Or front loaded horn ? I always wondered for FLH but I have not the free area for that in y living room so it solved the problem for me as an impossibility to go that path !
Clayton Shaw uses these Beyma 12BR70 in his new Caladan OB:
"Shaw’s Caladan boasts a 1.5" thick baffle made from kiln-dried, solid American hardwood. It employs twin 12" Beyma pro-audio bass/mid units and a 1 1/8" silk-dome tweeter. Frequency response is said to be 32hz–20khz, ±2db, the sensitivity 93db/W/m, and the nominal impedance 4 ohms. The Caladan stands 38" high by 18" wide by 10" deep, including the legs, and weigh 35lb. Isoacoustics Gaia iii feet are included."
https://www.stereophile.com/content/linear-tube-audioclayton-shaw-acoustic-lab
https://claytonshawacousticlab.com/697-2/
"Shaw’s Caladan boasts a 1.5" thick baffle made from kiln-dried, solid American hardwood. It employs twin 12" Beyma pro-audio bass/mid units and a 1 1/8" silk-dome tweeter. Frequency response is said to be 32hz–20khz, ±2db, the sensitivity 93db/W/m, and the nominal impedance 4 ohms. The Caladan stands 38" high by 18" wide by 10" deep, including the legs, and weigh 35lb. Isoacoustics Gaia iii feet are included."
https://www.stereophile.com/content/linear-tube-audioclayton-shaw-acoustic-lab
https://claytonshawacousticlab.com/697-2/
Last edited:
That's about what you can expect from such an inexpensive driver...
Since you mentioned using two 12" drivers, I will mention that in some of my own OB designs I have used four 8" drivers. This makes for a nice compact and square bass section. The 8" driver will not have issues like cone breakup or surround resonances, etc. in a passband that goes up to 400Hz. Using four you can wire series-parallel or even all in parallel for higher voltage sensitivity. Four 8" drivers with 5mm Xmax can operate down to and crossed over at e.g. 70Hz with pretty high SPL (for home audio) even when the baffle is not large. Even a relatively cheap 8" will not have issues below 400Hz...
Maybe this is an option you would consider?
That speaker will have a very F-d up power response. Crossing a 12" to a monopole 1" ish dome tweeter is not typically recommended because the radiation pattern will grow narrow just below and then widen to almost monopole above, the crossover point. I'm not seeing anything to "boast" about here...
Your system has too much drivers ... Four way is a problem to tune, and you want to go to five ways ?
I think with only three ways you can do something sounding good.
You don't need a 12" for the midbass, the 8" can do well the job. You have to equalize actively the 8" to use it in the range 130Hz-1300Hz LR4 slope recommended The anemic mid-bass is a bad equalization or a negative effect of other drivers. The equalization depends on the baffle size.
Note add a second 8" as suggested could be done
And the 18" or 4x12" covers the bass 20-130Hz. Keep thing simple.
I was already done this kind of system with an analog passive crossover and the result is stunning. Do not forget to align acoustic centers
I havent heard the speaker but apparently plenty of good reviews...I was just putting it as an example. Maybe they do sound weird...
For a sealed enclosure loudspeaker (by definition, though terms are sort of misnomers):
Q < 0.5........Over damped...........No overshoot to rest position after a step input, extended settling time, tapered low frequency response
Q = 0.5........Critically damped......No overshoot to rest position after a step input, minimum settling time without overshoot (not perfect transient response)
Q > 0.5........Under damped..........Overshoot to rest position after a step input
Q = 0.577.....Bessel Function.........Very slight overshoot to rest position after a step input, minimum settling time
Q = 0.71......Butterworth Function..Slight overshoot to rest position after a step input, lowest flat frequency response
And to make things a little more complicated:
https://www.geocities.ws/kreskovs/Box-Q.html
Well, this is a bit of a puzzle. I can't see any reason why the Faital 18HP1010 should be a failure in this application. But unless you've worked out what the problem with it is, you risk spening money and not fixing anything.
Can you share your equalisation/crossover on the Faital? Or for whole setup?
Can you share your equalisation/crossover on the Faital? Or for whole setup?
what does this number/calculation represent?
Yes, I would consider this. I do have 4x 8PE21 but these aren´t woofers but midranges. Not enough Xmax. Which 8" drivers have worked well for you in OB and having good midrange into 400Hz?
@ianbo , @lousymusician
Will get my measurements with Acourate and the correction applied. It´s a bit cumbersome because it´s in a computer I rarely use and need to bend over backwards to get it up and running. Hopefully I can get it done tonight. Otherwise tomorrow night.
Makes total sense to first figure out what is causing this, if you can help me from reading graphs.
Thank you!
You're welcome!
0.4 Qt = measured?, Specs say 0.31 Qt, audible difference WRT (mid) bass output.
Some of the cheaper Visaton drivers are nice in OB. The W300 is a good 12 for OB. Eminence Delta Pro 15A and the Deltalite 2512 are good sounding 15s. Newark has some cheap cast frame pro drivers which also sound good for higher Q applications
Someone mentioned the Beta8 for a mid. I fully endorse that driver. Its a hidden gem. Beautiful midrange. Likely the best pro mid under $100.
Someone mentioned the Beta8 for a mid. I fully endorse that driver. Its a hidden gem. Beautiful midrange. Likely the best pro mid under $100.
@ianbo , @lousymusician
Some measurements from when I developed the filters in Acourate.
First graph shows left and right sweep measurements using the digital xo, also the target curve against which the speaker response is hammered into shape:
You can see the massive drop of the 18" starting around 280Hz.
These are the xo:
Sorry for the picture quality and size...couldn't format well on Android.
Some measurements from when I developed the filters in Acourate.
First graph shows left and right sweep measurements using the digital xo, also the target curve against which the speaker response is hammered into shape:
You can see the massive drop of the 18" starting around 280Hz.
These are the xo:
Sorry for the picture quality and size...couldn't format well on Android.
I understood, sure your crossover is not working ! You don't have any open baffle equalization .... Sure you don't have any low mid and you can better with less.
This an example of crossover I done. I include the dipole equalization and delay correction.
With your numeric crossover you should do the same crossover response.
Bass correction is red curve - Green is mid curve and blue is tweeter curve : the only normal curve without dipole correction. The mid curve is a little more complex du to mid-bass correction.
To make dipole correction you should take into account the dimension of the baffle. Here the baffle is 30 cm wide, a dipole equalization 20-400Hz is done.
Note you can do a global dipole equalization (see Perry Marshall designs ) or for each drivers.
This an example of crossover I done. I include the dipole equalization and delay correction.
With your numeric crossover you should do the same crossover response.
Bass correction is red curve - Green is mid curve and blue is tweeter curve : the only normal curve without dipole correction. The mid curve is a little more complex du to mid-bass correction.
To make dipole correction you should take into account the dimension of the baffle. Here the baffle is 30 cm wide, a dipole equalization 20-400Hz is done.
Note you can do a global dipole equalization (see Perry Marshall designs ) or for each drivers.
The last time I used the 4x 8" drivers I used a Peerless HDS 8" woofer but they are no longer available. You could for example use the new Dayton 8" SIG225, which has some similarities to the Peerless driver and would perform well below 400Hz. I would definitely use a home audio 8" and not a pro audio driver in this case. I'm not sure what brands are available where you live, or I would give you some more recommendations.
@LewinskiH01 Thanks for sharing the pictures. Have I got this right: the first photo shows a sweep of L and R channels, each with all drivers playing. Acourate is crossing over between drivers, but no further equalisation has been applied. Is that correct? I assume that for listening, you apply a further equalisation filter to this response, to force it to fit the target curve - but that equalisation curve isn't shown. Correct?
You mention that you can see the Faital 18 dropping off below 280 Hz or so. Yes, you can. And that's exactly what you'd expect it to do, uncorrected, on a small baffle or playing nude.
The second photo shows crossover filters. Again, these don't look as if they include equalisation to fit the driver responses to a target curve.
I may be misunderstanding you, but if you are playing the system using the filters shown in the second photo, and no further equalisation, then you will be getting the response shown in the first photo. And that doesn't include any equalisation of the Faital for the open baffle loss (as Jerome69 said). That would certainly cause anaemic mid-bass. But maybe you do indeed have a further layer of equalisation, fitting the response to the target curve, but not shown in the photos?
You mention that you can see the Faital 18 dropping off below 280 Hz or so. Yes, you can. And that's exactly what you'd expect it to do, uncorrected, on a small baffle or playing nude.
The second photo shows crossover filters. Again, these don't look as if they include equalisation to fit the driver responses to a target curve.
I may be misunderstanding you, but if you are playing the system using the filters shown in the second photo, and no further equalisation, then you will be getting the response shown in the first photo. And that doesn't include any equalisation of the Faital for the open baffle loss (as Jerome69 said). That would certainly cause anaemic mid-bass. But maybe you do indeed have a further layer of equalisation, fitting the response to the target curve, but not shown in the photos?
Last edited: