I think question like "why would anyone build this project?" would need to be answered first and foremost to get to get a solid foundation for project and a finished design eventually. Who is the anyone in question, who is gonna build it and why? Otherwise there is a risk for million things and desires with contradiction and the end result is a mishmash no one wants to buildBasic stuff for any engineering project.
I suggested the CTA 2034 standard earlier just because that is some kind of guideline that attempts to somehow generalize typical listening situation and gives some measures how to judge the quality in simulation phase. In addition at least VituixCAD has the CTA 2034 standard stuff baked in so it is rather easy and inexpensive to iterate a design with. The software even provides a preference ranking number by Sean Olive to judge performance. Very easy to evaluate the performance development of any project before going to listening tests.
It could certainly form part of what a group design project might use to get member's outlook aligned around detailed objective criteria for the required performance of a speaker. What I was referring to early as the specification of what is to be designed and built. This is how formally trained and experienced engineers (less so technicians and "self-taught" engineers) would tend to go about a group project like this because it is the most efficient and effective way for people dialled into an objective view to get good results.
It is of course not the only way to go about things and for those pursuing a hobby for fun with a weak grasp of the relevant engineering (or interest in many cases) it may not be an appropriate approach or even a viable one. An obvious alternative is to quickly rough out a design based on what is around and ones feelings about what might work. Build, measure, another quick think about what might work, build, measure, etc... So long as the design moves forward it can be an enjoyable process. If it doesn't enthusiasm is likely to wane and it is time to move on to something else.
All the forum based group DIY speaker projects I have observed and/or briefly posted in have largely followed the latter approach. Most don't get past the first prototype with some not even reaching that stage. It is not necessarily the wrong way of going about things but it does tend to pull in people that want to go about things in this way and push out those that don't. It would be interesting to see a group project go about things differently but whether it will work any better who knows?
For this project things don't seem to quite fit at the moment when it comes to respecting the idea of a reference speaker (or at least my understanding of it!) and respecting the OP's wishes for main speakers for a room using an 8" woofer, 4" midrange, below standard range driver cost, world driver availability plus a few other things that have yet to be fleshed out in a spec.
In my case the reference part is important because it enables a considered objective approach which is part of my hobby interest. In which case a small 20 litre(ish) speaker for relatively close listening with sub support could match most of the OPs requirements although probably not all such as the budget. A spec needs to be written down so that everyone that opts to become actively involved knows and is happy with what is to be designed and built.
Well the idea grow on me from reading the first posts of this tread.
I was long into commercial speakers. But look around and most are 2-way designs. It is hard to experiences 3-ways, TL designs serial XO, time alignment, low vs higher order XO, etc.
If one can build those for moderate cost I suppose it is fun to make and learn first hand the differences in sound they will have. It's hard to have a change to explore all. So far I have heard an electrostatic only once in my life. OB and FR never...so I can not say if it is my thing or not....
This project can form a base to understand what you really like and guide you through your DIY journey.
If I would look purely to myself I would like to re-use the drivers I have (OW1 and 15W8530) and build from there...but I want to look beyond that and make it more accessible for beginners.
Alright I trust you. This clarifies some misunderstanding I had. Now I understand your stance on this. The main purpose of the project is to have a design process that involve anyone interested into a design process and keep the process interesting. Secondary goal is to try to achieve nice end result, a 3 way speaker consisting of typical ~1-4-8" drivers, which would be accessible to many in the future as well. Given the constrains this would not lead to best possible 3-way but hopefully well balanced 3-way given the constraints.
In this context this would be a reference project even though the end result wasn't the ultimate 3-way concept. A reference for people who would like to explore and design a 3-way system without too much audiophool foo involved, how to approach things and what might matter and what might not. And of course to help moondog as thread starter finish his project, providing the context and constraints.
Alright, I'd say it is a good intention and should be fun and I hope it would progress further than other threads you have observed in the past. I see you have grasp how to get the venture started and maintained
What others think? hifijim, allenB, other long time members and new comers, is your brain muscle tuned and typing fingers oiled ready to support the cause?
ps. I do hope the best but the interplay of things in a loudspeaker system is very complex and typing ideas down into sentences is very hard and will make long and winding posts which could make things very hard to understand, even though the individual concepts are rather simple by them selves. I've found that the stuff I read aeons ago have started to make sense only after sufficient understanding on most things involved is developed enough. Then it is rather easy to connect the dots and develop ideas within the head, but not sure I'd be able to write all that down into a text. Well, communication excercise never hurts.
Last edited:
Well, most of the stuff I hooked on was already discussed earlier in the thread. Should have read it first. A fool makes the same mistake again and again, not pay attention and then end up preaching the choir A little stir up doesn't hurt or does it? Sorry if I was on the rude side.
Will follow and comment on aspects when it looks like I've got something to bring to the table. Carry on!
A little stir up doesn't hurt or does it? Sorry if I was on the rude side. Will follow and comment on aspects when it looks like I've got something to bring to the table. Carry on!
Some random thoughts:
If we decide the tweeter needs a waveguide (or horn), that cost has to be included in the estimate. If this is the path we go, I recommend we assume the builder does not have access to CNC machining or to a 3-D printer, i.e. the waveguide has to be one which can be purchased.
As Tmuikku and Andy have discussed, this project is not about building the ultimate 3 way speaker... how could it be, with a driver budget of $600 and a design assumption of a basic workshop? .... But we should be able to design something which competes with commercially available speakers in the $4000 - $6000 retail price/pair range. The title to this thread "3 Way Reference Project" might be obsolete or misleading at this point.
Maybe we need to start a new thread? If we do, I would like to capture the discussion that has taken place since post # 471.
Regarding requirements, I am happy to play the role of "Requirements Gathering Person". I jumped in because I wanted to organize the ideas being tossed about. ... However, if anyone else would like to take over, please do, you will not hurt my feelings.
j.
If we decide the tweeter needs a waveguide (or horn), that cost has to be included in the estimate. If this is the path we go, I recommend we assume the builder does not have access to CNC machining or to a 3-D printer, i.e. the waveguide has to be one which can be purchased.
As Tmuikku and Andy have discussed, this project is not about building the ultimate 3 way speaker... how could it be, with a driver budget of $600 and a design assumption of a basic workshop? .... But we should be able to design something which competes with commercially available speakers in the $4000 - $6000 retail price/pair range. The title to this thread "3 Way Reference Project" might be obsolete or misleading at this point.
Maybe we need to start a new thread? If we do, I would like to capture the discussion that has taken place since post # 471.
Regarding requirements, I am happy to play the role of "Requirements Gathering Person". I jumped in because I wanted to organize the ideas being tossed about. ... However, if anyone else would like to take over, please do, you will not hurt my feelings.
j.
How are you going to decide this type of thing without a target performance specification for the speaker? If a group member fancies one and others don't object, let the OP to decide,...? Not intending to have a go but to illustrate that if you sort out a spec first you will have something to assess against and this kind of decision becomes more quantitative and straightforward.
One can purchase well designed and engineered 3 way active studio monitors for that sort of price. There are also a few well designed commercial passive 3 ways in the $2-3k range from the larger companies with an engineering outlook to some extent like KEF, some Harman brands, and no doubt a few others (not that familiar with passive home speakers). There are certainly lots of overpriced passive home audio speakers with modest technical performance and some well designed ones with greatly inflated prices but should we be comparing DIY speakers solely against them?
I would agree it needs either adopting (with a clear definition) or dropping.
I would suggest starting a new thread after you have gathered and agreed on the requirements with at least one or two people expressing active interest. This will enable the OP to define the project which should help reduce the number of posts suggesting random things outside the project scope and for the thread to be more about designing the speaker. Chatting around a subject obviously has it's place on a forum like this but vast amounts of it among a few posts contributing to a design is not ideal.
I expressed what I would like as outcome from this project.
Don't know how I can translate it more into needed requirements.
When I asked if the tweeter was going to need a wave guide, that is my short-hand way of saying: How are we going to write the requirement for directivity / dispersion. Because that requirement will determine whether or not we need a waveguide on the tweeter.
I expect there will be several competing designs that get proposed, and some may have waveguides, some may not.
But even if we wanted to establish a directivity / dispersion requirement, what would it be? Dr. Floyd Toole’s book is 550 pages, and he never defines a quantitative spec or limit on directivity, or sound power DI, or listening window response smoothness… He just provides qualitative guidance on what makes a good speaker good, and a bad speaker bad.
I feel it is really important to create our requirements with an awareness of what is possible. It is so easy to create a list of conflicting requirements which become impossible to execute. The spec writers need to have a vision of what a spec-compliant product would look like… it is hopefully not the best possible product, and certainly not the only possible product, but there should be at least one possible way to meet the spec in the minds of the spec writers.
My role at this time is to gather requirements, and in that role I am hesitant to suggest requirements until all the players have had an opportunity to voice an opinion. If this were MY project, I know what kind of speaker I would build, I have a thread dedicated to that. But this is not MY project, it is OUR project. So if anyone has a suggestion for a directivity requirement, or any other requirement, I am ready to capture it and add it to the list for future discussion.
I was simply making the point that a well-done DIY speaker with $600 in drivers will not be an ultimate performance 3-way, but it will usually compare well against commercially available passive speakers in the $4k-$6k range. Some of those commercial speakers will be better than our DIY, some will not, but our DIY will be comparable. I was not making any kind of value judgment on $4k-$6k speakers… whether they tend to be good sounding, or overpriced, or inferior, or whatever… I am managing expectations, that is all.
j.
Last edited:
There's a thread over at ASR where they propose an active speaker requirement. It's active, but from a design spec, its a good template. I've copied it over here so as an idea for "requirements capture", if we were to start a new thread.
Requirements for DIYAUDIO Upgradeable speaker concept.
1. The DIYAUDIO Upgradeable speaker should be passive and simple to build.
2. It will be a 2 way speaker, upgradeable to 3 way, or built as a 3 way from the outset.
3. Use of commercially available, knockdown or pre-made, is preferred, if possible.
4. The design should be reusable and extensible to allow for design variants.
There are 2 main extensible aspects that are expected.
Those would would like to start with a smaller speaker for smaller room can build the 2 way.
This speaker can be later converted to a 3 way, using the same mid-woofer, tweeter, and largely the same crossover components.
The second stems from bass alignment eg. bass reflex, convertible to sealed.
5. As such, the design is not required to operate with low power amplification. (eg. 10W or less)
6. Listening distance is targeted to be 2.5-3m, in a listening room of 75m^3.
7. The speaker should be able to have good overall performance whether it is located closer to a wall or out into the room.
8. The cost target for the drivers would be approximately $600 US retail per pair, and availability should be worldwide.
9 . The cabinet dimensions will be 1200x250x400mm at a maximum.
10. The design tools should be free but flexible enough to allow external data import/export. eg. VituixCAD and REW.
11. The design should be openly shared and FOR NON-COMMERCIAL USE ONLY.
Requirements for DIYAUDIO Upgradeable speaker concept.
1. The DIYAUDIO Upgradeable speaker should be passive and simple to build.
2. It will be a 2 way speaker, upgradeable to 3 way, or built as a 3 way from the outset.
3. Use of commercially available, knockdown or pre-made, is preferred, if possible.
4. The design should be reusable and extensible to allow for design variants.
There are 2 main extensible aspects that are expected.
Those would would like to start with a smaller speaker for smaller room can build the 2 way.
This speaker can be later converted to a 3 way, using the same mid-woofer, tweeter, and largely the same crossover components.
The second stems from bass alignment eg. bass reflex, convertible to sealed.
5. As such, the design is not required to operate with low power amplification. (eg. 10W or less)
6. Listening distance is targeted to be 2.5-3m, in a listening room of 75m^3.
7. The speaker should be able to have good overall performance whether it is located closer to a wall or out into the room.
8. The cost target for the drivers would be approximately $600 US retail per pair, and availability should be worldwide.
9 . The cabinet dimensions will be 1200x250x400mm at a maximum.
10. The design tools should be free but flexible enough to allow external data import/export. eg. VituixCAD and REW.
11. The design should be openly shared and FOR NON-COMMERCIAL USE ONLY.
Last edited:
I though that this was pretty clear in the first few posts.
Relatively cheap. Reasonably simple to build and with reasonable sound quality to be used as a "Yardstick" against which subsequent builds could be compared.
It shouldn't need killer bass otherwise we'd just be making more clones of the "Tarkus' so no need of -3dB at 30Hz [ or +6dB at 30]
Ballpark, off the top of my head -10dB @40 would work.
I stand by my remark that a smooth response tweeter with a low FS is needed as those are usually better with simpler cross-overs, perhaps even first order a single cap in line with or without resisters to drop the level.
High efficiency not needed and no real need for 110dB, 95dB is loud enough for most people most of the time in smallish rooms.
Bookshelf or tower or stand mounted as builders seem fit.
I had a pair of IMF Super Compacts way back, volume was never a problem, the drawback was simply lack of deep bass; we now have subwoofers that give us deep impact bass so why build bigger than needed?
I think this may be one of the things stopping this from moving forward. I didn't think it was very clear at all. There is a lot of subjectivism... An 8" woofer being "more than enough" to one person may be "not even worth bothering about for a 3 way" for another
Not a criticism but I think we need to define further.
Relatively cheap. Let's define that. Are we talking about driver's I've got leftover from another project or ones I have on my shelf? Or spending money on new drivers? And how much to spend (?How much US/AUD$ for drivers).
Bass extension for music- ?40Hz ?50Hz ?60Hz
Low Fs tweeter; NOT a problem- plenty on the market in the past decade all under US$50
Not high sensitivity. Let's define that. Shall we say Medium sensitivity at 85dB/2.83V. A pair should hit peaks of 100dB in a room
Small room- let's define small. 2x3x2.4m or 4x6x2.4m?
Cabinet volume- bookshelf or standmount or tower- well there's a big difference between 7L, 15L, 28L or 40L, and it affects sensitivity or bass extension (Hoffman's Iron Law).. so what size do we want? Is there an upper or lower limit?
Last edited:
So far, these are the tentative list of proposed requirements on the table for discussion. There has been very little discussion so far unfortunately, but hopefully that changes. I am most interested in hearing from Moondog55, kapelli, and MrHifitunes.
1) A TMW 3 way tower with an 8" woofer. This is reasonable, but with such a system we need to have realistic expectations about max SPL and sensitivity. Twin woofers would result in greater SPL and higher sensitivity.
Comment: Is there an upper size limit? Does it have to be a TMW configuration? Is a bookshelf/standmount acceptable?
2) A budget of $600. Is this for the drivers, or for everything? I would think that $600 for drivers is reasonable, but a total budget for everything of $600 starts to really limit our choices of drivers and crossovers. It would mean a driver budget of about $450, and that is rather stingy for 6 drivers.
Comment: Since there has been no objection, I am assuming that $600 is for drivers.
3) A choice of drivers which are readily available in most nations. From what I have gathered, Dayton is somewhat over-priced in Europe, and Seas is hard to get (expensive) in Australia. Are there any other driver OEMs that should be excluded?
Comment: Any other driver OEMs to be excluded? This is not something I can check from my desk in the USA...
4) SPL capability of 85 dB at 1 m, with a 20 dB headroom for peaks, which equates to 105 dB for peaks. Is this enough?
5) Minimum impedance of 3.5 Ohm
6) Basic workshop capabilities, assume power saw, router, drill, clamps.
7) Baffle edge diffraction control will be limited to bevels cut on table saw.
8) Bass extension minimum of F3 = 40 Hz. I would prefer a requirement for F10 rather than F3, but this is what has been proposed so far.
1) A TMW 3 way tower with an 8" woofer. This is reasonable, but with such a system we need to have realistic expectations about max SPL and sensitivity. Twin woofers would result in greater SPL and higher sensitivity.
Comment: Is there an upper size limit? Does it have to be a TMW configuration? Is a bookshelf/standmount acceptable?
2) A budget of $600. Is this for the drivers, or for everything? I would think that $600 for drivers is reasonable, but a total budget for everything of $600 starts to really limit our choices of drivers and crossovers. It would mean a driver budget of about $450, and that is rather stingy for 6 drivers.
Comment: Since there has been no objection, I am assuming that $600 is for drivers.
3) A choice of drivers which are readily available in most nations. From what I have gathered, Dayton is somewhat over-priced in Europe, and Seas is hard to get (expensive) in Australia. Are there any other driver OEMs that should be excluded?
Comment: Any other driver OEMs to be excluded? This is not something I can check from my desk in the USA...
4) SPL capability of 85 dB at 1 m, with a 20 dB headroom for peaks, which equates to 105 dB for peaks. Is this enough?
5) Minimum impedance of 3.5 Ohm
6) Basic workshop capabilities, assume power saw, router, drill, clamps.
7) Baffle edge diffraction control will be limited to bevels cut on table saw.
8) Bass extension minimum of F3 = 40 Hz. I would prefer a requirement for F10 rather than F3, but this is what has been proposed so far.
Thanks tktran303. So far I have not given much thought to the groundrules on how we would go through the design process. My one thought is that I am hoping we all can agree to use VituixCad2, but beyond that I have not thought about it. Your list touches on many important aspects of the design process.
I have been following the ASR thread from the beginning. It is very interesting, especially since Kimmo got involved.
I will try to merge your proposed requirements into the list.
j.
you can do an iteration on XRK's FAST by adding a sub under. His drivers meet your budget target. I also successfully used Visaton AL200 with SS 10F but that is a more expensive combo that would not leave $s for an additional driver. Btw I am still saving those drivers to try an LXMini-like config with an 8inch midbass (already tested the concept). Now if you wanted to engineer something like that, I believe that would be a value add to this forum; else there are already well engineered 3-ways with 8+4 combo (notably also in OB version).
10F/8424 & RS225-8 FAST / WAW Ref Monitor
10F/8424 & RS225-8 FAST / WAW Ref Monitor
Last edited:
An F-10 of 40Hz would be easier to reach wouldn't it?
Howabout an option for size; bookshelf for sealed and tower for ported?
So a woofer that could be used in either.
I'm just throwing words into this; as constrained personal circumstances mean no new drivers for me for a long while.
I blew both my budget and my WAF on on my last sub woofers purchase.
I have a favourite budget 4" driver but it's not available outside Australia/NZ as far as I know.
Jaycar for the Aussies here
Howabout an option for size; bookshelf for sealed and tower for ported?
So a woofer that could be used in either.
I'm just throwing words into this; as constrained personal circumstances mean no new drivers for me for a long while.
I blew both my budget and my WAF on on my last sub woofers purchase.
I have a favourite budget 4" driver but it's not available outside Australia/NZ as far as I know.
Jaycar for the Aussies here
Thanks tktran, this is an example of a neat list with some very desirable attributes.. but it has questionable utility as though it was not written in consultation.
So 3 way kind of becomes a moot point if you use the same midrange as the two way version.
The reason for making a 3 way if it doesn't need to be a 3 way (never mind the subwoofer region). Manufacturers often offer models/upgrades which are 2 and 3 way. It only really makes sense when the bigger 3 way models have a smaller mid and a larger woofer. The two ways have a small woofer that isn't small enough for the upper mids and isn't large enough for the bass.
So 3 way kind of becomes a moot point if you use the same midrange as the two way version.
Another example. The ability to play close to the wall is more than just setting the weight of the baffle compensation... the control is in the size. So number 7 suggests a design feature, then 9 puts a barricade right in the middle of it.
Would version of this be suitable?: SBA-861-PFCR
I'd suggest a smaller mid, lose Troels' 'Diffraction shelf', and put a sensible port on it....
shouldn't cost too much, 'specially if you don't fall for the audiophool xover parts
I'd suggest a smaller mid, lose Troels' 'Diffraction shelf', and put a sensible port on it....
shouldn't cost too much, 'specially if you don't fall for the audiophool xover parts
AllenB,
It's up the group what they want, in terms of a democratic affair.
The dimensions was an example of how the group may consider a size requirement or constraint.
I didn't get into the technical specifics of how a 2 way can be converted into a 3 way.
But it can be done, anytime a driver smaller than the another driver. eg. 2 way with 5.25" midwoofer might be good to about 60Hz, but add in a 8" and it can reach another octave, and spare that little mid-woofer of extra excursion requirements if you crossover at 200Hz-300Hz.
If the group wants to reuse components, then one way would be to build a 2 way, then upgrade to a 3 way later, by building the woofer cabinet below the 2 way...
eg. Roman J Bednarek's
Astorius for the
Asterion
Another example of a commercial design is the Kii Three vs Kii Three + BXT.
As for adjustable baffle step compensation (and corresponding tweeter level) so that a speaker may play nice far away, or close to a wall, again, it can be done.
I didn't want to get into the weeds about the technical aspects of how it can be achieved...
It's up the group what they want, in terms of a democratic affair.
The dimensions was an example of how the group may consider a size requirement or constraint.
I didn't get into the technical specifics of how a 2 way can be converted into a 3 way.
But it can be done, anytime a driver smaller than the another driver. eg. 2 way with 5.25" midwoofer might be good to about 60Hz, but add in a 8" and it can reach another octave, and spare that little mid-woofer of extra excursion requirements if you crossover at 200Hz-300Hz.
If the group wants to reuse components, then one way would be to build a 2 way, then upgrade to a 3 way later, by building the woofer cabinet below the 2 way...
eg. Roman J Bednarek's
Astorius for the
Asterion
Another example of a commercial design is the Kii Three vs Kii Three + BXT.
As for adjustable baffle step compensation (and corresponding tweeter level) so that a speaker may play nice far away, or close to a wall, again, it can be done.
I didn't want to get into the weeds about the technical aspects of how it can be achieved...
Last edited: