Introduction to designing crossovers without measurement

It's difficult to be sure with single measurements. What I would probably do at the start is try to ignore the peak near 800Hz, and listen to adjust the tweeter resistor. Set that value in the simulator and try to keep it there for the next stage.
I did tried a new measurement keeping the enclosure bit more higher from floor, however this time I am getting some dips, i took around 3 measurement and all of the showing similar. Moreover I get the same dip with tweeter has well around 1kh region
 

Attachments

  • RS180-8-31-07-2022.jpg
    RS180-8-31-07-2022.jpg
    95.9 KB · Views: 72
Hi Everyone,

Image you plan to build your next diy loudspeakers without measurements using Dayton drivers.

Most of the Dayton drivers have a smooth response across their operational pass band which means you could adopt a crossover without measurements using the approach as outlined by AllenB in this thread. This intuitive approach lends itself to listening and trying different things.
Please find below some crossover curves I created with 2nd and 4th order LR filters and 3rd order Butter-worth filters.
The crossover frequencies are 2,000 Hz, 2,400 Hz, 3,000 Hz, and 3,500 Hz.

By looking at the driver response curves published on the Dayton Audio website and the crossover curves attached below you can get an idea of what might happen. But the key as we know is in thelistening.

So over the past week I have given some thought to a switchable passive crossover box. The concept of the crossover box is to enable switching of different low pass and high pass crossover frequencies on the fly during subjective assessments.

Looking at the practicality of such an switch box I looked on the Parts Express website at low cost crossover parts sand at suitable low cost Lorlin 2 pole 6 position switches.

Using the available poles and positions of the switch I came up with the arrangement below. Referring to the simplified schematic each filter has an input and an output Node. I decided to use a seperate Lorlin switches for the input switching and the output switching of the low pass and high pass filters.

Grouping the filters I chose some nominal crossover frequencies which are common for a two way loudspeaker with a woofer and a tweeter.

Low Pass Filter Group

Switch One positions 1-6 Filter Input

3rd order 2000 Hertz, 2400 Hertz, 3000 Hertz, 3500 Hertz
4th order 2000 Hertz, 3000 Hertz

Switch Two positions 1-6 Filter Output

3rd order 2000 Hertz, 2400 Hertz, 3000 Hertz, 3500 Hertz
4th order 2000 Hertz, 3000 Hertz

High Pass Filter Group

Switch Three positions 1-6 Filter Input

3rd order 2000 Hertz, 2400 Hertz, 3000 Hertz, 3500 Hertz
4th order 2000 Hertz, 3000 Hertz

Switch Three positions 1-6 Filter Output

3rd order 2000 Hertz, 2400 Hertz, 3000 Hertz, 3500 Hertz
4th order 2000 Hertz, 3000 Hertz

Looking at the Dayton driver response data across the range I chose low pass crossover frequencies of 2000 hertz and 2,400 hertz for larger 8" or 9" woofers and 3,000 hertz and 3,500 hertz for smaller 6 1/2" woofers. The tweeter has identical high pass crossover frequencies.

I then chose 3rd order BW slopes and 4th order LR slopes. These slopes are more satisfactory for filtering out problematic woofer break up modes and a tweeter resonance frequency below the crossover point. The crossover region transitions are smaller and tend to have less on axis and off axis anomalies that 1st and 2nd order filters have.

In use the user will need to refer to the woofer Re and Le to calculate the conjugate RC values for the woofer rising impedance to ensure the LP filter functions correctly.

To spice things up I have also included a filter damping modification so compensate for a rising woofer or tweeter response in the crossover region. To be posted at a later date.
Variable 8 ohm L pads available from Parts Express can be added on the output of the tweeter high pass filter.
Assembling such a switch is perhaps best done on a a sheet of plywood using hotmelt glue to fasten the inductors and capacitors.
Flying leads,tag strips and liberal use of cable ties can connect the filters components to the respective switch contacts.


If there is sufficient interest we could put together a parts and assemble thread. I will post a more detailed schematic.


Ian
 

Attachments

  • Crossoverboxvoltagedrives.JPG
    Crossoverboxvoltagedrives.JPG
    373.1 KB · Views: 117
  • Crossoverboxschematic.JPG
    Crossoverboxschematic.JPG
    457.1 KB · Views: 116
  • Like
Reactions: 1 user
What help are you looking for?

If you slide the gate left, is there a point where the treble comes good?
I would like to understand if the measurement taken by me are correct? why I am getting the dips, wherein same driver in similar enclosure over the internet is flat till 4khz. Attaching the Tweeter (orange) and Woofer (red) raw SPL from REW
 

Attachments

  • TWEETER-REW.jpg
    TWEETER-REW.jpg
    224.4 KB · Views: 69
  • WOOFER-REW.jpg
    WOOFER-REW.jpg
    208.4 KB · Views: 63
I have a friend who has blown out his 15" sub in a horn ported box that is on the order of 4.21 Ft^3. He wants to replace with Rockville RVP15W8 and he is swapping to the Pyle 432 2" tweeter under the horn. This is a 20 year old box and they used a single pole crossover before with a crude Cap and Inductor configuration. I can't tell what model of sub it was but looks like a emenence 15" sub and an emenece tweeter. My question is I am trying to help him build a 2nd order crossover and I am not sure what or where to start with the cross over frequency, the RVP15W8 has a FR of 25 Hz to 1.5 KHz and the tweeter is 400 Hz to 9 KHz. Should the Crossover frequency be in the middle of the overlap of these two ranges between the sub and tweeter? I was thinking 900-1100 Hz range. Any help would be greatly appreciated. Also these speaker boxes are used solely outside at running events so not in a typically quality sound environment or audience.
 
diyAudio Moderator
Joined 2008
Paid Member
You've chosen a speaker with some different and special requirements.

With a speaker like this it's up to you to make sure the horn (not the compression tweeter but the horn it is attached to) is suitable to be crossed to the 15" woofer, otherwise you may have coverage problems with PA or balance problems with hifi. Normal crossover frequency for hifi use is between 700-1200Hz and your choice there is good. PA use is sometimes chosen based on output/power instead.

Using a single capacitor on the compression tweeter can run you into trouble with the impedance peaks, turning them into response peaks... Using an L-Pad can help in reducing those peaks. Using a second order filter can help in reducing the sensitivity to those peaks, so you're making a reasonable choice there.
 
diyAudio Moderator
Joined 2008
Paid Member
Sometimes it's important to keep the inductor resistance the same as before. That is to say that big differences could change the sound. No, it doesn't simply mean better sound when it is lower, as long as it has been designed for the resistance you choose.

Often for woofers lower is chosen to reduce heat and to reduce the change to the bass alignment/rolloff that has been created by the box. Say, less that 1/10 of 8 ohms.

For pro applications, heat is a more important consideration. Also, pro drivers are often more sensitive so there is actually less heat except when delivering higher output. Heat can melt insulation and cause inductor shorts or worse.
 
One more dumb question, if I am adding an L-pad to the tweeter high pass filter with the tweeter rated at 250 Watts RMS what size resistors do I need? Simple calculator shows for a 3 dB pad that the series resistor is 146 watt and the parallel resistor to the tweeter is 104 watts. Those are some huge resistors. This puts 250 watts across the tweeter. I must be making a mistake.
 
diyAudio Moderator
Joined 2008
Paid Member
I don't think you have, but that rating means less than it seems and you won't need that kind of power. Some compression tweeters can get very loud on less than 1W. You won't get them to 250W before they hit the stops.

I, for example, usually begin with the standard 5W resistors and check them (feel them) when I give them a workout.
 
Just found this thread by recommendation from another member and what a great thread with lots of good info for total novice like myself.

I am trying to build a crossover for DIY speaker for the first time and followed the instruction on page 1 and here are the calculations that I got.

[Woofer]
McCauley Sound 15" Woofer B1560 - 350W, 35 Hz - 1.2 kHz, 95db @ 8 ohm
http://www.mccauleysound.com/component_specifications.cfm?ID=120

  • Cut off 600Hz
  • Le 1.6mH (0.0016)
  • Rdc 6 ohm
  • Value of resistor = 6 x 1.25 = 7.5ohm
  • Value of capacitor = 0.0016 / (7.5)squared = 28.6uf
  • Value of Inductor = 8ohm / (6.3 x 600Hz) = 2.11mH

[Tweeter]
JBL S Tweeter Selenium ST200 - 70W RMS, 2 kHz - 20 kHz at 105db @ 8 ohm
https://www.parts-express.com/pedocs/specs/264-348--st200-spec-sheet.pdf

  • Cut off 2500Hz
  • New impedance = (8 ohm x 10) / (8+10) = 4.44 ohms
  • Value of Capacitor = 4.44 ohm x 12.6 x 2500Hz = 139,860. 1/139860 = 7.15uf
  • Value of Inductor = 4.44 ohm / (2500Hz x 3.15) = 0.56mH
  • Value of Resistor = 105db - 95db = 10db and lower 2 db extra in case = 12db lower which means multiply by 2 = 4.44 x 2 = 8.88 ohm resistor

[Midrange]
Faital Pro 6FE200 - 260W RMS, 85 Hz - 6 kHz at 95db @ 8ohm
https://faitalpro.com/en/products/LF_Loudspeakers/product_details/index.php?id=401020110


[What I could not figure out] o_O

+ I am not sure I should follow woofer calculation instructions or tweeter calculation for Mid-Range Crossover values. I read post #82 and could not understand.

+ Is 10 ohm good starting point to use to calculation new impedance for tweeter?

+ To find out the value of resistor for tweeter crossover, I used 12db (10db difference + 2 db extra) which should be two times of impedance. Have I understood correctly?


Again, thank you so much for putting great info for beginners.
 
diyAudio Moderator
Joined 2008
Paid Member
I am not sure I should follow woofer calculation instructions or tweeter calculation for Mid-Range Crossover values.
Can you do both?

I've decided that post #82 could do with some polishing so thank you for your feedback. I will eventually include schematics there as well.

I used 12db (10db difference + 2 db extra) which should be two times of impedance. Have I understood correctly?
The same as the impedance will get you approximately -6dB.

Your woofer is 95, so after your baffle step it will be 89, give or take for the room effects and EQ. So you could aim your midrange and tweeter there to begin with.
 
  • Like
Reactions: 1 user
Can you do both?

I've decided that post #82 could do with some polishing so thank you for your feedback. I will eventually include schematics there as well.


The same as the impedance will get you approximately -6dB.

Your woofer is 95, so after your baffle step it will be 89, give or take for the room effects and EQ. So you could aim your midrange and tweeter there to begin with.

Thank you, Allen. I tried to figure out and am lost how to do both... I guess that I need 2 inductors and 2 capacitors for mid-range instead of one of each?

If it is ever possible, would you be able to update Post #82 with example calculations like tweeter and woofer's? It would be a huge help to understand how to find out value of inductor and capacitors for midrange and how schematics changes from 2 way to 3 ways.
 
diyAudio Moderator
Joined 2008
Paid Member
I'm working on it now, but it will take some time.

How about an L-pad next to the driver to start. This will bring it down 6dB and smooth the impedance. It will also mean you don't need the same resistor and capacitor as the woofer circuit. (By the way, you can find L-pad calculators online just for now)

Next you can use the capacitor and inductor from the tweeter circuit, or just the capacitor.

Then add the woofer inductor. Total of 4 or 5 components. (I'll have to review this after making the demonstration graphics.)
 
  • Like
Reactions: 1 users
I'm working on it now, but it will take some time.

How about an L-pad next to the driver to start. This will bring it down 6dB and smooth the impedance. It will also mean you don't need the same resistor and capacitor as the woofer circuit. (By the way, you can find L-pad calculators online just for now)

Next you can use the capacitor and inductor from the tweeter circuit, or just the capacitor.

Then add the woofer inductor. Total of 4 or 5 components. (I'll have to review this after making the demonstration graphics.)

Thank you very much, Allen.

Please correct me if I am wrong and does L-Pad mean putting 4R in series with the tweeter and about 6R in parallel in this case and below components in total (7 components in total)?

Tweeter Circuit : 7.15uf cap, 0.56mH inductor, 4R in series & 6R in parallel resistors
Woofer circuit : 28.6uf cap, 2.11mH inductor, 7.5ohm resistor.

Midrange is whatever leftover from tweeter and woofer circuit filtered so just reverse the polarity and connect midrange driver positive to Tweeter negative , negative to woofer positive?