Many people say that dome midrange drivers have a sound quality which is unique and desirable. I have heard systems with dome mids that sounded very good, and others which were very ordinary. This thread discusses this topic in detail....link
For the past year, I have been thinking about the Bliesma M74A-6 driver, and simulating various options to use it effectively. To me, this driver offers an excellent way to experience the power and potential of a good dome mid in a high performance application. I like that it operates pistonically over its full range, and the first mode resonance is more than an octave above its likely pass band. This feature sets it apart from its silk and paper dome sisters. I like that it has high sensitivity and high SPL capability. I like that @HiFiCompass has measured very low harmonic and modulation distortion at high SPL. I like that the “A” aluminum dome version is considerably less expensive than the beryllium and textreme versions.
Thus, I started this project with just one primary requirement, that it use the M74A driver.
Index
For the past year, I have been thinking about the Bliesma M74A-6 driver, and simulating various options to use it effectively. To me, this driver offers an excellent way to experience the power and potential of a good dome mid in a high performance application. I like that it operates pistonically over its full range, and the first mode resonance is more than an octave above its likely pass band. This feature sets it apart from its silk and paper dome sisters. I like that it has high sensitivity and high SPL capability. I like that @HiFiCompass has measured very low harmonic and modulation distortion at high SPL. I like that the “A” aluminum dome version is considerably less expensive than the beryllium and textreme versions.
Thus, I started this project with just one primary requirement, that it use the M74A driver.
Index
I ordered a pair of Bliesma M74A-6 drivers from Solen, and my cost including shipping was $833 USD for the pair. This is expensive, but very comparable to other high performance drivers such as Purifi, Satori Textreme, and ScanSpeak Revelator/Illuminator.
The pair arrived a few days ago without damage or other drama, the shipping package was well put together and well cushioned. The two drivers shared a common box, and they had sequential serial numbers.
I made an impedance scan of the two drivers, and they are an extremely close match.
M74A-6 S/N 0966:
M74A-6 S/N 0967
Solid=0967 Dashed blue =0966:
j.
The pair arrived a few days ago without damage or other drama, the shipping package was well put together and well cushioned. The two drivers shared a common box, and they had sequential serial numbers.
I made an impedance scan of the two drivers, and they are an extremely close match.
M74A-6 S/N 0966:
M74A-6 S/N 0967
Solid=0967 Dashed blue =0966:
j.
Nice!
Looking forward to what you make.
By the way, you can export the impedance from DATS and import into REW to overlay in a way better than dats does.
Looking forward to what you make.
By the way, you can export the impedance from DATS and import into REW to overlay in a way better than dats does.
I can say that the midrange is really impressive. I learned about them from a youTube video about a recording studio that started building them into studio monitors.
They were very key to the success of my project as I required a midrange with a shallow cup so I could mount two back to back in the thinnest baffle possible.
I also purchased the midrange from Solen. The exchange rate was favorable. https://www.solen.ca/en/products/bliesma-m74a-6-74mm-aluminum-magnesium-dome-midrange-3inch
I built a 3-Way speaker using this driver for the PNWAS 2024 speaker contest in the Seattle area. Due to errors I made with my crossover design software it was not really complete in time for the contest. Since then I have it working reasonably well. My speaker is a more complex design as it is a bipolar speaker with the same drivers on the front and back to produce a nearly omni-directional directivity profile. It is an assembly of two identical woofer boxes and a separate baffle for the midranges and tweeters. I used the SB ceramic coated tweeter and 10" woofers from Vifa / Tympany and Dayton 12" passive radiators. I can provide more details if anyone is interested.
As my passive crossover became very complex I plan to eventually move to a pair of Hypex Fusion 3-channel plate amps with DSP crossovers.
The front of the top enclosure is red and the back is silver.
Each woofer box is identical and has a 10" woofer and a 12" passive radiator.
The black carpet covered woofer boxes I used for prototyping are from Parts Express.
They are intended for cars and are very inexpensive and will ship for free if you buy two.
Two of them stacked look pretty good.
They were very key to the success of my project as I required a midrange with a shallow cup so I could mount two back to back in the thinnest baffle possible.
I also purchased the midrange from Solen. The exchange rate was favorable. https://www.solen.ca/en/products/bliesma-m74a-6-74mm-aluminum-magnesium-dome-midrange-3inch
I built a 3-Way speaker using this driver for the PNWAS 2024 speaker contest in the Seattle area. Due to errors I made with my crossover design software it was not really complete in time for the contest. Since then I have it working reasonably well. My speaker is a more complex design as it is a bipolar speaker with the same drivers on the front and back to produce a nearly omni-directional directivity profile. It is an assembly of two identical woofer boxes and a separate baffle for the midranges and tweeters. I used the SB ceramic coated tweeter and 10" woofers from Vifa / Tympany and Dayton 12" passive radiators. I can provide more details if anyone is interested.
As my passive crossover became very complex I plan to eventually move to a pair of Hypex Fusion 3-channel plate amps with DSP crossovers.
The front of the top enclosure is red and the back is silver.
Each woofer box is identical and has a 10" woofer and a 12" passive radiator.
The black carpet covered woofer boxes I used for prototyping are from Parts Express.
They are intended for cars and are very inexpensive and will ship for free if you buy two.
Two of them stacked look pretty good.
Last edited:
Comparing the two units, Fs is within 0.6%, Re is within 1.5%, and Qes is within 0.7%. That is impressive.
A new hifijim build thread is like Christmas coming early. Looking forward to watching the speaker come to life!
Hooray, a new hifijim speaker story!!
That was some idea from me in CAD for a "value", compact crystal clear performance monitor:
1x T25A
1x M74A
2x Revelator Alu 26W/8867T00 ( in ~60l CB "Air suspension". Bucking Magnets TBD: Discovery 26W -> +10% BL, Revelator 15M -> 0% BL...)
Hypex Fusion Module + OLED Display
Space for passive impedance matching network, like serial notches and coils for further current distortion reduction
1x T25A
1x M74A
2x Revelator Alu 26W/8867T00 ( in ~60l CB "Air suspension". Bucking Magnets TBD: Discovery 26W -> +10% BL, Revelator 15M -> 0% BL...)
Hypex Fusion Module + OLED Display
Space for passive impedance matching network, like serial notches and coils for further current distortion reduction
What is going on at 3, 6, and 9 Khz? I realize these are really small deviations, but for $415 a driver, they should be damn near perfect.
Take a look on the backside - ring magnet with small closed back making a sealed cavity...
https://hificompass.com/en/reviews/bliesma-m74t-6-74mm-textreme-dome-midrange
https://hificompass.com/en/reviews/bliesma-m74t-6-74mm-textreme-dome-midrange
Last edited:
Early on I ran a lot of simulations with VituixCad diffraction tool. I found that the best directivity performance from 500 Hz up occurred with either a very wide baffle, or a very small baffle. The very wide baffle, on the order of 600+ mm wide, presents the tweeter with an infinite baffle within its operating range, and the baffle step for the midrange is well below the crossover from mid to tweeter. This works out really well. Unfortunately, my room cannot accommodate a speaker this wide, so this option is off the table.
There are several ways to configure a very small baffle. The goal is to make the flat baffle area around the tweeter and mid as small as possible. My approach in the past has been to make deep tapered bevels. However, I am in the mood to try something different this time. A trapezoidal baffle with 37 mm radius edges simulates quite nicely, and I have the capability to cut 37 mm radius edges. The best results are when the trapezoid is sized to just neatly fit the M74 and whatever tweeter I might use. I made a lot of simulations over the past year trying different variations of trapezoid baffle shapes, based on the size of the tweeter chassis and the spacing between the tweeter and mid. I was able to get very nice directivity performance from a variety of trapezoid baffle shapes. This convinced me that I had a lot of downstream design flexibility with the trapezoid shape.
As I thought about constructing a trapezoid box, or a rectilinear box with a trapezoid shape on top, I realized it would be easier to make a separate trapezoid enclosure for the mid and tweeter which would be positioned on top of whatever woofer system I decided upon. This configuration cleanly solves the potential issue of isolating the midrange and tweeter from woofer-generated vibration.
This is one of my latest diffraction simulations, to illustrate the kind of directivity performance which is possible. At this stage of the design, getting the diffraction and directivity performance optimized is the primary task. These idealized diffraction simulations show us the performance with idealized drivers, and real world drivers will inevitably be worse, so it is important to get this as optimized as possible.
The primary drivers of directivity for this type of speaker are (1) the directivity of the tweeter (2) the directivity of the midrange driver from 1k - 3k, (3) the size, shape, and edge treatment of the baffle, (4) The vertical spacing of the drivers, (5) the crossover frequency and slopes. Once the drivers are selected and the cabinet is designed, 4 of the 5 directivity influences are fixed, and the crossover is the only variable left open. So I want to optimize the power and directivity performance to the extent possible.
Here is one of my latest simulations of a trapezoidal baffle, combined with a 2nd order crossover at 3500 Hz and some baffle step eq. This kind of power and directivity performance is very easy to work with.
Small frame tweeter (scanspeak 3004 or Bliesma T25)
Bliesma M74A
There are several ways to configure a very small baffle. The goal is to make the flat baffle area around the tweeter and mid as small as possible. My approach in the past has been to make deep tapered bevels. However, I am in the mood to try something different this time. A trapezoidal baffle with 37 mm radius edges simulates quite nicely, and I have the capability to cut 37 mm radius edges. The best results are when the trapezoid is sized to just neatly fit the M74 and whatever tweeter I might use. I made a lot of simulations over the past year trying different variations of trapezoid baffle shapes, based on the size of the tweeter chassis and the spacing between the tweeter and mid. I was able to get very nice directivity performance from a variety of trapezoid baffle shapes. This convinced me that I had a lot of downstream design flexibility with the trapezoid shape.
As I thought about constructing a trapezoid box, or a rectilinear box with a trapezoid shape on top, I realized it would be easier to make a separate trapezoid enclosure for the mid and tweeter which would be positioned on top of whatever woofer system I decided upon. This configuration cleanly solves the potential issue of isolating the midrange and tweeter from woofer-generated vibration.
This is one of my latest diffraction simulations, to illustrate the kind of directivity performance which is possible. At this stage of the design, getting the diffraction and directivity performance optimized is the primary task. These idealized diffraction simulations show us the performance with idealized drivers, and real world drivers will inevitably be worse, so it is important to get this as optimized as possible.
The primary drivers of directivity for this type of speaker are (1) the directivity of the tweeter (2) the directivity of the midrange driver from 1k - 3k, (3) the size, shape, and edge treatment of the baffle, (4) The vertical spacing of the drivers, (5) the crossover frequency and slopes. Once the drivers are selected and the cabinet is designed, 4 of the 5 directivity influences are fixed, and the crossover is the only variable left open. So I want to optimize the power and directivity performance to the extent possible.
Here is one of my latest simulations of a trapezoidal baffle, combined with a 2nd order crossover at 3500 Hz and some baffle step eq. This kind of power and directivity performance is very easy to work with.
Small frame tweeter (scanspeak 3004 or Bliesma T25)
Bliesma M74A
Last edited:
I get good results with rectangular baffles, I didn't notice an advantage to trapezoidal baffles other than the aesthetics. There can be some advantage locating the driver off center.. Positioning the driver centered at 1/3 the width of the baffle, the diffraction effect can be reduced. There is a ratio left to right that aligns the dip from the left side with the peak from the right.
On the narrow baffle I loaded it into I had the baffle step to contend with in the pass band as I was running the midrange down to 450 Hz. A wide baffle is much easier to deal with when designing a crossover.
There are similar effects moving the driver up and down on the baffle.
Close to top edge
Lower on baffle seems to produce a smoother response.
On the narrow baffle I loaded it into I had the baffle step to contend with in the pass band as I was running the midrange down to 450 Hz. A wide baffle is much easier to deal with when designing a crossover.
There are similar effects moving the driver up and down on the baffle.
Close to top edge
Lower on baffle seems to produce a smoother response.
That mid+tweeter shape with the big roundovers is similar to what I tried before, probably the best sounding thing I've heard, eliminating that diffraction all the way down to the midrange is amazing. I'm currently struggling to come up with a solution that looks visually appealing. Good luck with the build.
Hi Jim,
I'm looking forward to following your design process.
The trapezoidal cabinets have been done in the past. The earliest recollection I am aware of is the JBL L250, designed by Greg Timbers for JBL, and released in 1982:
Some years later, in the mid-1990s here in Australia, some trapezoidal speakers were released by a Dutchman, whose name I unfortunately can't recall.
The company was called Mass Technologies, who were Australian importers for Vifa and Scan-Speak at that time. These model is fully active:
However, I find the fully trapezoidal cabinets a bit hard on the eyes.
Have you considered partially trapezoidal?
eg. Peak Consult (DK)
Having done my own small directivity study, I believe you are on the right path with the use minimal baffle near the drivers.
As @fluid has demonstrated, directivity is also shaped by the depth of the cabinet, although it is a bit harder to simulate in VituixCAD2 alone.
BEM by some other software eg. AKABAK may be necessary.
Finally, I also feel that wide baffles is not fully explored. In the days of wide but thinner televisions, is there a role for wider but thinner speakers? How thin can we push it? I spoke to @vineethkumar01 about this after seeing his stunning build
He felt that the minimum cabinet depth may be limited by the depth of the woofers...
Would a wide, but thin cabinet be something you might explore?
Finally, please consider woofers that give the sensitivity that goes towards matching your midrange and tweeter. Of course, you will need to balance box size, bass extension, and available use of amplifier power/EQ for your needs.
Having built a system with dual 12" woofers in a sealed/EQ box with ~110dB capability continuously. Since I've never even had my pre-amp control up to -6dBFS, I feel this is a was too much, in hindsight. Of course, it takes some time to figure out what 105, 110 or 115 dB sound like. I didn't study this before I designed it:
If I were to do this again, I would aim for something with at least 3dB less output capability.
Have fun!
I'm looking forward to following your design process.
The trapezoidal cabinets have been done in the past. The earliest recollection I am aware of is the JBL L250, designed by Greg Timbers for JBL, and released in 1982:
Some years later, in the mid-1990s here in Australia, some trapezoidal speakers were released by a Dutchman, whose name I unfortunately can't recall.
The company was called Mass Technologies, who were Australian importers for Vifa and Scan-Speak at that time. These model is fully active:
However, I find the fully trapezoidal cabinets a bit hard on the eyes.
Have you considered partially trapezoidal?
eg. Peak Consult (DK)
Having done my own small directivity study, I believe you are on the right path with the use minimal baffle near the drivers.
As @fluid has demonstrated, directivity is also shaped by the depth of the cabinet, although it is a bit harder to simulate in VituixCAD2 alone.
BEM by some other software eg. AKABAK may be necessary.
Finally, I also feel that wide baffles is not fully explored. In the days of wide but thinner televisions, is there a role for wider but thinner speakers? How thin can we push it? I spoke to @vineethkumar01 about this after seeing his stunning build
He felt that the minimum cabinet depth may be limited by the depth of the woofers...
Would a wide, but thin cabinet be something you might explore?
Finally, please consider woofers that give the sensitivity that goes towards matching your midrange and tweeter. Of course, you will need to balance box size, bass extension, and available use of amplifier power/EQ for your needs.
Having built a system with dual 12" woofers in a sealed/EQ box with ~110dB capability continuously. Since I've never even had my pre-amp control up to -6dBFS, I feel this is a was too much, in hindsight. Of course, it takes some time to figure out what 105, 110 or 115 dB sound like. I didn't study this before I designed it:
If I were to do this again, I would aim for something with at least 3dB less output capability.
Have fun!
Following hifijim's new adventure... 🙂
For the look, Wilson Watt / Sasha or Joseph Audio Pearl are showing that rather big speaker can look smaller if midrange head is done like trapezoidal pyramide...They also look very elegant to me. And could be done as a stack of basses with mid pyramide on top. It is easier to work with few boxes than with one very heavy...
For the look, Wilson Watt / Sasha or Joseph Audio Pearl are showing that rather big speaker can look smaller if midrange head is done like trapezoidal pyramide...They also look very elegant to me. And could be done as a stack of basses with mid pyramide on top. It is easier to work with few boxes than with one very heavy...
I am in love with Avalon cabs ( and the wide SF Stradivari). Compas or Indra style with double woofers but in sealed cab with dsp-active... The Compas in the pic has too large mid me thinks, a dome mid would make it the best possible narrow wde directivity 3-way! Double woofers are needed because of baffle loss, and they can be made .5 passively.
HTGuide Jon Marsh has this passive Ardent https://www.htguide.com/forum/missi...builds?41409-Wavecor-Ardent-Design-and-Build=
HTGuide Jon Marsh has this passive Ardent https://www.htguide.com/forum/missi...builds?41409-Wavecor-Ardent-Design-and-Build=
Hi,
Watch "on". As a beginner in diy speaker but an engineer I really like your thread and approach to design speaker.
Regarding baffle, maybe you already now it but there is a thread on ASR where sigberg audio presents its iterative step on sigberg manta design in order to integrate midrange driver on a wide baffle using cardiod slot If I had time I would like to test this idea.
Watch "on". As a beginner in diy speaker but an engineer I really like your thread and approach to design speaker.
Regarding baffle, maybe you already now it but there is a thread on ASR where sigberg audio presents its iterative step on sigberg manta design in order to integrate midrange driver on a wide baffle using cardiod slot If I had time I would like to test this idea.
Attachments
Something similar was already made. T25B + M74B + PTT6.5 NAA
https://yb-audio.pl/realizacje/destino/
https://yb-audio.pl/realizacje/destino/