Hi everyone,
Newbie here looking for some realistic & pragmatic advice on what to consider building for a long term project.
I currently have a full active Linn Akubarik "akitiv" system.
Linn — Akubarik Speakers
And a second Linn Majik 5.1 surround system.
I've been into hifi for ~20 years and the time has come for me to begin building my own gear. I have no wood working experience and very limited soldering experience however I'm keen and not phased by learning the skills needed to build myself (I think it will be fun!).
my long term aim is to build a high end active DSP 2.1 system to replace the Akubariks which significantly outperforms them.
I'm after neutral & transparent sounding system which will go in a dedicated listening room 5.38m (17'8") by 3.33m (10'11") & height 2.4m (7'10"), the room will have treatment complimented by DSP and 2-4 subs, all centered on a single listening position.
I'm considering buying a pre-amp with built in DSP e.g. Lyngdorf room perfect RoomPerfect | Lyngdorf Audio or possibly a DIRAC solution but would consider DSP modules in active speakers, whichever gives the best sonic performance.
I'm quite realistic that this will be a long term project & the learning curve is going to be steep.
I've looked into starting with a simple bookshelf kit to learn the ropes however I think the build looks quite trivial so think I should be more ambitious.
Budget is expected to be eye watering, anywhere up to £20k ($30k) or over, but will be spent in stages over 24-36 months.
Intention is to run the main speakers down to ~100Hz with the subs handling omni directional frequencies.
Ideally I'd like to be able to start with passive mains, running them via existing retail amps, and then make them active over time as I build the power amps. Introducing more & more subs as a second stage before the mains go active.
A lot of background I know, keen to get some realism so please be blunt with your feedback and suggestions.
lots of questions given the above but the 2 key ones right now are:-
1. What is the best speaker type choice for low bass extension speakers coupled with subs? - I'd consider any type including horn, open baffle, trad cabinet etc.
2. Is "jumping in at the deep end" and building a costly pair of main speakers a sensible & pragmatic approach for a newbie? - I'm not so keen on building something I'd never use just to learn the ropes.
I'd prefer an established & proven design and have been trawling the forum, usual designers & websites but don't seem to be able to find lots of choice for "high-end" narrow band mains.
I think part of the issue is that I don't yet understand/appreciate which mid and tweeter drivers are exceptional, and therefore which designs would be world class - Suggestions on designs greatly welcomed.
I'm still in research & fact finding mode at the moment, trying to make my mind up on the right approach.
Thanks in advance
Newbie here looking for some realistic & pragmatic advice on what to consider building for a long term project.
I currently have a full active Linn Akubarik "akitiv" system.
Linn — Akubarik Speakers
And a second Linn Majik 5.1 surround system.
I've been into hifi for ~20 years and the time has come for me to begin building my own gear. I have no wood working experience and very limited soldering experience however I'm keen and not phased by learning the skills needed to build myself (I think it will be fun!).
my long term aim is to build a high end active DSP 2.1 system to replace the Akubariks which significantly outperforms them.
I'm after neutral & transparent sounding system which will go in a dedicated listening room 5.38m (17'8") by 3.33m (10'11") & height 2.4m (7'10"), the room will have treatment complimented by DSP and 2-4 subs, all centered on a single listening position.
I'm considering buying a pre-amp with built in DSP e.g. Lyngdorf room perfect RoomPerfect | Lyngdorf Audio or possibly a DIRAC solution but would consider DSP modules in active speakers, whichever gives the best sonic performance.
I'm quite realistic that this will be a long term project & the learning curve is going to be steep.
I've looked into starting with a simple bookshelf kit to learn the ropes however I think the build looks quite trivial so think I should be more ambitious.
Budget is expected to be eye watering, anywhere up to £20k ($30k) or over, but will be spent in stages over 24-36 months.
Intention is to run the main speakers down to ~100Hz with the subs handling omni directional frequencies.
Ideally I'd like to be able to start with passive mains, running them via existing retail amps, and then make them active over time as I build the power amps. Introducing more & more subs as a second stage before the mains go active.
A lot of background I know, keen to get some realism so please be blunt with your feedback and suggestions.
lots of questions given the above but the 2 key ones right now are:-
1. What is the best speaker type choice for low bass extension speakers coupled with subs? - I'd consider any type including horn, open baffle, trad cabinet etc.
2. Is "jumping in at the deep end" and building a costly pair of main speakers a sensible & pragmatic approach for a newbie? - I'm not so keen on building something I'd never use just to learn the ropes.
I'd prefer an established & proven design and have been trawling the forum, usual designers & websites but don't seem to be able to find lots of choice for "high-end" narrow band mains.
I think part of the issue is that I don't yet understand/appreciate which mid and tweeter drivers are exceptional, and therefore which designs would be world class - Suggestions on designs greatly welcomed.
I'm still in research & fact finding mode at the moment, trying to make my mind up on the right approach.
Thanks in advance
I would start by adding subs (multiple) to your existing system. At this kind of price range you should be using proper loud speaker management systems, this unit is very highly regarded and has excellent specifications:
ASC48 - Linea Research
You should be using multiple subwoofers as it gives more even bass response throughout the room, I use 4 and am quite satisfied with the results. This article has some real world data on spatial variation reduction:
Data-Bass: Subwoofer Measurements
As your room is dedicated you should consider solutions like a double (or single with absorber) bass array:
Double bass array - Wikipedia
and Sofit mounting your main speakers:
Soffit-Mounting of Control Room Monitors
The subbass should be offloaded to the subwoofers, making the mains play low just increases their size and makes the bass drivers have higher excursion. If your system is correctly implemented an 80Hz crossover won't be localisable.
ASC48 - Linea Research
You should be using multiple subwoofers as it gives more even bass response throughout the room, I use 4 and am quite satisfied with the results. This article has some real world data on spatial variation reduction:
Data-Bass: Subwoofer Measurements
As your room is dedicated you should consider solutions like a double (or single with absorber) bass array:
Double bass array - Wikipedia
and Sofit mounting your main speakers:
Soffit-Mounting of Control Room Monitors
The subbass should be offloaded to the subwoofers, making the mains play low just increases their size and makes the bass drivers have higher excursion. If your system is correctly implemented an 80Hz crossover won't be localisable.
Last edited:
A bunch of money doomed to be "wasted". I advise you first try to improve the sound of Akubarik using dsp crossovers. Failing to do so would mean you are not talented enough to become a loudspeaker designer assuming there is something wrong with the Linn's.
If I ever wanted to invest such funds in this hobby, I'd for sure get a cnc router and get the load of unnecessary work off of my back.
If I ever wanted to invest such funds in this hobby, I'd for sure get a cnc router and get the load of unnecessary work off of my back.
For that type of money, i would contact a known speaker designer to design and build the speakers/system for me.
thanks for the response, adding subs is a good suggestion, thanks, that would make a lot of sense.
I've never come across this kind of system before, I've looked at the site but I still can't quite understand its purpose or what it's doing exactly. I can see it's providing DSP and managing the input/output, but how are the DSP functions configured? is it a mic based solution like DIRAC, or is it depending on manual configuration based on measurements & EQ setting from something like REW?
if it's manual I'm not sure I yet understand how you'd configure delay & phase - I've played with phase using REW and rephase before.
that's exactly the kind of thing I'm thinking of, though placement of the subs might differ to, I've read about a few different methods, not sure yet which will be best for me.
sofit mounting is a consideration, though I have a central fireplace and the speakers will need to be located in front of the alcoves at either side - I'm trying to decide if the alcoves are better for room treatment, not sure of the answer to be honest.
perfect!
That's a very worrying thing to say, I don't want to make a mistake, can you expand a bit on what the problem is that you see please.
I should have explained a bit more about the Linn setup, Linn have a proprietary DSP solution which is based on entering details of the room, speakers, listening position & sound absorption details of the materials.
it's works without a mic but does a pretty good job.
I've already used REW, rephase & EQ via Roon convolution filters to see if I can do better than the Linn solution and I can't improve on it.
a 3rd party solution might very well be able to, so coupled with added Subs might be a better 1st step.
CNC router might be a longer term idea, though I'm very fortunate to have access to a very well equipped wood shop.
I actually mentioned this as an option in the 1st draft of the post I wrote but decided to remove it, mainly due to lack of confidence in knowing if getting a custom design was the right thing to do, and partly because I've been influenced by reading the "so you want to design a speaker" thread which strongly recommends starting with proven designs 1st.
The problem I see is the presumption of needing exorbitant amount of funds to build a high quality dsp 2.1 system. The main issue is the implementation.
I should have been clearer in my 1st post sorry, having the funds available is not the same things as expecting to spend everything.
The point I was trying to get across is that the performance is the important factor, and if needed I can spend up to that amount, if I can get the performance I'd like cheaper then that's just fantastic!
This will be a dream system for me, I very much doubt I'll ever be able to afford to do anything better ever again (other then modest improvements over time).
I can't afford to throw money away on crazy things, and I can't afford to get things wrong in a major way.
and having made the dream system, I'll be able to sell on my old Linn system for a decent amount to get some of the money back, would expect this new system will cost around an extra £10k once that's done.
If I can't get a top class system for that I'm doing something wrong 🙁
Always keep in mind you/we are all creatures of limited capabilities prone to fall under influences of all kinds and it takes wisdom to discern what's right and what isn't because the world is not interested in your well being, rather in keeping you spending as much as possible and by doing so you may easily become a victim of it suffering the usual consequences. Start humbly, stay humble, learn from the experts ( speaker dave and gedlee aka Dr.Geddes) in this forum, practice performing loudspeaker measurements and learn to decipher what's room doing from what speaker is. Reading "Testing Loudspeakers" by Dr.Joseph D'Appolito may be a good investment, if you haven't already.
wise words.
i'm very conscious that there's a good and bad way to get to my dream end point and I'm very open to taking a less direct route, I'm just not sure what it is, any advice or suggestions greatly received.
starting humbly is very much what I had in mind, I'm unsure what's best as a 1st dip in the water though given the long term objective. Despite the budget I still don't want to waste money if I can avoid it and would much rather everything I do be a step towards the goal (ideally the step should be direct, but maybe an indirect step would be fine in some situations).
I've already been using REW to measure the room and I'll be introducing some room treatment once I've completed the build of a new partition wall, so already feel at home understanding the nuances between room and speaker - more practice needed but I'm already measuring.
I'll check out the book, thanks for the suggestion
You should have some drive units to begin any project. Active dsp filtering would get you going faster and simpler without having to worry about passive parts. You could use units from majik speakers.
Very important to understand the difference in the roughly two kinds of DSP you can by. The ones that has some kind of auto-tuning and the ones that has a manual dialing in function.
Simply refering to a DSP - says very little. Some DSP's can alter 3-4 PEQ's with a rough adjustment and only below maybe 250hz. Others can adjust a huge number of settings with 2-3 decimals in almost freely defined values of Q, gain and phase.
Also - beware of what can be corrected with a DSP and what cant. A DSP can manipulate the line-signal. It can never correct acoustics. So if the baffle has the wrong shape or size, the driver to big/small or the room plays tricks...... no amount of DSP power can fix this.
And..... measurements. Understand that they have to be good and done right. Or you are simply chasing your tail.
Dont go for the most expensive driver or the most crazy DSP/amplifier - just because of price, looks or data-sheets. Go for the gear that gets the job done - in your specific build. Which mean that you gotta make some choices about size, price, complexity, looks and skills too.
If you have a dream or burning wish because of something you red about a given product - please share it. Cause then it is much easier to figure out how to move on with the project 🙂
Simply refering to a DSP - says very little. Some DSP's can alter 3-4 PEQ's with a rough adjustment and only below maybe 250hz. Others can adjust a huge number of settings with 2-3 decimals in almost freely defined values of Q, gain and phase.
Also - beware of what can be corrected with a DSP and what cant. A DSP can manipulate the line-signal. It can never correct acoustics. So if the baffle has the wrong shape or size, the driver to big/small or the room plays tricks...... no amount of DSP power can fix this.
And..... measurements. Understand that they have to be good and done right. Or you are simply chasing your tail.
Dont go for the most expensive driver or the most crazy DSP/amplifier - just because of price, looks or data-sheets. Go for the gear that gets the job done - in your specific build. Which mean that you gotta make some choices about size, price, complexity, looks and skills too.
If you have a dream or burning wish because of something you red about a given product - please share it. Cause then it is much easier to figure out how to move on with the project 🙂
Its a manual DSP like the mini-DSP. It also can act as a DAC when driven by AES3 (pro level SPDIF) or Dante (realtime pro audio over IP standard). Its configured over the network using a computer (but stores it settings). Using these you can process all the signals for the power amps of your system for crossovers, eq room EQ etc. it does not support automatic EQ you have to design the filters using E.G REW or FIR designer. This type of DSP has a fixed process flow and you can look at the available functions on page 34 of the manual:
https://linea-research.co.uk/wp-content/uploads/LR Download Assets/User Guides/ASC User Guide.pdf
I don't personally use this DSP but its well regarded, the same company also does amps with similar DSP capabilities integrated. You might also want to consider Powersoft X4/X8 if you want a single box solution to amp+dsp.
I use a Symetrix 8x8 DSP + DIGIO which is an older free form DSP (as in the processing is not fixed. That could also be an option and is the most flexible but could be a bit of a steep learning curve. QSC also do free form DSPs:
Symetrix Radius - Open Architecture Dante Digital Signal Processors
Core 110f - Q-SYS Cores - Products, Peripherals & Accessories - Q-SYS Ecosystem - Products - Systems - QSC
I suggested the ASC48 as its analog output specifications are better than QSC/Symetrix and its easier to use.
Its worth downloading the QSC and Symetrix software and playing about offline to see what these DSPs are capable of. With enough units you can process the input/output signals for a whole stadium from microphones to the drive units.
I have never used auto-eq systems like Dirac to me they seem very limited, better than nothing but I want to choose my own filters!
*I would suggest building a new set of speakers rather than messing with the Linns as you will find that even exceptional drive units are not that expensive compared the finished Linn speakers systems !
Last edited:
Thanks for the advice, I'd rather buy some additional drivers than cannibalize the Linn speakers, I'm going to need to keep using those while I work on this project.
thanks for the advice, I generally spend quite a lot of time researching things, and ideally I'd like to listen as well, before I buy anything.
I don't really know exactly what I want to do yet, it's just a desire to try to build a world class system. I need to read and learn more about what's possible so I can start to piece together how I'm going to do it. I'm a long way off putting together a shopping list.
research, research, research
great thanks for the info, I can see having that much control would be very useful, something for me to think about.
Other than knowing I want DSP, I'm still yet to decide what's best for me, need to look into it in more detail to make my mind up, I think they only constraint is the ability to manage multiple subs.
I also want to run seperate DAC & streamer units so ideally would like a dedicated DSP
Well, were to start. I think there is already good advices posted.
On the onset, it seems rather easy to build a great loudspeaker. However, this is not the case.
I think you would be the first person ever, to achieve a perfect speaker as a first build, if starting from scratch. I think you should consider at least 5 complete builds from scratch, before getting close to at perfect speaker. Maybe only to find, that you did something wrong, needing to start all over. It is said that mastering a discpline takes around 10.000 hours of practice. Not a point that it takes that long to make a great sounding speaker, but it takes a long time fully grasping all details, not the least combining them all. Look for instance Troels Gravesens page. I bet it takes him way more than 100 hours completing a build, which he also states somewhere.
Also never neglect what you learn in practice. Give speakers a listen. People prefer different things. You can read about a product for a month, then when you actually listen to it, all you thought was true is wrong. This might change one more time, when you put the speaker into your room.
To answer your question 2) No it is not a pragmatic way.
This is by no means to scare you away from DIY speaker building, but expect several years to really nail it. You might find that all the money spent, but more importantly the time spent, would be no better than buying some decent used speaker, correcting them with DSP and calling it a day.
On the onset, it seems rather easy to build a great loudspeaker. However, this is not the case.
I think you would be the first person ever, to achieve a perfect speaker as a first build, if starting from scratch. I think you should consider at least 5 complete builds from scratch, before getting close to at perfect speaker. Maybe only to find, that you did something wrong, needing to start all over. It is said that mastering a discpline takes around 10.000 hours of practice. Not a point that it takes that long to make a great sounding speaker, but it takes a long time fully grasping all details, not the least combining them all. Look for instance Troels Gravesens page. I bet it takes him way more than 100 hours completing a build, which he also states somewhere.
Also never neglect what you learn in practice. Give speakers a listen. People prefer different things. You can read about a product for a month, then when you actually listen to it, all you thought was true is wrong. This might change one more time, when you put the speaker into your room.
To answer your question 2) No it is not a pragmatic way.
This is by no means to scare you away from DIY speaker building, but expect several years to really nail it. You might find that all the money spent, but more importantly the time spent, would be no better than buying some decent used speaker, correcting them with DSP and calling it a day.
Last edited:
I recommend reading the following:
"Interview with Speaker Design Guru David Smith" at tnt-audio com
"Interview HOBBY HIFI mit D'Appolito" in German at forum visaton de (first post by Laie)
A lot can be learned from these two humble and knowledgeable experts.
edit: I disagree that it is difficult to build a great loudspeaker. The problem is people are presenting things in that regard as very complex for the obvious reasons.
"Interview with Speaker Design Guru David Smith" at tnt-audio com
"Interview HOBBY HIFI mit D'Appolito" in German at forum visaton de (first post by Laie)
A lot can be learned from these two humble and knowledgeable experts.
edit: I disagree that it is difficult to build a great loudspeaker. The problem is people are presenting things in that regard as very complex for the obvious reasons.
Last edited:
Epic links!
I had no intention stating that things is more complicated than it is. It is not difficult for anyone to buy a decent DSP, setting up a simple crossover and getting something decent going within one hour. But to get something truly great sounding, requires experience, which can only happen over time.
To OP a great starting point for Multible-sub would be to buy a MiniDSP 2x4HD (not the old cheaper one, it is bad), and either buy some cheaper subwoofer or build some, then start experimenting. This experient might end with conclusion, that even the most expensive fullrange speakers will benefit from subwoofer(s). Multiple Subwoofers: Optimize Them With Multi-Sub Optimizer Software
I had no intention stating that things is more complicated than it is. It is not difficult for anyone to buy a decent DSP, setting up a simple crossover and getting something decent going within one hour. But to get something truly great sounding, requires experience, which can only happen over time.
To OP a great starting point for Multible-sub would be to buy a MiniDSP 2x4HD (not the old cheaper one, it is bad), and either buy some cheaper subwoofer or build some, then start experimenting. This experient might end with conclusion, that even the most expensive fullrange speakers will benefit from subwoofer(s). Multiple Subwoofers: Optimize Them With Multi-Sub Optimizer Software
Last edited:
I'll open by declaring myself a 'hobbyist' without the experience of many here, but I'll make some broader points. I think many would hope to be able to make the 'perfect system' with similar resources. I will follow your journey with interest if you choose to go ahead.
Fundamentally DIY struggles to compete with cheaper commercial gear (economies of scale) and also more expensive equipment (access to subspecialist knowdge, advanced production techniques and being able to 'spec' drivers from manufacturers), but in the middle of those 2 it can do very well.
At the more expensive end of hifi it becomes much less about 'better' and much, much more about preference. Sadly the only way you will work out if you prefer paper cones to metal or ceramic, OB to line arrays, tubes to class A, is a wealth of experience. If you know exactly what you want it makes sense to expend considerable resources. But if you're not sure what flavour is your favourite you may spend time effort and money only to discover you prefer something else.
If you still want to persue high end DIY (and I think you should) I would definitely suggest picking a design from a known successful high end designer rather than creating something from scratch. Toels graveson and Tony gee would be common places to start but you might struggle to spend more than £5k. Not many active designs about. If none of their designs is to your liking I think they both do commissions for a premium. In UK wilmslow have some designs with high end drivers but not heard much about them, maybe go demo. If you build a high end design, it will still be a great speaker even if it turns out you would have preferred a slightly different flavour.
Finally, spend some money on assessing and managing room acoustics and room plan. Both more important than credited for.
I think you'll see these themes recur in responses, in between advice relating to design which I'll resist chipping in with and leave to those more knowledgable than myself. Good luck.
Fundamentally DIY struggles to compete with cheaper commercial gear (economies of scale) and also more expensive equipment (access to subspecialist knowdge, advanced production techniques and being able to 'spec' drivers from manufacturers), but in the middle of those 2 it can do very well.
At the more expensive end of hifi it becomes much less about 'better' and much, much more about preference. Sadly the only way you will work out if you prefer paper cones to metal or ceramic, OB to line arrays, tubes to class A, is a wealth of experience. If you know exactly what you want it makes sense to expend considerable resources. But if you're not sure what flavour is your favourite you may spend time effort and money only to discover you prefer something else.
If you still want to persue high end DIY (and I think you should) I would definitely suggest picking a design from a known successful high end designer rather than creating something from scratch. Toels graveson and Tony gee would be common places to start but you might struggle to spend more than £5k. Not many active designs about. If none of their designs is to your liking I think they both do commissions for a premium. In UK wilmslow have some designs with high end drivers but not heard much about them, maybe go demo. If you build a high end design, it will still be a great speaker even if it turns out you would have preferred a slightly different flavour.
Finally, spend some money on assessing and managing room acoustics and room plan. Both more important than credited for.
I think you'll see these themes recur in responses, in between advice relating to design which I'll resist chipping in with and leave to those more knowledgable than myself. Good luck.
I'm under no illusion that there's going to be an enormous learning curve and that this might take a lot of time and effort to get to where I want to.
my (mis) perception though is that provided I don't do stupid things like damage a cone, lots of the DIY build mistake should be fixable e.g. bad soldering can be redone, cabinets can be remade, drivers rehoused, crossovers swapped out etc.
I'd also expect to get better & faster with each build, I can see myself building 4 subs, the 1st will take a long time, but by the 4th I should be a lot faster & be at a higher standard - I have no issue then with going back to the 1st and remaking it given what I've learnt.
The main issue then is time and effort, which is something I'm willing to pour into the project, after all this is the fun side for me, win-win hopefully!
Thanks have read the 1st interview, it matches my expectations, and makes me want to start building something! if anything it's just adding fuel to the fire 🙂
in my OP I asked about which speaker design would pair best with a multi-sub setup, I'm still not sure what the answer is but reading this interview I'm again thinking line arrays and horns sound really interesting.
The other thing that reading this interview brings out again is my main reason for DIY, it's a very strong driver for me:- I want total control, big attention to detail, use the best parts, pick the perfect materials.
That might mean that I lean more towards design myself at some stage, but I'm a long way off dipping my toe in that water. for now I'd be content using existing designs and where I have the latitude to choose materials or components I'll use that to pick what I prefer.
I'm pretty sold on the idea of sub integration and plan to try and number of different options as I add more and more subs.
Kipman725 mentioned double bass array, I'd also be interested in Welti and Geddes approaches too, as well as the more traditional single & double sub.
think trying these different approaches work really well with DIY, I can try a different approach each time I produce a new sub.
Thanks for the software link, I've already been looking at it.
I've got a hard linkage between room treatment and subs for my listening room - goal is to try to reduce the room treatment by using subs instead.
something that I'll be working on at the same time as the DIY build.
I take your point on preference, my goal still is to get out and listen to lots of gear before I decide on what I'm going to build for my main speakers and amps.
for Subs I'm not sure if this is as big a consideration, sub design seems to me to converge on a much smaller number of optimal designs, so I think it might be more viable to pick the most appropriate for me and build 1.
As I've already said, I'm still in fact finding mode, trying to figure out what the best approach is for me before I kick things off in anger.
I am a little concerned about the high end DIY versus buying from a manufacturer comment though, are you effectively saying the off the shelve drivers used for DIY can't compete on sound quality terms with unavailable custom designed drivers used in high end retail gear? That could be a show stopper for me 🙁
it's important that I can DIY build something for my available budget which is better than I could by off the shelf - I'd even think that buying 2nd hand falls within that statement....
Last edited:
They do want you to think like that. That doesn't make it true.
Why not get a PC with multichannel card, run a free DSP program like DRC-FIR with REW and a media player including convolution and experiment with what you've got, learning the ropes.
Next, take a good look at your room. decide what treatment options are available and determine what kind of speakers could work in that space to get great results.
Don't go running blindly after High End, learn a bit more about speakers and how it all works first. Not all that seems really great is that great anyway.
If we're allowed again, visit different setups if you can. Get a feel of what it is you want and/or like sound wise. Go for function, not glamour.
Before I forget to mention it, DSP isn't a fix all cure. It will do exactly what you tell it to do. Knowing what you want it to do is the bigger secret to figure out.
It can really help make a system wonderful, but it all starts with a sane plan based on acoustic principles. It can't make a wart shine.
On this forum we have some guests that know a thing or two, members like gedlee, Tom Danley, speaker dave (which happens to be the same David Smith that was linked earlier), among others, all post here and will bring you valuable lessons if you go look for it.
Last edited: