Hi, I would like to build a speaker similar to the Granada G2.
It is using 2 x AE-12 Dipole, a Scan Speak 12MU and 2 x D3004/662000. One in the front and one on the back.
The G2 is active, but I will prefer a passive filter.
Sensitive 2.83v
Dipole-12 8 ohm 91db, 4 ohm 94db
12MU 4 ohm, 90db
D3004/662000 4 ohm 91db
Can I mix 8 and 4 ohm’s drivers?
I am open for suggestions, just want a passiv filter
http://aespeakers.com/shop/dipole/dipole12/Acoustic Elegance Dipole12 woofer for Open Baffle Bass
https://www.scan-speak.dk/product/d3004-662000/D3004/662000 – Scan-Speak A/S
https://www.scan-speak.dk/datasheet/pdf/12mu-4731t00.pdfhttps://www.scan-speak.dk/datasheet/pdf/12mu-4731t00.pdf
It is using 2 x AE-12 Dipole, a Scan Speak 12MU and 2 x D3004/662000. One in the front and one on the back.
The G2 is active, but I will prefer a passive filter.
Sensitive 2.83v
Dipole-12 8 ohm 91db, 4 ohm 94db
12MU 4 ohm, 90db
D3004/662000 4 ohm 91db
Can I mix 8 and 4 ohm’s drivers?
I am open for suggestions, just want a passiv filter
http://aespeakers.com/shop/dipole/dipole12/Acoustic Elegance Dipole12 woofer for Open Baffle Bass
https://www.scan-speak.dk/product/d3004-662000/D3004/662000 – Scan-Speak A/S
https://www.scan-speak.dk/datasheet/pdf/12mu-4731t00.pdfhttps://www.scan-speak.dk/datasheet/pdf/12mu-4731t00.pdf
Shortly, it is not possible to make this kind of OB speaker with passive filter, it needs electronic response corrections - DSP EQ.
Don’t want to sound harsh, but based on your questions I would strongly encourage you to find a finished project and repeat it. These are pretty expensive drivers to experiment.
There are many OB speakers without DSP. Spatial Audio and The Pure Audio Project are both using 15" drivers without DSP.
I know I am a newbie but I have people that can help me. Just collecting information.
I know I am a newbie but I have people that can help me. Just collecting information.
You can do it (passive), but the result will be very insensitive.
This is a good simulator for basic pressure losses open baffle: (..put your driver size on the baffle you create, put the virtual mic. in front of the virtual driver at 2 meters distance on the baffle you created, and then check the "open baffle" box.)
Tolvan Data
Note that there are some added compromises with regard to dispersion between mid and tweeter in a typical 3-way. (..in that you need a much more efficient driver for the mid. which is usually larger in diameter - this makes getting your tweeter closer to the mid-point of that mid. much more difficult unless it's a coaxial design.) Generally the solution for this is more drivers (as in a 4 or 5 way loudspeaker).
To counteract both problems of insensitivity and crossover dispersion:
Ex. Tweeter +
smaller diameter midrange to get you to about 2 kHz (or higher) like the Scan Speak.
Then a more appropriate larger driver to cross-over to the pressure loss point of the smaller diameter midrange that will also have less pressure loss via a weaker motor (Qts of .6 or higher) and be a LOT more efficient to take up the "slack" of the efficiency loss in the lower midrange and upper bass (..though usually you aren't going much lower than 150 Hz with this sort of driver - with 150 Hz as a crossover point). However, the larger the baffle, the lower you can go (..and also the lower the upper midrange can go before it needs a high-pass filter).
Ex.
FaitalPRO | LF Loudspeakers | 8FE200 (8Ω)
or a lot more money:
215 Signature Extended bass-midrange driver 35W / 96.5dB / 8 Ohms - Supravox
This then leaves you with the bass - and bass usually needs a lot of pressure "boost" to combat the pressure loss (..again, model with the Edge). HOWEVER, "active" bass (that goes into the midrange) can be a solution - this of course puts you back into an "active" design, if only semi-"active". Alternatively with a LOT more surface area (with something like dual 18" drivers) and a lot more efficiency you could compensate with passive filter and reach fairly near the average to as low as 35 Hz.
..in any event: this is all EXCEEDINGLY COMPLEX for someone without a LOT of knowledge to get a good result without using a proven design.
This is a good simulator for basic pressure losses open baffle: (..put your driver size on the baffle you create, put the virtual mic. in front of the virtual driver at 2 meters distance on the baffle you created, and then check the "open baffle" box.)
Tolvan Data
Note that there are some added compromises with regard to dispersion between mid and tweeter in a typical 3-way. (..in that you need a much more efficient driver for the mid. which is usually larger in diameter - this makes getting your tweeter closer to the mid-point of that mid. much more difficult unless it's a coaxial design.) Generally the solution for this is more drivers (as in a 4 or 5 way loudspeaker).
To counteract both problems of insensitivity and crossover dispersion:
Ex. Tweeter +
smaller diameter midrange to get you to about 2 kHz (or higher) like the Scan Speak.
Then a more appropriate larger driver to cross-over to the pressure loss point of the smaller diameter midrange that will also have less pressure loss via a weaker motor (Qts of .6 or higher) and be a LOT more efficient to take up the "slack" of the efficiency loss in the lower midrange and upper bass (..though usually you aren't going much lower than 150 Hz with this sort of driver - with 150 Hz as a crossover point). However, the larger the baffle, the lower you can go (..and also the lower the upper midrange can go before it needs a high-pass filter).
Ex.
FaitalPRO | LF Loudspeakers | 8FE200 (8Ω)
or a lot more money:
215 Signature Extended bass-midrange driver 35W / 96.5dB / 8 Ohms - Supravox
This then leaves you with the bass - and bass usually needs a lot of pressure "boost" to combat the pressure loss (..again, model with the Edge). HOWEVER, "active" bass (that goes into the midrange) can be a solution - this of course puts you back into an "active" design, if only semi-"active". Alternatively with a LOT more surface area (with something like dual 18" drivers) and a lot more efficiency you could compensate with passive filter and reach fairly near the average to as low as 35 Hz.
..in any event: this is all EXCEEDINGLY COMPLEX for someone without a LOT of knowledge to get a good result without using a proven design.
Last edited:
You can do it passive with those drivers but you will have to do a lot of measuring to get it right. If you have help from someone who knows how to design dipole speakers then its not an issue.
You could add a Minidsp HD for a few hundred dollars for room correction and bass EQ and have a very good solution.
And you'll need a reasonably powerful amp either way.
You could add a Minidsp HD for a few hundred dollars for room correction and bass EQ and have a very good solution.
And you'll need a reasonably powerful amp either way.
Last edited:
Hi Martin
I did some study on dipole speaker design recently. A passive filter is possible, but the sensitivity will not be that high. It depends on the F3 that you want and the woofer baffle width you choose. Sensititvity will be around 80 dB, 2.83 Vrms, 1m in full space for the AE Dipole 12-D16 (8 Ohm) on a 50 cm wide baffle and a F3 = 40 Hz. 83dB for the AE Dipole 12-D8, 4 Ohm version.
On my website you can find a recent dipole design, VCL EX21 – VCLLabs. Maybe it gives you some extra inspiration.
I have plans to make a 4-way dipole design very soon. Using the AE Dipole 18-D8, the Supravox 215RTF64, the Scanspeak 12MU and the Mundorf AMT17D2.2. I will design a passive version, an analog active version with opamps or tubes and a digital version using biquads, for miniDSP for example. The design will appear on my website soon.
Last edited:
How can the sensitivite be so low? the woofers are reated at 91db at 2.83V.
Also the spatial audio m3 sapphire is 91db
As you can understand from Paul and Scott posts M3 Sapphire 91db sensitivity and Freq Resp: 30Hz cannot be thru. Because of dipole cancellation bass response with this size OB baffle will start fall below about 300-400 Hz with slope 6 dB per octave. It is possible to compensate with passive filters as Paul make in his built but you lose then about 15 dB in sensitivity if compensations goes down to 30 Hz.
Martin, please study the basics of dipoles carefully! Looks like you don't know much...
Dipolplus - Alles über offene Schallwände
Tech
Electro-acoustic models
Dipolplus - Alles über offene Schallwände
Tech
Electro-acoustic models
The 91 dB sensitivity you find in the datasheet of the woofer is the SPL on infinite baffle, the half space response. For a dipole, the woofer on a finite baffle in full space, the SPL becomes different.
For a better understanding, look here: Electro-acoustic models.
Edit: like Juhazi just posted already 🙂
To have an idea of the dipole response, in attach a simulation in Leap, AE Dipole 12-D16 on a 50 cm x 50 cm baffle in full space at 1m, 2.83 Vrms. The pink curve is the infinite baffle response, the green one the dipole response. For dipoles, the correct response has to be measured, but the simulation is rather close to it.
For a better understanding, look here: Electro-acoustic models.
Edit: like Juhazi just posted already 🙂
To have an idea of the dipole response, in attach a simulation in Leap, AE Dipole 12-D16 on a 50 cm x 50 cm baffle in full space at 1m, 2.83 Vrms. The pink curve is the infinite baffle response, the green one the dipole response. For dipoles, the correct response has to be measured, but the simulation is rather close to it.
Attachments
Last edited:
so this is not true Trio15 Classic, Open Baffle Speakers by PureAudioProject - PureAudioProject
Also many are using 8W 300B amps with the M3 Sapphire. And the have good bass.
I will read more but I are all the reviews lies?
Also many are using 8W 300B amps with the M3 Sapphire. And the have good bass.
I will read more but I are all the reviews lies?
The woofers used in the Trio15 Classic are Eminence OB-A15neo. Their sensitivity is 96 dB, 2.83Vrms 1m on infinite baffle. So two of them parallel have 102 dB.
Placing them in a baffle of 56 cm wide and supposing the baffles are roughly 20 cm deep, the SPL is about 87 dB at 30 Hz, 1m 2.83 Vrms in simulation. That means that the sensitivity of the dipole speaker for F3 = 30 Hz is about 90 dB at 1m, 2.83 Vrms in full space.
Measuring them standing on a floor (half space), the sensitivity will be 6 dB more, being 96 dB. Some manufacturers specify the sensitivity in half space.
So, I don't think the sensitivity will be 96 dB in full space.
Placing them in a baffle of 56 cm wide and supposing the baffles are roughly 20 cm deep, the SPL is about 87 dB at 30 Hz, 1m 2.83 Vrms in simulation. That means that the sensitivity of the dipole speaker for F3 = 30 Hz is about 90 dB at 1m, 2.83 Vrms in full space.
Measuring them standing on a floor (half space), the sensitivity will be 6 dB more, being 96 dB. Some manufacturers specify the sensitivity in half space.
So, I don't think the sensitivity will be 96 dB in full space.
They compensate somewhat with a higher Qts (Qe) driver, and likely some added resistance. Along with 2 woofers in parallel. Along with any gain the 300B amp provides via the impedance rise at the driver's Fs.
This is in addition to the loading on-baffle and in-room.
Despite all of that - it doesn't mean that the bass response will be near the average (which is usually determined at around 1 kHz) at low freq.s, it likely "tapper's off" in output the lower you go. and is almost non-existent (audibly speaking) below 30 Hz (especially when considering loudness curves). In this respect then you could consider it a lie.
Again: exceedingly complex. 😉
I'd start with the bass 1st - it's the most challenging. (..that with the ability to measure in-room w/ an spl reference). This is usually REW + uMIK 1 microphone.
These seem decent for the task if desiring only a passive result (which usually requires a large iron core inductor for the eq. typically larger than 20 mH in series with the driver):
BIANCO-18SW450
SB Audience Bianco 18SW450 - first impressions (open baffle)
Last edited:
Have you read what Troels says and shows about it?
"Actually I'm only going to show two measurements as they are pretty similar to OBL-11 and not least because measuring frequency response on an open baffle is a pain due to rear wave cancellation. What's show above is the MT @ 0.5 meter distance with 1/12 octave and 1/3 octave smoothing, red and yellow respectively. Green is nearfield response of bass driver. What we can get from this is a fairly even response vs. frequency around 93-94 dB/2.8V, 1 meter. "
Well, fairly even 93-94 dB but only above 200Hz! Nearfield measurement of the bass shows there is dipole loss compensation applied, but relative spl level is a big question, typical nearfield is taken at 1cm distance!
Last edited:
If we look at the OBL-15 measurements then it looks like there is no bass 🙂
But OB is very hard to measure and it’s measured in his workshop.
It doesn't look good.
But read his listening impression. Best bass ever, better than any other speaker he has built.
Okay I surrender 🙂 Active XO with some baffle step compensation would be the best, but it’s possible to build OB without.
I know that a DSP with miniDSP or DEQX HD5 is very good.
I just can’t accept the dacs and amp’s used in such a configuration.I like a good big tube amp, grammophon and a super dac.
Offuse you could use 6 tube amps and 6 dacs but the cost and heat would be tremendously
But maybe I cloud try something like this: [url]https://www.xkitz.com/products/linkwitz-riley-2-way-active-crossover-fully-balanced-xover-2b[/URL]
As I understand it the reason so many are using 15” is to use the second to compensate for the baffle step loss. Is that right?
Thanks everyone for making me a little wiser and keeping a good tone.
But OB is very hard to measure and it’s measured in his workshop.
It doesn't look good.
But read his listening impression. Best bass ever, better than any other speaker he has built.
Okay I surrender 🙂 Active XO with some baffle step compensation would be the best, but it’s possible to build OB without.
I know that a DSP with miniDSP or DEQX HD5 is very good.
I just can’t accept the dacs and amp’s used in such a configuration.I like a good big tube amp, grammophon and a super dac.
Offuse you could use 6 tube amps and 6 dacs but the cost and heat would be tremendously
But maybe I cloud try something like this: [url]https://www.xkitz.com/products/linkwitz-riley-2-way-active-crossover-fully-balanced-xover-2b[/URL]
As I understand it the reason so many are using 15” is to use the second to compensate for the baffle step loss. Is that right?
Thanks everyone for making me a little wiser and keeping a good tone.
Most informative DIY open baffle build thread ever!
Hi Martin,
I think you will benefit greatly from reading this thread... My open baffle dipole with Beyma TPL-150
It is the single best source of real world DIY open baffle design combined with lots of other great ideas, experiences, experiments and input from other top DIY guys.
I love open baffle covering the 80Hz to around 2.5Khz band IF you have at least 1 meter ( best with 2 meters) free space behind your speakers which you can fill with large plants and room treatment on the walls.
Plus you will need to use DSP to integrate the subs and high frequencies with accurate time and frequency domain FIR filters and room Eq.
Basically that is what Stig Eric discovered and he has almost unlimited resources!
In my experience listening to dozens of DIY (including my own) and commercial open baffle designs, the "purist" open baffle designs using passive crossovers, valve amps and a pair of high Qts (usually with rubber surrounds... Yuk!) 15 or 18 inch drivers per side all sound compromised, muddy and gutless. They give the modern, correctly designed open baffle speakers a bad reputation.
Hope the Stig Eric thread is of interest.
Hi Martin,
I think you will benefit greatly from reading this thread... My open baffle dipole with Beyma TPL-150
It is the single best source of real world DIY open baffle design combined with lots of other great ideas, experiences, experiments and input from other top DIY guys.
I love open baffle covering the 80Hz to around 2.5Khz band IF you have at least 1 meter ( best with 2 meters) free space behind your speakers which you can fill with large plants and room treatment on the walls.
Plus you will need to use DSP to integrate the subs and high frequencies with accurate time and frequency domain FIR filters and room Eq.
Basically that is what Stig Eric discovered and he has almost unlimited resources!
In my experience listening to dozens of DIY (including my own) and commercial open baffle designs, the "purist" open baffle designs using passive crossovers, valve amps and a pair of high Qts (usually with rubber surrounds... Yuk!) 15 or 18 inch drivers per side all sound compromised, muddy and gutless. They give the modern, correctly designed open baffle speakers a bad reputation.
Hope the Stig Eric thread is of interest.