12" Woofer and 2.5" Wideband Cabinet Conception / Crossover

Impedance trace from DATSV3.
Blue is magnitude.
Red is phase.
Green is real.
Purple is imaginary.

BB_impedance_sweep_2023-05-31_02.png


DATS Linearity test traces
Green = -9 dBu
Orange = -1 dBu
Whatever blue color that is = 10 dBu
BB_DATS_linearity_202305-31_01.png


Hope this helps.

Will be looking to get the laptop and REW going hopefully later.
 

Attachments

Peaks and Valleys pulled from the text files:

Standard measurement at 10 dBu (minor differences across 5-6 measurements).
46.581 16.093 -19.693
80.635 6.599 -12.550
82.998 6.602 -11.441
313.358 11.564 -9.114
703.464 4.667 -8.066
724.077 4.669 -6.206
1933.055 7.058 4.986
4597.605 5.803 11.712
4732.325 5.802 12.633

The other two just for fun.

-1 dBu
47.946 16.910 -18.006
80.635 6.573 -12.913
313.358 11.580 -9.076
322.540 11.567 -12.275
703.464 4.674 -8.124
724.077 4.672 -6.274
1878.024 7.076 5.491
1933.055 7.076 4.934
4732.325 5.789 12.445

-9 dBu
47.946 17.476 -14.942
80.635 6.555 -13.165
313.358 11.578 -9.038
322.540 11.573 -12.389
703.464 4.648 -8.093
1878.024 7.067 5.592
1933.055 7.066 4.959
4870.992 5.787 13.595



 
Yes.

For the small driver, I probably have individual measures saved from before, I would just need to pull it to look up which one.

I did not measure the woofer individually. If I need to, I will put it on the list.

Neighbor's boy was either mowing or asking for help when his mower quit, so no REW tonight.
 
Awesome great job !!

Performing as expected. In a nutshell around 5.8 to 19.6 ohms

I was curious how close real world Ql would be and was assuming new out of the box. The woofer
Fs would be slightly higher.
For the most part Fb was looking to fall around 39 to 41 according to sim.

Pretty much how it came out.
Its sealed so I was unsure to use 10 or 7 for Ql

But the impedance variation is very little in sim between Ql 10 or 6 but notable.

compared to sim predictions and real world it seems around good old standard 7
and Fs of the woofer out of the box not more than 6 Hz higher.
 
I have near-field measurements with mic at center of wideband and woofer right at edge of baffle/rubber surround. That is inside the garage, facing out, but full of clutter. I was not able to get 1m outside measurements, it started sprinkling 30 seconds after I got everything carted out and set-up.

I am not sure of what to post or how to process the data. I know how I did it for myself and the little gift speakers I was making, but those were single driver speakers. Let me know what you would like.

Here is a capture of the spl trace for the woofer (one of three runs).
BB Woof Near 1 Jun 1.png


When I moved the mic for the wide-band, the level needed adjusting, so I made another woofer trace using the same amplified volume setting as the wide-band.
BB WB Near matched Jun 1.png
BB Woof Near matched Jun 1.png


I meant to make some pictures of the impulse trace, but apparently forgot.

I will see if the REW project files upload.
 

Attachments

Soon as the weather is nicer in your part.

Everything summed from the wideband be cool.
and far you wanna go off axis. 15 , 30 , 45 degree

filter Q looks correct.
Wideband looks smoother than data sheet.

You are familiar with these drivers
is the high-end usually behave like that or we seeing something else

otherwise it looks smoother than datasheet. and that 3k bump is artifact seen in
the 4 ohm version. far as I could tell from factory FRD

Definitely not a REW expert myself, many others will chime in.
crossover points look about right. full sum should be interesting.
 
Last edited:
Not sure how comparable, but here are the results from a pair of ported spheres made from Target bowls, the "fluff" was not true Acousta-Stuff, it was something called "Renewliner" and it wasn't teased out near as well. In this one co-mingled graph with Near Port traces, the Near Driver traces, and the Far traces, then the combined and aligned trace from two separate speakers.

Sphere_Outside_Damping_Tests.png


If I have been doing it correctly, I might guess that the near driver trace is pretty smooth, but I believe that is only useful up to the lowest point the 1m far measurement can get to before the first reflection.

Back to the speaker at hand (have you named her yet?):

Here are the impulse traces from the near measurements. The wideband looks familiar:

Wideband
BB WB Near matched Jun 1 impulse.png
BB WB Near matched Jun 1 impulse step.png


The woofer impulse isn't familiar to me, but without more experience, I am not sure if this is a reasonably expected outcome or not.

Woofer
BB Woof Near matched Jun 1 impulse.png
BB Woof Near matched Jun 1 impulse step.png


Edit: I hooked it up in the house next to a 3 way with advertised sensitivity of 86.5 dB (2.83v/1M), this speaker took a bit more power to get to the same perceived loudness. Wife said it had better thump. It might be that the woofer has a great initial impulse, but that second hump is some <thing> that may be tamed with damping?
 
Last edited:
yes , sensitivity is low around 82 dB

all in the balance act of getting the bass and smoothing the baffle
step on the mid.

It should drink up a little bump on volume knob pretty easy
or yeah, .... a lot of bump. compared to 86 dB

But that was the game to try to get in sub bass territory.
with high Qts woofer. Basically the size and baffle is shooting
for the thump.
And yeah long as its not too boomy.
Its sealed so yeah it could be stuffed
light batten on the walls. but lower area is were it could be really stuffed.

only if its too boomy. pretty much got the bump already there
that most EQ in.
140 liters was the right choice im thinking.
If it was too small I think that is when things get too boomy.
 
Crossover seems fine.
Impedance curve would likely show any mistakes.
pretty much got it. That part is spot on to the sim

Between sim and real life, I was already playing around with the notch.
So depending how it all sums together. One cap can be moved up in down in value
to fine tune if needed.

Otherwise had suspicion off axis would either drop or be magical with the fullrange.
in the highend. Didnt expect the top to measure like that.
I left space on the baffle for tweeter. it can go to 3 way as well.
Otherwise it looks smoother than the datasheet ...weird
 
Well, here is a picture of the 1m measurements.

Can't guarantee super accuracy, tripod mic stand on a gravel driveway, but I tried to keep it within an inch of perfect.

One note, I realized I never rounded over the baffle edges (I only have one router and I left it set up for the relief circles), so they are still square. And still no damping in main box.

BB Outside Traces Averaged Uncombined.png


I will have to figure out how to combine these into the single trace (per angle), I followed a script for the driver/port, not sure how to do that here. I might have to use yesterday's matched amp setting traces, these aren't matched.

Kids (3 & 4) have hit the danger zone with the wife, so I need to pack it in for today and get the kids so we can finish chores and get out of the house.
 
One note: I have a USB microphone and while I used a reference speaker for timing, I moved the microphone twice but not the reference speaker, so apparently they aren't going to be good for importing into VirtuixCAD.

Here are the REW files for one set and a set of averaged traces from three measurements.

I might get another chance next week, I just need to figure out a better way to collect the rotated measurements. If the center of rotation is the center of the baffle front edge, that makes a turn table seem impractical as it would be a pretty sizable turn-table and invite the reflections I am trying to avoid. Otherwise I am moving the step ladder every time anyway, so I am not sure how a turn table helps? I am not a big video watcher, but I might have to try to find one so I see the set-up.
 

Attachments

Last edited:
I like it very much. Even with just the one, I was super happy in the garage until the chicks hatched (so low volume and no dust generation out there for 6 weeks). After seeing how the second one goes, I might rebuild this one (to fix some little things) or try it with ply or foam/ply.

Caveats: I don't have the trained ear to faithfully use the more expressive words. I have occupational hearing loss and a definitive notch that coincides with the main cicada sound (I cannot hear them).

In the garage, I can hear a different [tone/voice] compared to a JBL Control 25AV. Not the ideal set-up for comparison at the moment, so I can't tell if it's placement or character. Maybe later I will get them facing out of the door on a switch (again at low volumes only).
https://jblpro.com/products/control-25av

I had it on the dining room table with a Philharmonic BMR which are new to me. So there isn’t a historical familiarity with either really. I wouldn't think that's a fair comparison, but it was a whole lot closer than I would have thought. Note: level matching was impossible, but I did have them connected with separate switches.
https://philharmonicaudio.com/BMR.html

Wife said:
White Dragon’s Big Girl had more low end thump.
Since they were sitting together on her dining room table, her main issue was the size.
Overall balance was better in the BMR.

My take:
Truth be told, I was torn.
The biggest advantage FOR ME of the BMR was in [midrange fill].
Otherwise, I am not so sure I didn’t prefer the sound of WD-BG on some things.

Other notes: Before stuffing the wide-band, I heard the expected hollow/echo. I thought I would still hear it now because there is still nothing in the bigger box - but it wasn't jumping out at me. The different [tone/voice] is more audible to me.

During the sweeps, there was a [shudder/resonance] as it got started, which left me wondering if I should have put a weight on top of it. I never heard it in music, only on the step ladder. On the other hand it wasn’t apparent in the traces, but I could clearly see the wind ripple the lower bass measures (I threw those away), so maybe it wasn't that much of an issue.

Hope that helps!
 
Last edited:
A better comparison with the JBL Control 25AV. Again, sensitivity is preventing a good level match and the overall levels are pretty low to protect the peepers (but mostly to keep peace in the house).

I can hear a difference something like the [midrange fill], but here I much prefer White Dragon’s Big Girl than the JBL. Which is great because that's what they are replacing.

If there is some reference to listen to and compare - like training wheels for critical listening ears, happy to try it.
 
Awesome, that is very interesting. Good to know the big girl gives the JBL a run for the money.

Definitely got to thank CANNON-FODDER for sticking through the frustrations of any project and getting a test cabinet
and crossover finished. Then squeezing in measurements with random rain storms.
Pretty darn good, after getting through the intimidation of the crossover. Think were moving you up in Rank.

WD-BG is great and straight forward acronym. pretty funny and good name.

reminds me a lot of my teen years.
Dad gave me his pioneers, sealed 12" with white cone. No ports high Qts speaker likely.

I had some small home speakers branded Altec with some basic Silk domes. way better highend than the Pioneers.

But everyone always asked me to hook up the big Pioneers instead. Always a favorite, with basic paper cone tweets.

I always wanted the Bass of the pioneers. And then cleaner highs with just a simple silk dome.

It was a concept, I was having fun with. Pretty interesting almost over night. Someone came in and built one.
It pushed me and got some good motivation to dip into SketchUp and do the plans for the Foam.

Think I need to dive in again and work on some wood / plywood plans.
Likely tackle a tweeter option as well.
 
  • Like
Reactions: CANNON-FODDER