axis response curves track so we have a defined off axis response in the horizontal plane.

In the vertical plane things are okay as the on an off axis response of the woofer and mid dome are symmetrical. With the high frequency crossover point to the tweeter the thing that matters is the Design Point for the measurements on axis. The Design Point is normally the Listening Height and is normally on axis with the tweeter. At a listening distance of 2 metres or greater the a variation of 15 degrees is not going to effect the response in the crossover region.

What we can do to minimise and disturbance in the crossover is to look at a steep crossover slopes. We will take about that soon.
 
With the tweeter we have some variations in the overall flatness that needs to be considered. This can be addressed on the crossover network by building TILT into what is referred to as the voltage drive or voltage curve going to the tweeter.

We can see the on axis and the 15 degree off axis tracks well with the on axis response and the 30 degree response holds up well being only 2 db down and 10 hertz. There are no issues with the proposed 3.2 khertz crossover point to the mid dome. Level matching to the mid dome driver will need be applied. This dome tweeter is a 4 ohm ohm driver..
 
I have attached below the driver impedance curves so we can see the variations. Note that a driver with an 8 ohm labelled impedance is never strictly an 8 ohm driver. The significance of this is that we need to account for the actual impedance and the variations in the impedance when we calculate our crossover network filters .

I have posted the overall system impedance so you can see what happens with the network in place.
 

Attachments

  • DaytonImpedancetutorial.JPG
    DaytonImpedancetutorial.JPG
    180.7 KB · Views: 63
  • Daytonindividualdriverimpedancetutorial.JPG
    Daytonindividualdriverimpedancetutorial.JPG
    159.8 KB · Views: 59
  • Like
Reactions: 1 user
Now we need to scratch our heads and figure out exactly where the drivers are going to be placed on the baffle.

Toaster79 has already outlined his proposed concept. I have take this further and carefully placed the drivers on the baffle with real measurements of the distances between the drivers. The reason for doing this is that for the crossover slopes to work correctly we need to know the distance between the drivers. This is because the phase response of the crossover slope is dependant on the distance between the drivers. So it needs to be defined. The difference in the alignment of the driver voice coils also needs to be defined. Together with the lateral spacing of the drivers we call this the X, Y, Z co-ordinates.

Ideally the closer the tweeter is the the mid range dome the better.

Next we have to think about crossover slopes. In a two way system where there is at least one octave or more of response overlap between the woofer and the tweeter you can use gentle crossover slopes. Its much simpler than the steeper crossover slopes to calculate parts values for and they can sound really good. However, in a three system with the crossover points only just over two octaves apart we need to use steeper slopes so the operating bandwidth of the midrange driver is clearly defined.

For this reason I decided to use Linkwitz Riley 4th order crossover slopes. A really important thing is to understand that this refers to the acoustic response of the driver. Not the electrical response. A lot of people mistake LR filters for the electrical response. This only ever applies with text book drivers which in practise don't exist.

The crossover network is if you like the engine room of the loudspeaker system. Its like the conductor and decides what goes where.

So what really matters?

Matching the driver levels - you can do this by ear as a diy exercise or use the data provided by Dayton Audio as in this case or you can use REW to make your own measurements. The later is not hard to do. Listening is the most important thing to do. The reason is that even the best simulations rely on assumptions and measurements that may not be applicable to your enclosure baffle and room placement of your diy loudspeaker system.

Correctly functioning crossover filters - you can attempt a cookbook design if you can measure your driver impedance with for example a DATS3 loudspeaker tester. This will save you a lot of time making adjustments when using RC and RL conjugate networks for impedance soothing which will make manual calculations work much better in practise. Then if your driver has a known smooth response your crossover slope will be reasonably close. The problem however will be your baffle and what it does to the smooth response of your drivers. The way to test your crossover is to reverse the polarity of one driver and listen for a null in the sound. This can easily be done with a measurement using REW.

Consider your driver frequency response characteristics per my earlier posts. Early listening in your loudspeaker design is the key. Trust your ears. If it does sound right try something different. What I have stimulated here will require careful listening and most likely some adjustments. Your early listening will instantly tell you if you have enough bass, if the tweeter is too bright or the mid range is too forward or recessed.

Please see the attached simulated crossover response using the principles discussed.

How did I do it? I used the power of the computer to produce this simulation. I have not introduced the 2nd woofer to keep the simulation as simple as possible. Adding the 2nd woofer causes a lot of phase grating in the response with the 600 hertz crossover point. A much lower crossover point of 200- 250 hertz would be required to operate both woofers over the same bandwidth. The possibility exists for adding the 2nd woofer as helper woofer over a limited bandwidth using a 1st order low pass filter if a baffle step compensation is required. Dual 8 ohm woofers would too sensitive unless wired in series. That might work in a D'Apolito arrangement with a woofer above and below the mid cone and a wider baffle to mount the tweeter beside the mod dome. I have such as system and it is exceptional.

I hope those following this thread find these posts useful.

Ian
 

Attachments

  • Daytonbafflelayout.jpg
    Daytonbafflelayout.jpg
    28.8 KB · Views: 96
  • DaytonFRresponseTutorial.JPG
    DaytonFRresponseTutorial.JPG
    209.2 KB · Views: 84
  • Daytonvoltagedrivetutorial.JPG
    Daytonvoltagedrivetutorial.JPG
    199.2 KB · Views: 108
  • 822B7475-77A8-4015-9710-E55CC34869C7.jpeg
    822B7475-77A8-4015-9710-E55CC34869C7.jpeg
    743.6 KB · Views: 95
Last edited:
  • Like
Reactions: 1 user
Moving on to more engine room stuff it’s often a good idea to plan your diy loudspeaker with the woofer. The woofer does all the heavy lifting and it’s low frequency capability will underpin the overall efficiency and tonal balance of your design.

So how do you select your woofer ?

No one woofer is going to give you everything on your wish list such as low bass extension, high sensitivity, smooth response past 3 k hertz and good off axis response.

Looking at the maximum low frequency acoustic output which is 106 db at 40 hertz a larger or multiple drivers will increase the output. The trade off is enclosure size, complexity and cost. Sometimes our egos want bigger. But bigger also invites the WAF.

A dedicated sub woofer or multiple subs will give you far better low frequency output and extension than any floor standing design.

Typically a 6 1/2 inch woofer or bass/ mid driver will be easier to crossover at 3Khertz than an 8” woofer. An 8-10@ woofer will typically crossover over at 1.8-2.4khertz.

Lots of modern tweeters will function down to 2 khertz. But distortion starts to rise with lower crossover frequencies and matching the subjective sound of a woofer to a tweeter is important.

All woofers are governed by the laws of physics. We all dream of feeling 30 hertz but the reality is a 6 1/2”-8” woofer is best for a 45-50 hertz output at full power. A floor wall boundary can increase output below 100 hertz by +6 db. This can equate to an extension down to 35 hertz in a correctly tuned design. But don’t expect a 6 1/2” woofer to rattle your windows. So consider your room placement and take advantage of the low frequency extension from room placement and rear facing ports.

As a guideline the smaller the woofer the lower it’s sensitivity will be and the less bass output level it will be capable of at full power. Always check the maximum acoustic output at low frequencies. Not the small signal output.

There are exceptions such as subwoofers that are specifically designed for low frequency outputSome of the new Purifi woofers are setting new performance bench marks in low frequency output.

Below are some screenshots from a simple smart phone loudspeaker box designer called Speaker Box Lite. It is pre loaded with a lot of driver data from many different brands.

These curves are a simulation of the Dayton woofer selected by Toaster79.


Looking at the maximum low frequency acoustic output which is 106 db at 40 hertz a larger or multiple drivers will increase the output. The trade off is enclosure size, complexity and cost. Sometimes our egos want bigger. But bigger also invites the WAF.

A dedicated sub woofer or multiple subs will give you higher low frequency output and lower extension than any floor standing design.
 

Attachments

  • 67EF0927-A63A-4A62-81C6-AB5E8401C684.jpeg
    67EF0927-A63A-4A62-81C6-AB5E8401C684.jpeg
    95.8 KB · Views: 55
  • 78D13365-7183-4839-B03C-DE9243A29377.jpeg
    78D13365-7183-4839-B03C-DE9243A29377.jpeg
    100.2 KB · Views: 49
  • 962E770A-3808-4178-B1D7-57B7ADFE5CA6.jpeg
    962E770A-3808-4178-B1D7-57B7ADFE5CA6.jpeg
    92.5 KB · Views: 49
  • C65B183D-763B-4371-955B-C354CACA241C.jpeg
    C65B183D-763B-4371-955B-C354CACA241C.jpeg
    90.8 KB · Views: 55
  • B0925E48-FF37-47C2-A8FC-D61640442A9A.jpeg
    B0925E48-FF37-47C2-A8FC-D61640442A9A.jpeg
    122.1 KB · Views: 61
Last edited:
  • Like
Reactions: 1 user
Moving on to more engine room stuff it’s often a good idea to plan your diy loudspeaker with the woofer. The woofer does all the heavy lifting and it’s low frequency capability will underpin the overall efficiency and tonal balance of your design.

So how do you select your woofer ?

No one woofer is going to give you everything on your wish list such as low bass extension, high sensitivity, smooth response past 3 k hertz and good off axis response.

Looking at the maximum low frequency acoustic output which is 106 db at 40 hertz a larger or multiple drivers will increase the output. The trade off is enclosure size, complexity and cost. Sometimes our egos want bigger. But bigger also invites the WAF.

A dedicated sub woofer or multiple subs will give you far better low frequency output and extension than any floor standing design.

Typically a 6 1/2 inch woofer or bass/ mid driver will be easier to crossover at 3Khertz than an 8” woofer. An 8-10@ woofer will typically crossover over at 1.8-2.4khertz.

Lots of modern tweeters will function down to 2 khertz. But distortion starts to rise with lower crossover frequencies and matching the subjective sound of a woofer to a tweeter is important.

All woofers are governed by the laws of physics. We all dream of feeling 30 hertz but the reality is a 6 1/2”-8” woofer is best for a 45-50 hertz output at full power. A floor wall boundary can increase output below 100 hertz by +6 db. This can equate to an extension down to 35 hertz in a correctly tuned design. But don’t expect a 6 1/2” woofer to rattle your windows. So consider your room placement and take advantage of the low frequency extension from room placement and rear facing ports.

As a guideline the smaller the woofer the lower it’s sensitivity will be and the less bass output level it will be capable of at full power. Always check the maximum acoustic output at low frequencies. Not the small signal output.

There are exceptions such as subwoofers that are specifically designed for low frequency outputSome of the new Purifi woofers are setting new performance bench marks in low frequency output.

Below are some screenshots from a simple smart phone loudspeaker box designer called Speaker Box Lite. It is pre loaded with a lot of driver data from many different brands.

These curves are a simulation of the Dayton woofer selected by Toaster79.


Looking at the maximum low frequency acoustic output which is 106 db at 40 hertz a larger or multiple drivers will increase the output. The trade off is enclosure size, complexity and cost. Sometimes our egos want bigger. But bigger also invites the WAF.

A dedicated sub woofer or multiple subs will give you higher low frequency output and lower extension than any floor standing design.
 
A little bit about me.

I arrived at Diyaudio.com back in the early 2000’s and built a number of Pass Aleph diy Aleph amplifiers including the BOSZ preamp and the Moamps active crossover. I have had a long interest in diy loudspeakers and built a clone of the famous JBL 4343 4 way monitor back in 1979. In the past 20 years l have contributed to the popularity of JBL legacy 4 way monitors and provide technical advice on upgrading and building clones of these studio monitors to fanatics in the EU, Asia Pacific and North America. More recently in the past few years l have collaborated with Greg Timbers (retired JBL chief engineer), Nelson Pass and Douglas Self in the design of custom analogue active crossovers for pre determined loudspeaker systems.

l am semi retired and live on the Sunshine Coast in Queensland Australia.
 
  • Like
Reactions: 1 user
Some useful tool box kinks for diy loudspeaker builders.

http://www.sengpielaudio.com/calculatorVoltagePower.htm

The above link has some very useful calculatiors for determining voltage and power and power gain.

https://sound-au.com/articles/enclosures.htm

Elliott Sound pages needs no introduction as one of the most credible sources of factual explanation on all things diy audio.

http://www.mh-audio.nl/Loudspeaker.html

Hi Fi Audio design has evolved to become one of the most comprehensive diy audio design links with numerous tool box calculators.

These tool box links give you the diy loudspeaker builder a degree of certainty in designing, planning and execution of your projects.

A few thoughts on graphs and measurements from a highly experienced loudspeaker designer. Scroll down to the sun heading on measurements.

http://www.troelsgravesen.dk/The-Loudspeaker.htm
 
Where do l sit on the spectrum?

My own system uses an 18” woofer. But l do listen to other systems. I am not suggesting this is the only avenue to diy audio bliss. Far from it. A good loudspeaker if often about an elegant and simple design in its execution.

https://audio-database.com/JBL/speaker/4345-e.html

Taking a leaf from Troels Gravesen my own experience is that driver efficiency is an important factor in appreciation of life like sound reproduction.

Why does a bookshelf loudspeaker sound like a bookshelf loudspeaker (except maybe an ATC powered monitor)?

Efficiency is about the ratio of power conversion to acoustic output and heat. For every 3 db increase in driver sensitivity you need 50% less amplifier power for the same power level. Heating of the voice coil results in an increase in the dcr of the coil. The result is power compression.

Linearity is also important at low frequencies. The more linear our loudspeaker is at low frequencies (below 100 hertz) the bigger it will sound and bass instruments will sound more life like. Don’t be afraid of employing a diy subwoofer(s) to improve the linearity of your system. You can then focus your design on higher overall sensitivity. Low Frequency Linearity is the ability to hold up the spl at low frequencies as level increases without dynamic or thermal limitations.
 
  • Like
Reactions: 1 users
Touching on the subjectivity of a loudspeaker please don’t under estimate the impact of cone diaphragm materials and fabrication techniques on what you actually hear.

A long time ago there was a great article in Sound Practices that described the key determining factors effecting a loudspeaker diaphragm’s behaviour.

It goes something like this.

Stiffness
Self Damping
Low density

Here is a great link on cone fabrication materials and performance.

https://audioxpress.com/article/speaker-cones-fabrication-materials-and-performance

Without understating it these things have a significant influence on what your ears like and don’t like.

This link is a good summary of technical terms of a loudspeaker driver.

https://eminence.com/pages/support__understanding-loudspeaker-data

Everything about a loudspeaker driver is a set of compromises. Finding a complimentary set of compromises is the key.

The choice of drivers is entirely yours. Over the years paper, polypropylene, mineral loaded polypropylene, Kevlar, carbon fibre, aluminium, titanium, silk, beryllium and blends of synthetic fibres have been used to fabricate diaphragms. The complete list is much longer.

Various coatings can be applied to modify the characteristics of a cone or tweeter diaphragm.

Let your ears be the judge. Not your eyes!

When ever l can l find it enlightening to listen to un amplified acoustic instruments.

The piano, a trumpet, drums, cymbals, acoustic guitar, double bass and many other instruments have a unique sound. Identifying what gives you your emotional connection to a musical instrument and focusing on that in your diy loudspeaker will bring you joy.
 
  • Like
Reactions: 1 user
I was looking back at the start of this thread and it occurred to me that one way of making quick subjective crossover adjustments without measurements is to make a crossover switch box.

The ability to be able to make quick comparisons when tweaking your crossover is vitally important. Once you start tweaking or design by trial and error it can get confusing once you change two or more crossover values.

Even it you decide to use crossover network software like Xsim a crossover switch box allows you to try some options you would not otherwise be able to subjectively audition.

Sometimes if your lucky everything falls into place. But it’s nice to be able to make quick subjective comparison with some adjustments to confirm your design is working as intended.

The crossover switch box is a bunch of inductors and capacitors of different values that can be selected with a rotary switches for different crossover points. This can be arranged for both high and low pass filters. With the inclusion of the tweeter resister and the woofer impedance compensation calculations as previously discussed by AllenB you can have a number of adjustments to evaluate. Lpads will allow continuous adjustment to the levels.

During the week l will look at a two way loudspeaker and provide a schematic with the parts values.

To reiterate always listen before you set your loudspeaker design in stone. Rarely will a design on paper be ready for musical enjoyment with without some empirical adjustments.
 
  • Like
Reactions: 1 user
Erratum.

In my previous posts I advised that I was unable two obtain a satisfactory outcome for a dual helper woofer where baffle step could be a problem with a 650 hertz crossover point.

On looking back at my work today I discovered an error in the simulation! This is the problem with simulations in that sometimes you can make blind mistakes and then dismiss what could be a solution to your design.

So please find attached some graphics exported from Leap 5 with the 2nd helper woofer used to level off the loss in bass due to baffle step. The baffle step was simulated using a diffraction simulator for this example.

The relevance to this thread is that you would incorporate a 2nd woofer and empirically design a 1st woofer crossover using listening tests in your room. A text book calculation on an initial inductor value used in the helper woofer low pass filter is quite acceptable. You can modify the response using a damping resister between 100R and 22R to smooth the Summed overall response. I used a DCR od 2 ohms for a 12mH inductor. This is quite acceptable. Please note what you should use a seperate woofer chamber for the helper woofer as the different voltage drivers to each woofer could result in non linear distortion.

Note : In these curves I used an electrical -6 db LF shelf filter to simulate the effect of baffle step in the SPL response.
 

Attachments

  • Daytonhelperwoofersystemspl.JPG
    Daytonhelperwoofersystemspl.JPG
    352.3 KB · Views: 71
  • DaytonHelperwoofersystemimpedancefile.JPG
    DaytonHelperwoofersystemimpedancefile.JPG
    313.8 KB · Views: 66
  • 4939E49D-30AF-401F-ACD2-E33420D452C2.jpeg
    4939E49D-30AF-401F-ACD2-E33420D452C2.jpeg
    677.8 KB · Views: 56
  • 6C131267-371E-45EE-ABDB-6866C0F7A79C.jpeg
    6C131267-371E-45EE-ABDB-6866C0F7A79C.jpeg
    553.1 KB · Views: 65
  • Like
Reactions: 1 user