Let’s make a dedicated thread for this project to keep things consistent. Links to relevant threads are appreciated! 🙏
This project will proceed at its own pace (likely slowly). The goal is to build a relatively risk-free, "safe bet" speaker with minimal iterations and prototyping.
Project Goals
I’ve been eager to start a new project. My current Audio Note setup sounds wonderful but has reached its peak development for now. I don’t expect this build to replace that system, but for variety’s sake, it would be fun to build something myself. While I’m not experienced in advanced speaker design theory, I understand basic principles and have a good grasp of furniture and component design as well as common manufacturing methods and materials.This project will proceed at its own pace (likely slowly). The goal is to build a relatively risk-free, "safe bet" speaker with minimal iterations and prototyping.
Ballpark Concept
- Size: Bookshelf/monitor, max height ~500 mm
- Enclosure: Sealed
- Configuration: 2-way preferred for simplicity, but I’ll consider 3-way if benefits (like lower f3) are predictable and budget-friendly
- Drivers: Bass driver 8" or 10" but curious…
- Amplification: Hypex Fusion or equivalent
- Budget: Max €2000 per pair (excluding labor and cabinet materials)
Design Rationale
- Sealed Enclosure: Works better in my space than vented solutions. I value the responsiveness of sealed designs over absolute bass extension.
- Target Frequency Response: Ideally around 40 Hz in-room. If that’s not achievable, the design should allow easy future integration of a bass module or active subwoofer for possible future, higher SPL situations like psychedelic garage parties with beer bongs and people in weird clothing.
- Performance Focus: Prioritize good value for money and performance within budget constraints.
- Listening Environment: Apartment with neighbours, 3-meter listening distance, typical listening level < 75 dB.
(Very Loose) Inspirations & Preferences
- Consider a [sealed configuration] Antelope Atlas i8 as very rough reference
- Dynaudio Focus 10 as another…
- An ideal, financially completely unrestrained (and boring) option would be a Dutch & Dutch 8c, delivered and custom setup at my home while drinking my expensive wine in my tweed jacket and claiming to know many things others don't
- Paper-cone woofers preferred for their pleasant sound, though not mandatory
- Controlled directivity is a plus but not essential
Current Component Ideas
- Woofer:
SB Acoustics SB23NRXS45-8
(8-inch) - Tweeter:
Wavecor TW030WA11
(low fs of 425 Hz and a waveguide) - Amplification:
Hypex Fusion FA122 - Cabinet: ~25 liters
- Crossover: Target ~1.6 kHz
Questions for the Community
- Is the proposed 25-liter cabinet size appropriate for the SB Acoustics woofer?
- Thoughts on driver matching, particularly the Wavecor tweeter and crossover frequency?
- Any other driver/enclosure recommendations within the budget?
- Would you rather fight one horse-sized duck or 100 duck-sized horses?
https://www.usaudiomart.com/details/650160496-active-bookshelf-speakers/
In case you might be interested
In case you might be interested
Seems interesting and very appropriate! Shame about the limited shipping 🤷‍♂️
I could very well consider that design đź‘Ť
I could very well consider that design đź‘Ť
Fwiw,
You can’t go wrong with Hypex.
Give some thought to getting the FA253. It’s a bit larger footprint and 3 channels but if you’re buying new the difference between the 253 and 122 isn’t too bad. Then you always have the option of going to a 3way design. SB acoustics makes some excellent well priced drivers. Remember too with dsp you can really enhance a middle of the road driver to quite acceptable levels, in all respects. Imo top of the line drivers ( Satori and Purifi come to mind) may not add quite as much value. If you go with Hypex their filter design software is pretty good but will take time to learn and polish.
Best of luck
I’ll be following along
You can’t go wrong with Hypex.
Give some thought to getting the FA253. It’s a bit larger footprint and 3 channels but if you’re buying new the difference between the 253 and 122 isn’t too bad. Then you always have the option of going to a 3way design. SB acoustics makes some excellent well priced drivers. Remember too with dsp you can really enhance a middle of the road driver to quite acceptable levels, in all respects. Imo top of the line drivers ( Satori and Purifi come to mind) may not add quite as much value. If you go with Hypex their filter design software is pretty good but will take time to learn and polish.
Best of luck
I’ll be following along
I don't like the hypex dsp in those plateamps, and would more insist on a minidsp flex, In my experience that dsp is very buggy and unreliable. The amps are good, but plate amps die fast, so, if you use them, make a seperate box (don't use the amp as back of the chambre) so you can replace the amp with a different model without compromising the speaker box itself.
A MiniDSP Flex with seperate amps (Ncore or other) will be better, but then i will blow your budget if you also want Ncore amps. The cheapest i know that are good are 400€ for 2channel 75w/8R. That could work for nearfields without a ub if you have 2 amps, but the minidsp flex and amps toghetter are already about 1500€, so 500€ left for the rest...
I would use a different tweeter, the SB26ADC in an Soma Sonus waveguide coupled to a woofer like the SB24WO24P-4 can be great, An Dayton RS180-4 could also work with limited bass. The WO24P gets the response you want without eq, the RS180 will need eq but is a lot cheaper (and still good).
A MiniDSP Flex with seperate amps (Ncore or other) will be better, but then i will blow your budget if you also want Ncore amps. The cheapest i know that are good are 400€ for 2channel 75w/8R. That could work for nearfields without a ub if you have 2 amps, but the minidsp flex and amps toghetter are already about 1500€, so 500€ left for the rest...
I would use a different tweeter, the SB26ADC in an Soma Sonus waveguide coupled to a woofer like the SB24WO24P-4 can be great, An Dayton RS180-4 could also work with limited bass. The WO24P gets the response you want without eq, the RS180 will need eq but is a lot cheaper (and still good).
I've been playing around with a couple Hypex plate amps (FA253) for about a year now. The software is finicky, but I haven't had any issues once they're set and just using them.
waxx's suggestion for a "safe bet" speaker pairing is spot on.
A 3-way about the size of your inspiration speakers size wise is going to require a coaxial. They can get weird.
waxx's suggestion for a "safe bet" speaker pairing is spot on.
A 3-way about the size of your inspiration speakers size wise is going to require a coaxial. They can get weird.
Thanks everyone 🙏 Much food for thought...
I have only a few daily windows of opportunity to follow up so please bear with me.
What kind of enclosure and which drivers are you using / contemplating?
I wasn't aware of Hypex reliability issues. What's a typical plate amps lifespan and how do they usually fail?
I have considered a Flex for preamp duty and I'm also curious about what Dirac can bring to the table. I think you make a good argument regarding reliability and separate amps but the I'm not fond of the added clutter. I can live with the initial setup complexity and finicky software, but after that I would prefer these to be as "plonk and play" as possible.
I think your driver suggestions fit the brief but I have a few general reservations:
I could consider increased size for a 3-way but I think ultimately, a 3-way would add complexity to component selection, setup and budget. I think a 3-way, in my case, would have to be a relatively proven design.
I have only a few daily windows of opportunity to follow up so please bear with me.
Can you elaborate on implementation and why this would be interesting? Being quite expensive, a pair of those would leave very little budget for remaining components.
Oh nice, please share!
What kind of enclosure and which drivers are you using / contemplating?
Thanks!
I wasn't aware of Hypex reliability issues. What's a typical plate amps lifespan and how do they usually fail?
I have considered a Flex for preamp duty and I'm also curious about what Dirac can bring to the table. I think you make a good argument regarding reliability and separate amps but the I'm not fond of the added clutter. I can live with the initial setup complexity and finicky software, but after that I would prefer these to be as "plonk and play" as possible.
I think your driver suggestions fit the brief but I have a few general reservations:
- I have no experience in waveguides. I understand their benefits but with so many options, which one would you choose and why? The Soma Sonus graphs mention "PS"; what is that?
- My experience with aluminum domes is less than flattering. I'm sensitive to 2 kHz - 5 kHz range and I've found they are not always the smoothest operators. I also understand that waveguides can exaggerate their deficiencies even further unless they are well designed.
- The SB woofer is at the upper limit of the budget but not completely out of consideration.
- The Dayton is interesting. How would you estimate in-room bass response compared to the SB?
I could consider increased size for a 3-way but I think ultimately, a 3-way would add complexity to component selection, setup and budget. I think a 3-way, in my case, would have to be a relatively proven design.
Here it is
Thread '3-way to active - Hypex FA253 - learning project'
https://www.diyaudio.com/community/threads/3-way-to-active-hypex-fa253-learning-project.421061/
Thread '3-way to active - Hypex FA253 - learning project'
https://www.diyaudio.com/community/threads/3-way-to-active-hypex-fa253-learning-project.421061/
I would overthink the whole project for some reasons.
If you speculate on using a subwoofer in the near future, a 8" mid-bass driver is a bad choice for a two way. Any 8" driver will have serious cone break up when driven up to 2000 Hz. You only use an 8" if you want a power 2-way that can do stand allone 40Hz with quite some level.
If you want real sound quality, better use a smaller mid-bass not more than 6". Now these have very limited bass output, but you can make this an advantage. Simply use two of them, the tweeter in the middle. Yes, Joe's old principle. If you are deep into HIFI, you will find very little serious and founded critic of the theory Joseph A. D'Appolito developed. There have been some modifications on the rules over the years, but it is still one of the best ways to build a HIFI speaker.
From personal experience I prefer it over other ways to position drivers on a baffle, not only for the great sound stageing. If you (sometimes) like life like SPL, nothing compares to a speaker that simply doesn't produce audible distortion at deafening levels. Usually you except rising distortion with rising SPL, that is why you ruin your speakers with too much alcohol ingested. Ah, sorry, I forgot, you Finn's don't drink. No point for you. That distrotion doesn't happen with such constructions, when you restrict them to 80 Hz. There is even a compromise to use them stand allone: Do them in individual vented volumes and close these for subwoofer use. You got the budget to use two very high quality 5" for example. You get nearly +6dB from them, for double cone area and double amp power, which should get close to a single 8" driver. By the way, I prefer a sup under every main speaker if possible, so you basically have a separated 3-way speaker.
I'm very critical on plate amps. (I don't talk about their much to high price point.) So first, integrating them into a speaker distracts you from an acoustically optimal cabinet. Second, heat and vibration limit their life im most cases, not to speake about negative impacts on the components performance. Sad to say, not only with cheap examples...
Once you got over the "it is soo neat to have all in a single box argument" you realize that a mains line and the signal cables problems are more anoying than a 4-strand speaker cable. OK, now you got much more options than the usual Hypex answer.
You will see that there are many better options if you go for separate amps and DSP, maybe even some (PA) amps with integrated DSP. Symetric XLR connections come for free with PA products, not as luxury item like in HIFI. So ground noise is no problem.
Thomann is your friend even near the Arctic:
https://www.thomann.de/fi/kaiuttimi...LCJjdXJyZW5jeSI6MiwibGFuZ3VhZ2UiOjN9&reload=1
If you can put a few amp modules and power supplies in a case and wire them up, even better.
Hope that gave you some new ideas.
If you speculate on using a subwoofer in the near future, a 8" mid-bass driver is a bad choice for a two way. Any 8" driver will have serious cone break up when driven up to 2000 Hz. You only use an 8" if you want a power 2-way that can do stand allone 40Hz with quite some level.
If you want real sound quality, better use a smaller mid-bass not more than 6". Now these have very limited bass output, but you can make this an advantage. Simply use two of them, the tweeter in the middle. Yes, Joe's old principle. If you are deep into HIFI, you will find very little serious and founded critic of the theory Joseph A. D'Appolito developed. There have been some modifications on the rules over the years, but it is still one of the best ways to build a HIFI speaker.
From personal experience I prefer it over other ways to position drivers on a baffle, not only for the great sound stageing. If you (sometimes) like life like SPL, nothing compares to a speaker that simply doesn't produce audible distortion at deafening levels. Usually you except rising distortion with rising SPL, that is why you ruin your speakers with too much alcohol ingested. Ah, sorry, I forgot, you Finn's don't drink. No point for you. That distrotion doesn't happen with such constructions, when you restrict them to 80 Hz. There is even a compromise to use them stand allone: Do them in individual vented volumes and close these for subwoofer use. You got the budget to use two very high quality 5" for example. You get nearly +6dB from them, for double cone area and double amp power, which should get close to a single 8" driver. By the way, I prefer a sup under every main speaker if possible, so you basically have a separated 3-way speaker.
I'm very critical on plate amps. (I don't talk about their much to high price point.) So first, integrating them into a speaker distracts you from an acoustically optimal cabinet. Second, heat and vibration limit their life im most cases, not to speake about negative impacts on the components performance. Sad to say, not only with cheap examples...
Once you got over the "it is soo neat to have all in a single box argument" you realize that a mains line and the signal cables problems are more anoying than a 4-strand speaker cable. OK, now you got much more options than the usual Hypex answer.
You will see that there are many better options if you go for separate amps and DSP, maybe even some (PA) amps with integrated DSP. Symetric XLR connections come for free with PA products, not as luxury item like in HIFI. So ground noise is no problem.
Thomann is your friend even near the Arctic:
https://www.thomann.de/fi/kaiuttimi...LCJjdXJyZW5jeSI6MiwibGFuZ3VhZ2UiOjN9&reload=1
If you can put a few amp modules and power supplies in a case and wire them up, even better.
Hope that gave you some new ideas.
Don't know your budget, but it's superior to the SB23 in sensitivity, power handling, sealed Fb and xmax (about 45hz for 50l). Would need a tweeter that can go low; should match one of the Wavecor 30mms. Looks lower distortion from spec sheet vs hificompass measurements of the SB23, as well, though the charts are not apples to apples. I had some of the NRX1 SB15s and they're not my cup of tea in the midrange; I wouldn't go for them again; not in a 2-way, but that's me.
Last edited:
@Turbowatch2 Thanks for your input. While I appreciate the out-of-the-box reasoning, you are suggesting a very different proposition to my original idea.
Allow me to elaborate further:
I'm familiar with a few commercial D'Appolito designs and to my memory they did sound very good. I understand why they are a common studio monitor design. I think their greatest benefits are a predictable phase behaviour, and generally a very precise sound.
But ever since I've had my AN-K's (sealed, 8") I've been fascinated by the idea of a "modern interpretation" of that design. There's something about a larger, sensitive woofer, in a sealed box that I really enjoy and that I haven't heard reproduced by smaller drivers, even in those D'Appolitos. Furthermore, soundstage and imaging are not that much of a focus in my listening. I much prefer the presence of a larger woofer at the cost of certain technicalities. I also think larger drivers generally sound better at lower listening levels.
Point being, I realise there are probably technically more competent concepts for a speaker, but I'm not sure yours the direction I want to go. So, while I'm a little hesitant about your proposition of smaller drivers for a few subjective reasons, I don't disagree with your reasoning from an objective, performance point of view.
However, the build phase is still probably quite far on the horizon and as I consider your suggestions, the brief may change for a variety of reasons 🤷‍♂️
Regarding the subwoofer(s): I'm hoping to manage without and I want these speakers to perform reasonably well on their own, in my current listening environment. But I may want to add a subwoofer later as life circumstances change and evolve.
As for breakup and beaming, correct me if I'm wrong, but to my understanding a 1.6 - 1.8 kHz XO should be completely viable for an 8" woofer to perform well. It obviously places higher demands on tweeter selection and matching but to my knowledge it should be doable with fairly predictable results.
Ports are a hard no, please see reasoning in my original post.
I think there are a few ways to reduce the risks you bring up regarding plate amplifiers. One of them being an attachable but external and mechanically isolated chassis as per @waxx suggestion. That would bring opportunities to further improve cooling and ultimately reliability. It would also add upgrade potential should I decide to go the Flex - separate amps route in the future, without spoiling the speaker cabinet.
Best,
//S-P
Allow me to elaborate further:
I'm familiar with a few commercial D'Appolito designs and to my memory they did sound very good. I understand why they are a common studio monitor design. I think their greatest benefits are a predictable phase behaviour, and generally a very precise sound.
But ever since I've had my AN-K's (sealed, 8") I've been fascinated by the idea of a "modern interpretation" of that design. There's something about a larger, sensitive woofer, in a sealed box that I really enjoy and that I haven't heard reproduced by smaller drivers, even in those D'Appolitos. Furthermore, soundstage and imaging are not that much of a focus in my listening. I much prefer the presence of a larger woofer at the cost of certain technicalities. I also think larger drivers generally sound better at lower listening levels.
Point being, I realise there are probably technically more competent concepts for a speaker, but I'm not sure yours the direction I want to go. So, while I'm a little hesitant about your proposition of smaller drivers for a few subjective reasons, I don't disagree with your reasoning from an objective, performance point of view.
However, the build phase is still probably quite far on the horizon and as I consider your suggestions, the brief may change for a variety of reasons 🤷‍♂️
Regarding the subwoofer(s): I'm hoping to manage without and I want these speakers to perform reasonably well on their own, in my current listening environment. But I may want to add a subwoofer later as life circumstances change and evolve.
As for breakup and beaming, correct me if I'm wrong, but to my understanding a 1.6 - 1.8 kHz XO should be completely viable for an 8" woofer to perform well. It obviously places higher demands on tweeter selection and matching but to my knowledge it should be doable with fairly predictable results.
Ports are a hard no, please see reasoning in my original post.
I think there are a few ways to reduce the risks you bring up regarding plate amplifiers. One of them being an attachable but external and mechanically isolated chassis as per @waxx suggestion. That would bring opportunities to further improve cooling and ultimately reliability. It would also add upgrade potential should I decide to go the Flex - separate amps route in the future, without spoiling the speaker cabinet.
Best,
//S-P
Waveguides help with dispertion (like a horn, but without gain) and make it easier to let a tweeter play low. The Soma Somus waveguides have a whole (very long) tread with all the info here but are a proven concept: https://www.diyaudio.com/community/threads/open-source-waveguides-for-cnc-3d-printing.318190/
About in room bass response estimation, that is always a guess, but in general in small room you got to 10dB bass gain due to the wall loading. your room has the sizze of my office more or less, and i did play that woofer (with a dayton tweeter) here as sealed while messing with someone else's speaker (to fix it) and it did give good bass. But the real proof of the pudding is off course the eating. It's a good driver with good specs for your purpose and it sims like it could get (without room gain) an F6 of 45Hz, so not that far of your mark, and easy to eq to your mark in nearfield.
About in room bass response estimation, that is always a guess, but in general in small room you got to 10dB bass gain due to the wall loading. your room has the sizze of my office more or less, and i did play that woofer (with a dayton tweeter) here as sealed while messing with someone else's speaker (to fix it) and it did give good bass. But the real proof of the pudding is off course the eating. It's a good driver with good specs for your purpose and it sims like it could get (without room gain) an F6 of 45Hz, so not that far of your mark, and easy to eq to your mark in nearfield.
@super-panda I understand you points quite well. Anyway, there is no pistonical working 8" driver in this world, so from some frequency up there is cone break up of some kind. Some designs even perfected this problem, so at some point only the voice coil and dustcap (dome) are moving.
If you accept this, everything is fine. There is one "But". You got to pick an 8" driver that has a very friendly break up, so the result is what you expect. There are many 8" that will get nasty once you go higher than 1000 Hz, even if you DSP the hell out of them. Usually these will have a cone on the light side, which results in a quite limited bass response. You will have to live with some of these limitations. The whole life is a compromise...
The same is with your personal "vented" problem. A very subjective view that certainly is not right for any construction. I'm with you if you tell me there is a giant army of boomy, bloated, boxy sounding vented constructions out there. On the other hand you can find some of the most linear, well sounding boxes with a vent on the back or a passive radiator. As I say: supjective. If you have the right driver, a vented construction reduces distortion in the low end, compared to a closed, equalized one with the same linearity. Closing the vents in case you add a sub is nothing bad, in fact it is accepted even in high end combinations.
I understand you have this picture of a large 2-way on your mind. The lower you can cut off the tweeter, the simpler the search for the woofer will get. The downside is the maximum SPL the tweeter can deliver. Choosing the TW030WA driver is a possible direction. Look on some (independent) measurements of the distortion. You should not use it under the frequency where it rises sharply. An active setup has the advantage to make an 18dB/oct or higher low cut possible, without the sound getting boring, as too many passive crossover parts tend to do.
Your 8" will not be the subwoofer type with a high mms. There are quite some options, even with plastic or metal cones, even as I would look for paper first.
This seems to be the most benevolent material, even as it is mixed with carbon, Kevlar or banana fibre today.
If you want a cheap set-up for experiments, maybe get two BRU5 TPA3255 amps from Aliex. These have a build in DSP. The downside is, you only can program both channels identically as far as I have understood. So you need one for the low and one for high end. For less than 40$ each (plus some power supply you may already have), these are quite nice and powerfull enough, depending on the supply. If everything works out well, you may replace them with HYPEX later.
If you accept this, everything is fine. There is one "But". You got to pick an 8" driver that has a very friendly break up, so the result is what you expect. There are many 8" that will get nasty once you go higher than 1000 Hz, even if you DSP the hell out of them. Usually these will have a cone on the light side, which results in a quite limited bass response. You will have to live with some of these limitations. The whole life is a compromise...
The same is with your personal "vented" problem. A very subjective view that certainly is not right for any construction. I'm with you if you tell me there is a giant army of boomy, bloated, boxy sounding vented constructions out there. On the other hand you can find some of the most linear, well sounding boxes with a vent on the back or a passive radiator. As I say: supjective. If you have the right driver, a vented construction reduces distortion in the low end, compared to a closed, equalized one with the same linearity. Closing the vents in case you add a sub is nothing bad, in fact it is accepted even in high end combinations.
I understand you have this picture of a large 2-way on your mind. The lower you can cut off the tweeter, the simpler the search for the woofer will get. The downside is the maximum SPL the tweeter can deliver. Choosing the TW030WA driver is a possible direction. Look on some (independent) measurements of the distortion. You should not use it under the frequency where it rises sharply. An active setup has the advantage to make an 18dB/oct or higher low cut possible, without the sound getting boring, as too many passive crossover parts tend to do.
Your 8" will not be the subwoofer type with a high mms. There are quite some options, even with plastic or metal cones, even as I would look for paper first.
This seems to be the most benevolent material, even as it is mixed with carbon, Kevlar or banana fibre today.
If you want a cheap set-up for experiments, maybe get two BRU5 TPA3255 amps from Aliex. These have a build in DSP. The downside is, you only can program both channels identically as far as I have understood. So you need one for the low and one for high end. For less than 40$ each (plus some power supply you may already have), these are quite nice and powerfull enough, depending on the supply. If everything works out well, you may replace them with HYPEX later.
My experience is similar,
While I haven’t changed the filters I have found the Hypex to be very consistent. No issues whatsoever. I have 2 pairs of active speakers. Both the 122 and the 252 plate amps have been excellent.
Both pair of speakers originally came with power ice plate amps (minidsp) and both had issues that eventually led to replacement. I was never able to determine the root cause of the failure. Maybe separating the dsp from the amp as a stand alone module is a good idea.
Best