That is for on wall with sub support. This is for on the floor with the option not to use subs. The smaller size of an on-wall leans away from using cardioid while the larger floorstander speaker leans towards it. Mind you if the top was large enough to reach 80 Hz then that could be the on wall speaker. Move it to sit on a bass extender for higher SPLs. Perhaps a 12" coaxial like above? But that won't use up some of the drivers I have sitting on shelves. I don't seem to be learning some lessons about group projects.
I was just stating an example of a spec. In my particular usecase, I have those speakers for nearfield application, where I sit hardly 1m away from them. In that case, a clean 86dB at 1m was plenty enough (though the speaker itself is capable of much more). Your spec looks better in terms of defining a spec for this project.
I just mean't that including a DSP in this project in general might make it easier and somewhat future proof given that DSPs and DSP amps for home speaker use are becoming more popular these days. I didn't mean restricting to any particular type of DSP or DSP amp. That I think any eventual user of this speaker can take a call I guess. Given that the Raspberry Pi is a capable DSP, a version of the speaker can use that. Somebody who already has a minidsp with separate amps can use that as well (if FIR filtering is not strictly needed in the crossover), etc.
I guess someone has to handle this part of the project if we are to develop a really optimized speaker in terms of above-mentioned aspects. I am personally a beginner to ABEC/AKABAK, so I would not be of much use in terms of speaker development using these tools.
Just for inspiration, a couple of recent cardioid projects from ASR..
I find reading the posts of the developers of the speakers very useful, especially the attached project PDF document in one of the threads:
1) https://www.audiosciencereview.com/...ith-double-waveguide-and-cardioid-bass.51866/
Project document here: https://audiosciencereview.com/foru...aveguide-and-cardioid-bass.51866/post-1964728
2) https://www.audiosciencereview.com/...on-using-global-optimization-algorithm.61208/
One more old thread:
3) https://www.audiosciencereview.com/...adiation-via-lateral-slots-like-d-d-8c.37863/
I find reading the posts of the developers of the speakers very useful, especially the attached project PDF document in one of the threads:
1) https://www.audiosciencereview.com/...ith-double-waveguide-and-cardioid-bass.51866/
Project document here: https://audiosciencereview.com/foru...aveguide-and-cardioid-bass.51866/post-1964728
2) https://www.audiosciencereview.com/...on-using-global-optimization-algorithm.61208/
One more old thread:
3) https://www.audiosciencereview.com/...adiation-via-lateral-slots-like-d-d-8c.37863/
Two types of modular, floor-standing tower-style controlled directivity speakers
1) The slim version (From Ascilab)
2) The fat version (https://ggntkt.de/en/model-m3/technische-daten/)
the slim version and the fat version, both can be optimized to match relevant technical specs as shown above.
Ultimately, it is the tastes/preferences of the builder that matter which type he/she can tolerate for the use case.
1) The slim version (From Ascilab)
2) The fat version (https://ggntkt.de/en/model-m3/technische-daten/)
the slim version and the fat version, both can be optimized to match relevant technical specs as shown above.
Ultimately, it is the tastes/preferences of the builder that matter which type he/she can tolerate for the use case.
Attachments
The flat version could be returning to the original id (a monkey box type of speaker) without compromise on sound. So maybe see if you can keep the 12"-5"-1" (more or less) config in a version of that, with cardioide response. That would be a great thing, and would be something i may be interested in building it if it's not to hard to do. It's not fully true to the concept of the 70's monkey coffin, but close enough maybe to be the 21st century version of it.
First we need to firm up on the main objectives such as position in the room and the degree to which the room is included in the design. Size is another. Then with a reasonably firm idea of what we are trying to achieve the range of driver configurations can be considered. I would like to do this via reasonably accurate simulation so that trading the pros and cons can be done in a quantitative rather than qualitative way. An alternative is to crack on and build a prototype, measure, building another, measure,... while seeking to keep ones spirits up.
The width of the wide speaker is 500mm with 100mm clearance either side (at a guess - needs checking) giving 700mm. Near a wall it would need 100mm clearance and will form a cavity which may cause issues. It is likely better out in the room. In my current living area I have got enough clear space near a wall for one speaker but the other is too tight. Plus it only uses an 8" woofer which with cardioid sound cancellation may fail to be clean enough at standard levels at 3-4m. We might need to go a bit bigger or drop the technical performance requirement.
The thin speakers fit comfortably in the room. Not that keen on the looks though. A 300-400mm square section might look better and would fit though not comfortably with one of the speakers.
Another option might be to design for optimum performance out in the room, support reasonable against the wall operation and put some wheels on.
The elephant in the room with a modern 3-way speaker is the crossover. I can collect 3 drivers for less than 200€ that will challenge any commercial HIFI product for more than 1000€ a piece. The problem is a passive crossover. While the OEM pays maybe 30€ for a complete, 3-way quality crossover on a PCB, even the coil of the low frequency part of a DIYS product will cost more than that. So any price-performance ratio is ruined by the extremely overpriced passive components that will cost a fortune for a state of the art 3-way speaker.
There is a way out, which in the minds of most HIFI interested folks is expensive and complicated: the active speaker.
Thanks to our Chinese friends, we can do a high quality active set-up for less than a passive crossover, with all the advantages of a DSP.
We can get 6 channels of DSP amplification with a solid output of 2x300W rms for about 200€. So thats 100€ per side. If you think the TDA3255 or any D-type is a bad amp, don't read any further.
Now things look better. The DSP doesn't care about the low crossover frequency of a woofer. Try to do a 200Hz 12dB/oct or even lower and steeper crossover in passive and you will understand. The coils and caps needed will be more expensive than all your high quality drivers combined.
There is a (still) very cheap DSP TDA3255 amp at Aliex from Wondom/ Sure/ Berybak. You will get it delivered to your door for less than 50$.
Combine 3 of those with a 600 Watt SMPS and you are done.
I know there are some that hate such active solutions, but if you are objective, not even 1% of DIYS speaker builder are able to tame a 3-way passive crossover. We can often see those passive x-over simulations made by people that trust in them. They usually use a huge number of components to (theoretically) deliver a flat response. Simulations are a great tool for people that know how to do a crossover from scratch. Those that only can do it with a simulation will not see how useless some of the expensive parts are. When you really build such a simulation and measure your finished speaker, be prepared for a bad surprise: it will be not flat at all. Now you have spend all your money on the passive parts and don't know what has gone wrong.
With your DSP you simply change a few settings and can correct at no extra cost, with your passive crossover gone wrong you have to buy more expensive parts... until you give up and live with a half well working speaker.
There is a way out, which in the minds of most HIFI interested folks is expensive and complicated: the active speaker.
Thanks to our Chinese friends, we can do a high quality active set-up for less than a passive crossover, with all the advantages of a DSP.
We can get 6 channels of DSP amplification with a solid output of 2x300W rms for about 200€. So thats 100€ per side. If you think the TDA3255 or any D-type is a bad amp, don't read any further.
Now things look better. The DSP doesn't care about the low crossover frequency of a woofer. Try to do a 200Hz 12dB/oct or even lower and steeper crossover in passive and you will understand. The coils and caps needed will be more expensive than all your high quality drivers combined.
There is a (still) very cheap DSP TDA3255 amp at Aliex from Wondom/ Sure/ Berybak. You will get it delivered to your door for less than 50$.
Combine 3 of those with a 600 Watt SMPS and you are done.
I know there are some that hate such active solutions, but if you are objective, not even 1% of DIYS speaker builder are able to tame a 3-way passive crossover. We can often see those passive x-over simulations made by people that trust in them. They usually use a huge number of components to (theoretically) deliver a flat response. Simulations are a great tool for people that know how to do a crossover from scratch. Those that only can do it with a simulation will not see how useless some of the expensive parts are. When you really build such a simulation and measure your finished speaker, be prepared for a bad surprise: it will be not flat at all. Now you have spend all your money on the passive parts and don't know what has gone wrong.
With your DSP you simply change a few settings and can correct at no extra cost, with your passive crossover gone wrong you have to buy more expensive parts... until you give up and live with a half well working speaker.
Hallo mein lieber Turbo, I believe your assumptions are not that accurate, not as much as you would believe them to be.
The truth is somewhere in between. Passive crossovers can present a problem in certain cases,
and not in other ones. I do not have a problem with using NPE's in x/o circuits, not in the least.
I do have a rule that makes me draw the line (x/o point) over which I won't go. I like passive filters, sized properly,
because they don't break down and sound transparently, and I am not sure this could be said for cheap active dsp.
The truth is somewhere in between. Passive crossovers can present a problem in certain cases,
and not in other ones. I do not have a problem with using NPE's in x/o circuits, not in the least.
I do have a rule that makes me draw the line (x/o point) over which I won't go. I like passive filters, sized properly,
because they don't break down and sound transparently, and I am not sure this could be said for cheap active dsp.
OK. The next task would seem to be to quantitatively evaluate the possible driver layouts, the narrow, squarish and wide cabinet configurations, how to adapt for the presence of the front wall, where to place the nulls, options to include the floor, etc... Hopefully with this information it will enable the pros and cons to be better evaluated and a short list (of hopefully one) to go forward for a more detailed design.
Although there is clearly current interest in cardioid designs, the active crossover and BEM/FEM simulations are rather working against a group project. Does anyone wish to get actively involved with this kind of thing, for this particular design over the next few weeks? Or should we accept this is more of an open solo project? I am happy either way but would like to sort it.
Although there is clearly current interest in cardioid designs, the active crossover and BEM/FEM simulations are rather working against a group project. Does anyone wish to get actively involved with this kind of thing, for this particular design over the next few weeks? Or should we accept this is more of an open solo project? I am happy either way but would like to sort it.
IMHO, for mid experienced diy builder (like me) active dsp makes sense for stereo aplication, especially cardioid design. For HT, pasive is better option from price/performance point of view.
My concern about active is limited durability. Passive speaker can work 20-30years easily, with active, you might have troubles in 5-10 years.
In stereo you can buy new dsp plate amps, but for me as HT fan, this will be nightmare. So if you can design reasonably small pasive speaker with controlled directivity a range 100Hz up, it will be masterpiece - because most speakers in HT is placed nearby the wall and will gain a lot from sbir elimination.
My concern about active is limited durability. Passive speaker can work 20-30years easily, with active, you might have troubles in 5-10 years.
In stereo you can buy new dsp plate amps, but for me as HT fan, this will be nightmare. So if you can design reasonably small pasive speaker with controlled directivity a range 100Hz up, it will be masterpiece - because most speakers in HT is placed nearby the wall and will gain a lot from sbir elimination.
It depends how you go about things. Performing DSP on a computer means perhaps buying a cheap system on a chip or maybe using an old one that has become surplus to requirements. Computers come and go but that's modern life. Buying decent older second hand AVRs that lack current connectivity and have low value as AVRs is a cheap way to get amplifier channels. Alternatively a few stereo amplifiers that have become surplus to requirements though not everyone has a bunch in cupboards.
What do tend to fail a lot are cheap class D amplifiers and SMPSs. There would seem to be little reason for DSP electronics to fail unless it is part of such boards. If such failures are likely to be a significant problem then pay more for more reliable hardware.
If DIY is a hobby how bothered is one likely to be when an old project from decades ago fails or drifts way out of spec? In the past I have pulled a few parts and binned the rest with little to no effort made to fix or maintain anything. My interests had moved on. But I guess this may not be the case for everyone.
Well, for me active version means plate amp with dsp and dac onboard, anything else is like Frankenstain 😉
For a manufacturer it might be a cheap approach but for a DIYer it is quite the opposite. £1500 for a pair of 3 channel hypex amps with DSP that lack the channels and possibly the flexibility to tackle this project. They might fit well with premium/prestige projects using premium/prestige drivers but less so more budget conscious projects. No problems - people want all sorts of different things from their hobby interests.
A 3-way plate amp with DSP is going to make this a very expensive project. Someone mentioned reliability. Amps build inside speakers, may it be full range or subs, are the most unreliable of all amps you can buy. Keep the amp out of the cabinet and you will erase the two causes for them failing so often: Heat and vibration. Be happy if your plate amp lasts longer than 4 years. Repair is hardly impossible to integration, SMPS , SMD parts and no shematics released by any brand.
There are so many small low cost, high quality amps on the market, placing some of them somewhere should be no problem. At least if you are able to build a cabinet of living room quality and install some speakers and wires. If you are not into DIYS anyway, why rant against a DIYS solution that will give a superior result. This is a modular system, nothing Frankenstein.
There are so many small low cost, high quality amps on the market, placing some of them somewhere should be no problem. At least if you are able to build a cabinet of living room quality and install some speakers and wires. If you are not into DIYS anyway, why rant against a DIYS solution that will give a superior result. This is a modular system, nothing Frankenstein.
Or just a 6 wire cable instead of 2 for each speaker and a little more elaborate amplifier... not so frankenstein... or?
//
I guess it can be done okish, but combo pc - 2xminidsp with power supply, 4x small tpa3255amp (each with own power brick), box to turn it all on/off, 230v grid to everything, lot of rca cables, 3-4 cables to each speaker... it is hard to do it elegant/living room usable - every box has different design and dimention, you have to keep some space around for cooling... but ok, for stereo, it can be done. And now imagine HT, where you usually need 9-13 speakers + subwoofers + preamp+ ...
There are multichannel amps arround for this ,but they are (relative) expensive if you want decent quality. Even a stack of cheap but good fosi amps and a dsp will get you close to 1K added cost to the project. Therefor a passive version is certainly needed for many.
DSP got many advantages, but the big disadvantage is that it makes a system too complex for many, and to expensive and not fitting the living room also. A passive speaker set has only one power supply (not counting sources and a preamp if you use that) and 2 speaker cables and is plug and play. That is more important for many.
Even a very close friend, who is a music freak and not a technoob at all (she's a high level coder as job, and allround diy'er next to that) does not want an active system for that. She considered the JBL M2, but the complexity of the system let her at the end take the JBL 4367. It may not be as good (very relatively), but it's easy to use with one amp, and no thinkering needed for her needs. I did install Dirac room correction because her amp (NAD M33) had it build in. So one box, one power cable, and two speakers with speaker cable. Everything is well build in and cables are hidden so it fit the asthetics of her living room. A active system, even only a 2 way does not fit there.
DSP got many advantages, but the big disadvantage is that it makes a system too complex for many, and to expensive and not fitting the living room also. A passive speaker set has only one power supply (not counting sources and a preamp if you use that) and 2 speaker cables and is plug and play. That is more important for many.
Even a very close friend, who is a music freak and not a technoob at all (she's a high level coder as job, and allround diy'er next to that) does not want an active system for that. She considered the JBL M2, but the complexity of the system let her at the end take the JBL 4367. It may not be as good (very relatively), but it's easy to use with one amp, and no thinkering needed for her needs. I did install Dirac room correction because her amp (NAD M33) had it build in. So one box, one power cable, and two speakers with speaker cable. Everything is well build in and cables are hidden so it fit the asthetics of her living room. A active system, even only a 2 way does not fit there.
https://www.soundimports.eu/en/hype...UNAbx3O4shlzJrBNjUb_0nJ_n5IBNqoY4HnHCRoQL4vNH
Roughly 650€ piece. Hypex FA series are widely used and well known among diy-folks. Decent dsp and NCore amps, can take analog or digital in. Jus two wires - power and signal.
But I understand desire for passive version, more classic and just plug and play. Bass response modding per room is lost.
Roughly 650€ piece. Hypex FA series are widely used and well known among diy-folks. Decent dsp and NCore amps, can take analog or digital in. Jus two wires - power and signal.
But I understand desire for passive version, more classic and just plug and play. Bass response modding per room is lost.
The hardware is currently a tiny raspberry pi board with 8 channel DAC hat with 4 wires going to the input of an AVR (to be 1 wire if things work well enough) and 4 pairs of wires going to the yet to be designed and built speakers (likely to be 2 pairs of wires). The setup controlled from a mobile, tablet, PC across the home network, a touch screen (if I buy one) or plugging in a keyboard, mouse, screen, TV.
How well the system works w.r.t. electrical interference and robustness has yet to be determined but being modular swapping components for better ones is straightforward as is using the system for other speakers. Excluding amplifiers the cost is comparable with that of a decent passive crossover.
How well the system works w.r.t. electrical interference and robustness has yet to be determined but being modular swapping components for better ones is straightforward as is using the system for other speakers. Excluding amplifiers the cost is comparable with that of a decent passive crossover.