Design introduction - The Modern Minimus 7 (MM7)
Hi everyone. I'd like to share a design that I'm really excited about. As a kid I loved the Minimus 7 when used with the matching receiver with EQ. The EQ makes the small speaker sound really good with great bass for such a small speaker. I figured, wouldn't it be fun to do a modern version of the classic speaker and see how much more performance we can squeeze out with DIY?
The original speaker is a 1" dome + 4" midwoofer in a 2L sealed box. Let's copy that form factor but with modern twists. The drivers will showcase the advancement in modern driver design over the past 50 years.
Components:
Originally, the speaker had the same front dimensions as the Minimus 7 wood version, which was bigger than the aluminum version. The reason is to simplify the build. Then I realized, the original Minimus 7 had surface mounted drivers, and if I also do surface mounted drivers, I can truncate the Purifi woofer frame and make the front baffle smaller without making it any harder to build. I can also push the woofer and waveguide right next to each other to save space, something I can't do with a flush mount since they need a small amount of space in between for rigidity. This shaved off 0.7” on both the 5" width and the 8" height, now down to 4.3" wide x 7.3" tall x 6" deep (1" deeper than before to maintain box volume). This doesn’t sound like much, but going from 5” to 4.3” is actually 14% smaller and it looks noticeably smaller. Now the front dimensions are exactly the same as the classic aluminum Minimus 7. I really like small speakers, and I love that I found a way to make it smaller without making it harder to build, which is usually the problem when trying to squeeze every bit of size out of them.
The original speaker had a rectangular faceplate for the tweeter. To match that visual style, I will be using a 4” rectangular waveguide for the tweeter based off of Somasonus’s 4” waveguide. Please excuse the crappy CAD drawing and the green waveguide that isn't edge to edge, I suck at CAD, but you guys get the picture. This brings an improvement in the sensitivity, directivity, and distortion performance of the Bliesma T25B tweeter, while following the visual style of the original speaker. Form and function, win!
While the drivers are no longer flush mounted, and this is not ideal acoustically, I am deciding to go this route in order to pay homage to the look of the original speaker, and because the negative effects of diffraction is minimized due to the use of a waveguide and there’s actually very little surface area left on the baffle that isn’t covered by the woofer or waveguide. It’s a small compromise I’m happy to make. This speaker isn’t meant to be completely acoustically optimal. The priority is to make something fun that pay homage to a classic.
Simulations are very promising. The Purifi 4" is an amazing woofer with such a long Xmax and strong motor for such a small driver, exactly what I need for deep bass output from a small box. With 100W/ch, 2L is all that's needed to get essentially everything out of the woofer. That's reasonable on the amplifier power and on the driver's power handling capabilities. This is possible from the very strong motor on the Purifi woofer giving it great deep bass efficiency in a small box.
The simulation shows about 98dB at 40Hz and 93dB at 32Hz from a pair. Subtract 2dB for losses like power compression and PR losses and there is still over 95dB at 40Hz before room gain and room modes. To get a rough comparison, because the Purifi has more than 2x the clean excursion of a normal 4" woofer, it plays more like a normal 6" woofer. Not bad!
In future posts I will go into much greater detail on the design choices to maximize performance out of a speaker, particularly a small speaker. I look forward to sharing more. I think this speaker will end up being a pleasant surprise to see how far things have come since 1977
Hi everyone. I'd like to share a design that I'm really excited about. As a kid I loved the Minimus 7 when used with the matching receiver with EQ. The EQ makes the small speaker sound really good with great bass for such a small speaker. I figured, wouldn't it be fun to do a modern version of the classic speaker and see how much more performance we can squeeze out with DIY?
The original speaker is a 1" dome + 4" midwoofer in a 2L sealed box. Let's copy that form factor but with modern twists. The drivers will showcase the advancement in modern driver design over the past 50 years.
Components:
- Tweeter: Bliesma T25B, one of the finest small faceplate 1" tweeters available.
- Waveguide: Modified Somasonus 4" waveguide for the Bliesma T25B in a rectangular frame rather than oval
- Woofer: Purifi 4X to get the best midrange and bass performance out of a 4" driver.
- Passive radiator: Purifi 4" passive radiator on the back tuned to 33Hz to help get ~5-6dB more bass at the 30-50Hz range.
- DSP and amplifier: Ashly FX125.4. 4x125W FIR DSP amplifier. The speaker will be active with DSP, and the aim is to get a flat response to an unusually low 32Hz in order to cover 99% of musical bass. DSP is a wonderful modern technology to allow for much deeper than normally possible bass out of small speakers.
Originally, the speaker had the same front dimensions as the Minimus 7 wood version, which was bigger than the aluminum version. The reason is to simplify the build. Then I realized, the original Minimus 7 had surface mounted drivers, and if I also do surface mounted drivers, I can truncate the Purifi woofer frame and make the front baffle smaller without making it any harder to build. I can also push the woofer and waveguide right next to each other to save space, something I can't do with a flush mount since they need a small amount of space in between for rigidity. This shaved off 0.7” on both the 5" width and the 8" height, now down to 4.3" wide x 7.3" tall x 6" deep (1" deeper than before to maintain box volume). This doesn’t sound like much, but going from 5” to 4.3” is actually 14% smaller and it looks noticeably smaller. Now the front dimensions are exactly the same as the classic aluminum Minimus 7. I really like small speakers, and I love that I found a way to make it smaller without making it harder to build, which is usually the problem when trying to squeeze every bit of size out of them.
The original speaker had a rectangular faceplate for the tweeter. To match that visual style, I will be using a 4” rectangular waveguide for the tweeter based off of Somasonus’s 4” waveguide. Please excuse the crappy CAD drawing and the green waveguide that isn't edge to edge, I suck at CAD, but you guys get the picture. This brings an improvement in the sensitivity, directivity, and distortion performance of the Bliesma T25B tweeter, while following the visual style of the original speaker. Form and function, win!
While the drivers are no longer flush mounted, and this is not ideal acoustically, I am deciding to go this route in order to pay homage to the look of the original speaker, and because the negative effects of diffraction is minimized due to the use of a waveguide and there’s actually very little surface area left on the baffle that isn’t covered by the woofer or waveguide. It’s a small compromise I’m happy to make. This speaker isn’t meant to be completely acoustically optimal. The priority is to make something fun that pay homage to a classic.
Simulations are very promising. The Purifi 4" is an amazing woofer with such a long Xmax and strong motor for such a small driver, exactly what I need for deep bass output from a small box. With 100W/ch, 2L is all that's needed to get essentially everything out of the woofer. That's reasonable on the amplifier power and on the driver's power handling capabilities. This is possible from the very strong motor on the Purifi woofer giving it great deep bass efficiency in a small box.
The simulation shows about 98dB at 40Hz and 93dB at 32Hz from a pair. Subtract 2dB for losses like power compression and PR losses and there is still over 95dB at 40Hz before room gain and room modes. To get a rough comparison, because the Purifi has more than 2x the clean excursion of a normal 4" woofer, it plays more like a normal 6" woofer. Not bad!
In future posts I will go into much greater detail on the design choices to maximize performance out of a speaker, particularly a small speaker. I look forward to sharing more. I think this speaker will end up being a pleasant surprise to see how far things have come since 1977
Last edited:
Just spitballing here but there was an Australian only small speaker [ whose model number I forget] that used a 4" woofer plus a co-axial tweeter on a 4" midrange. I'd try to emulate that if better bass in a tiny box was the aim
I did a revamp a few years back called the Synchaetas. Used a PRV 4MR60-4 for the woofer and an ND25 tweeter. I got an F3 sealed at about 100Hz and an F10 in the 50Hz range, and a worthwhile sensitivity of 86dB.
I need to get the info posted in the InDIYana armada pages.
I need to get the info posted in the InDIYana armada pages.
A small comment on this draft:
In my opinion, a passive radiator with a Vd equal to that of the active bass driver unnecessarily restricts the level stability. It would be better to have twice as much Vd compared to the active bass driver.
In my opinion, a passive radiator with a Vd equal to that of the active bass driver unnecessarily restricts the level stability. It would be better to have twice as much Vd compared to the active bass driver.
The choice to use a single passive radiator is also an intentional choice and an enormous amount of consideration went into this. The standard 2x PR is not as ideal as conventional wisdom would indicate, and 1 PR is not as bad as conventional wisdom would indicate
- In terms of displacement, 1 PR is not as bad as it seems. The woofers would only be used to about 7-8mm Xmax, and the PR has 14mm Xmech. Call it 12mm usable Xmax on the PR, and it is about 1.7x the displacement of the woofer.
- Real world displacement is lower than simulations for a number of reasons. It assumes PR's do not experience Kms compression. It assumes the woofer does not experience BL and thermal compression. In this tiny box the woofer will experience at least 1-2dB of power compression at max power, and 1-2dB from BL compression. Even though it looks like 1 PR isn't enough from simulations, after accounting for losses, 1 PR is enough all the way to the tuning frequency and it is only below tuning does the single PR actually truly run out of excursion.
2 PR's are also have more practical issues in the real world that people don't realize until they start pushing things hard.
Also, 2 PR's must be put on the side baffles of the enclosure, which doesn't look good in my opinion because it'll cover most of the side panel given how small the speaker is. I prefer the look of all wood on the side panels and the single PR at the back is nice and hidden away. It looks a lot better. It also prevents the box from being braced.
When you look at all the design choices, a single PR is not nearly as compromised as it looks
- In terms of displacement, 1 PR is not as bad as it seems. The woofers would only be used to about 7-8mm Xmax, and the PR has 14mm Xmech. Call it 12mm usable Xmax on the PR, and it is about 1.7x the displacement of the woofer.
- Real world displacement is lower than simulations for a number of reasons. It assumes PR's do not experience Kms compression. It assumes the woofer does not experience BL and thermal compression. In this tiny box the woofer will experience at least 1-2dB of power compression at max power, and 1-2dB from BL compression. Even though it looks like 1 PR isn't enough from simulations, after accounting for losses, 1 PR is enough all the way to the tuning frequency and it is only below tuning does the single PR actually truly run out of excursion.
2 PR's are also have more practical issues in the real world that people don't realize until they start pushing things hard.
- Enormous amounts of weight has to be put on the PR if 2 were used, about 120g, which is probably a little too much for a 4" PR.
- This one is the real kicker that few know about. 2 PR's does not actually give 2x more displacement over a single PR. It gives only maybe 1.25-1.5x more displacement. Because PR's are purely mechanical devices, they are extremely sensitive to parameter matching. This means when 2 PR's are used in the same enclosure, the Xmax of each PR is greatly reduced because at some level of excursion, the PR's start becoming nonlinear, and start to drift away from each other and 2 PR's excursions no longer match each other, and the enclosure effectively acts as if it only has 1 PR. The point at which this happens is well below Xmech of the PR, usually at half excursion. Usually most PR's cannot be used anywhere near to their Xmech without making bad noises, so 2x PR's still give a good amount of displacement increase. Purifi PR's can, so 2x PR's don't actually give any meaningful increase in displacement over 1 PR. The only real benefit is the stability from dual opposing PR's, but even that is gone once the PR's go nonlinear at high enough excursions
Also, 2 PR's must be put on the side baffles of the enclosure, which doesn't look good in my opinion because it'll cover most of the side panel given how small the speaker is. I prefer the look of all wood on the side panels and the single PR at the back is nice and hidden away. It looks a lot better. It also prevents the box from being braced.
When you look at all the design choices, a single PR is not nearly as compromised as it looks
The enclosure ALWAYS acts as if it has only one PR or port in a vented system. The multiples act as one tuning with one total mass.
PRs arent required to be flanking the woofer, some designs place both PRs on the rear panel. However, in a small design proposed such as these, this will likely make them oscillate a little on their own.
DanP built a small 2way such as these a couple years ago with singles and the LD25X tweeter, but his volume was closer to 3 ltrs, IIRC.
PRs arent required to be flanking the woofer, some designs place both PRs on the rear panel. However, in a small design proposed such as these, this will likely make them oscillate a little on their own.
DanP built a small 2way such as these a couple years ago with singles and the LD25X tweeter, but his volume was closer to 3 ltrs, IIRC.
I'm not saying that it couldn't work with only one passive radiator, just that the speaker will fall somewhat short of the capabilities of the active driver when it comes to the possible maximum SPL. It's just a compromise that everyone is free to make. 🙂
However, I would like to point out another possible problem, namely that the tuning frequency could be too low if only one passive radiator is used, even without an additional weight screwed on. The only way to raise fb would then be to use a second one.
In addition to the requirements regarding the displacement volume of the passive radiator, this could be another reason why I have not yet seen a Purifi build with fewer than two passive radiators per active bass driver (although I certainly haven't seen them all 😏).
Have you ever checked this with a simulation programme? The Enclosure Tool from VituixCAD is very well suited for simulations, even of constructions with passive radiators.
However, I would like to point out another possible problem, namely that the tuning frequency could be too low if only one passive radiator is used, even without an additional weight screwed on. The only way to raise fb would then be to use a second one.
In addition to the requirements regarding the displacement volume of the passive radiator, this could be another reason why I have not yet seen a Purifi build with fewer than two passive radiators per active bass driver (although I certainly haven't seen them all 😏).
Have you ever checked this with a simulation programme? The Enclosure Tool from VituixCAD is very well suited for simulations, even of constructions with passive radiators.
it’s all about max volume displacement of the PR relative to the woofer. However, the total PR mass for a given tuning scales with Sd squared so having multiple PRs means that each need more mass. this tubes their free resonances down which gives more bass extension. The combined response has a null at the free air PR resonance (PR is completely transparent at that frequency and short circuits the pressure)
Yes, I know that. What I don't know: what does that have to do with my last post?
But I also don't want to rule out the possibility that my post may have been misleading. My mother tongue is not English and I also use translation programmes.
But I also don't want to rule out the possibility that my post may have been misleading. My mother tongue is not English and I also use translation programmes.
Here's my thoughts.
If you do any of those things it's no longer a Minimus clone.
The attributes that made the Minimus special were its price performance ratio and tiny size.
Change either of those and it's just another small 2-Way speaker
If you do any of those things it's no longer a Minimus clone.
The attributes that made the Minimus special were its price performance ratio and tiny size.
Change either of those and it's just another small 2-Way speaker
I built speakers using the Minimus 11 cabinets. I used a wave guide and big CD. I’m going to replace the wave guide with a flat ribbon tweeter just to get more internal cabinet space to allow the woofer to go lower. Just something to think about.
I just saw an advertisement for this Minimus 7 speaker
In 1978 it was $49.95 each...
So die cast aluminum cabinets, off-shored to the far East ("custom manufactured in Japan) has been happening since at least 1977...
In 2025, $US49.95 is equivalent around US$250 each, so US$500 a pair
In 1978 it was $49.95 each...
So die cast aluminum cabinets, off-shored to the far East ("custom manufactured in Japan) has been happening since at least 1977...
In 2025, $US49.95 is equivalent around US$250 each, so US$500 a pair
I think you refer to Braun L300. I believe it is the smallest 3-way speakers : ).
Attachments
Was working at Rat-shack when these were out...burned a set up "exploring" ultimate power handling. That tight volumes of the M7 is the key to those working as good as they do...don't be breaking loose the driver seal while they are playing...even at moderate volumes you'll hit Xmax mechanical right away. I wouldn't bother with the passive radiator...at those low frequencies, the distances are so short, when the driver advances positive creating that pressure wave, as it travels forward & curls around the sides...halfway along the side it's going to encounter the rarefied pressure wave from the radiator...never to advance into the open air.
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Rick...
---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------Rick...
No; the RatShack was very different. I no longer have pictures as I flipped it on eBay almost immediately for about $150-, It was definitely a Tandy product
The woofer runs full range, and the "Deluxe L-C crossover" is this:
So the sound sucks
The measurements suck:
No mid-bass, and infant screaming piercing treble
How were people designing 2 way speakers in 1977?
FFT?
Ear?
Doug Rife's MLSSA was not released until a decade later until 1987. It was a 12bit ADC "audio interface" that you installed into your computer (USB hadn't been invented yet), capable of sampling in excess of 100KHz for measurements past 40KHz.
The iNTEL 80386 processor had been released 2 years prior in 1985. MLSSA needed the additional 80387 math co-processor which cost US$800 in 1987… $2000 in today’s money) just to run. By 1990, Stereophile had adopted it, was "It's cheap. Just $2750 buys the complete package: a long card to fit in one of a PC's expansion slots, and all the software."
By 1991, PCs were getting better- a PC with a math-processor cost US$1500, and MLSSA now costed $3000= US$4500 => US$10,000 in 2025
Audiophiles think that measurements don't matter. But they're not the ones making anything. Just consuming.
From processors to loudspeakers. Simulation, manufacturing and verification IS important in R&D.
But I'm probably preaching to the converted.
So the sound sucks
The measurements suck:
No mid-bass, and infant screaming piercing treble
How were people designing 2 way speakers in 1977?
FFT?
Ear?
Doug Rife's MLSSA was not released until a decade later until 1987. It was a 12bit ADC "audio interface" that you installed into your computer (USB hadn't been invented yet), capable of sampling in excess of 100KHz for measurements past 40KHz.
The iNTEL 80386 processor had been released 2 years prior in 1985. MLSSA needed the additional 80387 math co-processor which cost US$800 in 1987… $2000 in today’s money) just to run. By 1990, Stereophile had adopted it, was "It's cheap. Just $2750 buys the complete package: a long card to fit in one of a PC's expansion slots, and all the software."
By 1991, PCs were getting better- a PC with a math-processor cost US$1500, and MLSSA now costed $3000= US$4500 => US$10,000 in 2025
Audiophiles think that measurements don't matter. But they're not the ones making anything. Just consuming.
From processors to loudspeakers. Simulation, manufacturing and verification IS important in R&D.
But I'm probably preaching to the converted.
Last edited: