Heya .. new to the diy speaker building. I am very interested in Adire's technology and have been impressed with their car audio products. Now I am interested in building a high quality home speaker system. I want full-range reference speakers. Hopefully I can figure out how to implement a digital 4-way stereo crossover system on my PC, because I don't want to mess with passive crossovers (is this a mistake?)
Anyway, what do you think of this configuration? Any comments? Driver details here: http://www.adireaudio.com/
2 Extremis 6 drivers and a Vifa XT ring radiator tweeter in a sealed MTM configuration at the top of the tower.
1 Koda 8 Midbass driver below.. sealed.
1 Shiva 12" sub driver below that .. sealed.
I hope to crossover like this:
Shiva .. up to 60hz
Koda 8 .. 60-200hz
Extremis 6 Drivers .. 200-2khz
Vifa Tweeter .. 2khz+
I am thinking about using a Rotal 40x8 @ 8 ohm amplifier which puts out 65x4 @ 4ohms .. which is what the impedence of these drivers is. I will probably buy a separate plate amp for the Shiva.
Please comment! This all sounds awesome in my head, but I'm a newb.
Anyway, what do you think of this configuration? Any comments? Driver details here: http://www.adireaudio.com/
2 Extremis 6 drivers and a Vifa XT ring radiator tweeter in a sealed MTM configuration at the top of the tower.
1 Koda 8 Midbass driver below.. sealed.
1 Shiva 12" sub driver below that .. sealed.
I hope to crossover like this:
Shiva .. up to 60hz
Koda 8 .. 60-200hz
Extremis 6 Drivers .. 200-2khz
Vifa Tweeter .. 2khz+
I am thinking about using a Rotal 40x8 @ 8 ohm amplifier which puts out 65x4 @ 4ohms .. which is what the impedence of these drivers is. I will probably buy a separate plate amp for the Shiva.
Please comment! This all sounds awesome in my head, but I'm a newb.
If you're using the Extremis and the Shiva, I don't think you need to have the Koda8, since the Extremis should be able to play down to a reasonable level for the Shiva. Also the XT tweeters that I know of will not play to 2K as well as many other tweeters would.
I'm also confused about the electronics end. You have an 8 channel Rotel amp? I wouln't advise using your computer as a source for reference level. If you wanted to go active then maybe using a CD player and a pro crossover unit like something from Behringer would work. I don't know if there's many cheap 3-4 way crossovers. But then you would also have to have seperate inputs for each channel on the Rotel amp.
If you wait long enough, I'm sure there will be passive two-way designs using the Extremis and some tweeter.
I'm also confused about the electronics end. You have an 8 channel Rotel amp? I wouln't advise using your computer as a source for reference level. If you wanted to go active then maybe using a CD player and a pro crossover unit like something from Behringer would work. I don't know if there's many cheap 3-4 way crossovers. But then you would also have to have seperate inputs for each channel on the Rotel amp.
If you wait long enough, I'm sure there will be passive two-way designs using the Extremis and some tweeter.
kfr01 said:
Well, to be honest, I don't thinkt you should make 4-way system if you are new to DIY speaker building, and specially not with the combination of those units. Any idea of how big this monster would get?
No offence, but I would recommend doing something simple first to get some feeling for speaker DIY'ing, before plunging into such a massive project (don't underestimate a digital crossover, it is not simpler that a passive one!). Maybe a 2-way with the extremis as said above.
Thanks for the comments guys. I think I might try doing without the Koda 8. Why is the computer a bad source for reference level?
By digital crossover I meant a software + soundcard based solution. There has to be something like this out there.
One of you asked which amp I plan on using ... it is: http://www.rotel.com/products/specs/rmb1048.htm 8 channel amplifier.
By digital crossover I meant a software + soundcard based solution. There has to be something like this out there.
One of you asked which amp I plan on using ... it is: http://www.rotel.com/products/specs/rmb1048.htm 8 channel amplifier.
Get in line behind me.
I'm hoping someone comes up with a MT design when the Extremis 6 ships out. Adire will have something but I dont know when or how much the tweets will cost. Dont blame me for not trying though, I did my best with my limited knowledge.Kfr,
If I did my maths right, you could build a 2-way with the Extremis and then build a Tumult for about the same money (not including any amps).
BTW, the BL curve for the Extremis is up now. Very flat of course.
Telling me something is extremely complex doesn't really help. I have the wood working skills. I plan on skipping the passive crossover learning curve and step. Since I'm going sealed I don't think the volume will be out of control - especially if I remove the Koda 8 from the setup. Any other _real_ reasons why this won't work? I mean, thanks for warning me about the complexity .. but it would be better to give me -reasons- as to why it will be so complex or why I should not try it.
Also, one of you commented that 2khz is too low for the vifa ring radiator. http://www.d-s-t.com/vifa/data/xt25tg30-04d.htm
500hz is the resonant frequency and vifa lists 1.5-40khz as the usable frequency range. Won't 2khz by fine at 24db xo slope?
Also, one of you commented that 2khz is too low for the vifa ring radiator. http://www.d-s-t.com/vifa/data/xt25tg30-04d.htm
500hz is the resonant frequency and vifa lists 1.5-40khz as the usable frequency range. Won't 2khz by fine at 24db xo slope?
kfr01 said:
Okay, if you really want this... Then I think I should give you reasons why you should things differently (or not)
- first of all, I would not use the extermis in a MTM design. The woofer (even if they are great) are much to big, with means you have to cross way to low (probably under 2 Khz). I would recommend two smaller 4" or 5" woofers. You can use a higher X-over frequency and the smaller woofer have a better off axis response.
- Lose the Koda, but you already knew that
- I guess the X25 will handle the 2 Khz, but you should look at the distortion at those freuqencies, too!
- Don't use amps with the same power ratings for the various uints! The tweeter will not need nearly as much power as the shiva will. I would recommend to give the shiva lots of power (300W or so), and the tweater (and mid units), a little less (I guess 100W would be more than enough).
So, a MTM with 4" or 5" woofer, with a shiva (maybe on the side), will give you a better result, and if you use a Linkwitz transform, it doesn't need to be that big.
Sill: don't underestimate the digital X-over!
You cannot skip this learing curve, since with a digital X-over, you still have the same concepts! You still have to measure the units and simulate and try the filters. The only difference is that you don't have analog components to play with. You have a some kind of software to make the filters. It might seem less work, but I don't think it will be! I hope you will use a hardware digital X-over, and not your soundcard...(note that they are expensive)
Building a X-over is actualy the most difficult part of making a speaker. If it is bad, your speaker will be bad! I recommend sending a lot of time on research on this topic. Spend lots of time with tweaking the X-over! Fortunatly with a digital X-over, you can do that without the need to spend extra money
Good luck
Well you can read 'extremely complex' as 'you won't likely get a good result unless you are an expert at crossover design'. There are many talented speaker designers out there who don't even attempt a 3 way until they have 5 years expeience under their belts. In short - it would be a great learning exercise, but the odds that you would get the kind of 'reference system' you expect are fairly low, at least early on. Active or passive, a 4 way system is extremly complex to design. However, with enough time and experience, I expect you could get a nice result. It just won't be a 'run a few simulations, then plug and play' type process. If you decide to go ahead , you might look at something like this:
http://www.behringer.com/DCX2496/index.cfm?lang=ENG
controlled by some easy to use software like LSPCad. This would at least allow you to hear changes in your crossover in realtime, shortening your experimentation/tweaking process greatly.
http://www.behringer.com/DCX2496/index.cfm?lang=ENG
controlled by some easy to use software like LSPCad. This would at least allow you to hear changes in your crossover in realtime, shortening your experimentation/tweaking process greatly.
The XT tweeter has great freq response but the distortion level is a different story. This thread has some comments directly related to that tweeter: http://www.diyaudio.com/forums/showthread.php?s=&threadid=41780&perpage=10&pagenumber=3
So maybe push up the crossover freq or look into a different tweeter. And the Extremis probably can go higher than 2K without a problem. I know on the website it talks about using 3.5K.
I have used the 3.5mm jacks on my soundcard and converted it to RCA's to my reciever. I can't say I liked it, because the connection was "iffy" and I had to turn up the receiver higher to match the level of listening compared with my CD player.
Do you have a soundcard now that you want to use or have one picked out?
So maybe push up the crossover freq or look into a different tweeter. And the Extremis probably can go higher than 2K without a problem. I know on the website it talks about using 3.5K.
I have used the 3.5mm jacks on my soundcard and converted it to RCA's to my reciever. I can't say I liked it, because the connection was "iffy" and I had to turn up the receiver higher to match the level of listening compared with my CD player.
Do you have a soundcard now that you want to use or have one picked out?
Excellent. THANK YOU for sharing your full opinion with me. I really do appreciate it.
You mentioned that you hoped I wouldn't be using a soundcard solution. Why?
If that doesn't seem reasonable how would you alternatively feel about analog active pro-audio crossovers?
It just really _really_ seems that passive crossover design is like reinventing the wheel. Maybe I just haven't been into diy speaker building long enough to know all the reasons for passive crossovers, but I love my sound quality all-active car audio setup. It was so easy to adjust the crossover points and slope and listen to the results.
Again, thanks for the feedback and opinions.
You mentioned that you hoped I wouldn't be using a soundcard solution. Why?
If that doesn't seem reasonable how would you alternatively feel about analog active pro-audio crossovers?
It just really _really_ seems that passive crossover design is like reinventing the wheel. Maybe I just haven't been into diy speaker building long enough to know all the reasons for passive crossovers, but I love my sound quality all-active car audio setup. It was so easy to adjust the crossover points and slope and listen to the results.
Again, thanks for the feedback and opinions.
Ever measure your car audio setups?
Or is it "dial till it sounds best with the given components"? In which case, it may still sound horrible to many.
Understanding how drivers interact is important. I can give you a flat response that sounds horrible in most listening positions by making a point of using destructive interaction to flatten response where two drivers interact, which might sound OK in one spot, but go off-axis and it'll just be lumpy and horrible sounding.
You are welcome to do what you feel like. If you really only care if YOU like the results, I suppose it matters not. But you're here asking, which means it's not just for you.
C
Or is it "dial till it sounds best with the given components"? In which case, it may still sound horrible to many.
Understanding how drivers interact is important. I can give you a flat response that sounds horrible in most listening positions by making a point of using destructive interaction to flatten response where two drivers interact, which might sound OK in one spot, but go off-axis and it'll just be lumpy and horrible sounding.
You are welcome to do what you feel like. If you really only care if YOU like the results, I suppose it matters not. But you're here asking, which means it's not just for you.
C
Are you dead set on using an analog active crossover? Because otherwise you could use your PC as a source, feed the digital out to a behringer 2496 like I posted earlier, and control the 2496 via RS232 interface and LSPcad. I think for a reference system, at the very least you want to use a digital out from your PC, as the analog outs on most soundcards, especially popular ones, measure and sound quite awful.
I am in a similar situation as you, about to build a 'reference' system using my PC as the frontend. Unfortunately I can't afford the setup I just described, or any of what you've described for that matter, so I will likely feed the digital out from my PC to a Griffin powerwave, and build a very simple 2-way speaker for nearfield listening. To keep the crossover design as simple and foolproof as possible (I have no measuring equipment outside a RS SPL meter), I hope to use a widerange driver crossed to either a supertweeter or an active bass section, depending on where the wideband driver I end up with is weakest. This way I can (hopefully) keep the crossover out of the midrange zone, as I understand it is very easy to screw up a crossover at those frequencies. I built my first kit almost 10 years ago, and have been reading about this stuff for over 10, and to be honest I feel like I understand just enough to make a royal mess of a speaker , so I am still very cautious in my approach.
I am in a similar situation as you, about to build a 'reference' system using my PC as the frontend. Unfortunately I can't afford the setup I just described, or any of what you've described for that matter, so I will likely feed the digital out from my PC to a Griffin powerwave, and build a very simple 2-way speaker for nearfield listening. To keep the crossover design as simple and foolproof as possible (I have no measuring equipment outside a RS SPL meter), I hope to use a widerange driver crossed to either a supertweeter or an active bass section, depending on where the wideband driver I end up with is weakest. This way I can (hopefully) keep the crossover out of the midrange zone, as I understand it is very easy to screw up a crossover at those frequencies. I built my first kit almost 10 years ago, and have been reading about this stuff for over 10, and to be honest I feel like I understand just enough to make a royal mess of a speaker
, so I am still very cautious in my approach.
Luke (Uof Iowa) said:
Sure the Extermis can do that, but in a MTM setup, that is not wize. A normal MT with the shiva is very possible though! Besides, I never saw any off axis responses of the Exteremis.
kfr01 said:
Well, I meant the soundcard as a X-over. You need a lot of outputs, and massive CPU power (and of cource the right software, which is not very commom). If you have a good card and just use it as a source for the X-over, why not !
that would not be much cheaper than a digital X-over I guess. You could of cource make your own analog active X-overs
. But still, a digital X-over is much more flexible.I can imaging the ease of the active X-overs. But in a 3 of 4-way system it is not just as easy as setting a x-over point ans slope, and listening if it is okay. You have to measure it to get a real reference system. It would be a shame not to do that with such a system.
cjd said:
I've measured it with TrueRTA and flattened it with 10 bands of parametric eq. I think it sounds decent.
Please don't think I'm trying to say that I 'know it all'.. Gosh, far from it. That is WHY I'm here asking questions and trying to learn. Is it stupid to expect to put together a decent system using easily adjustable commercial active or digital (computer) crossovers? What do I need to do to understand how drivers interact as you recommend?
Thanks
I'll cast a supporting vote on the 'computer-based dsp crossover' approach. I also use this approach, and think it has a ton of advantages. However, I think what most people here are trying to do is just make it clear that crossover design is for the most part crossover design, and using a computer or other active techniques doesn't really make it 'easier', merely 'different'.
The best advice I can give is to seriously consider buying SoundEasy - it is a full speaker measurement/design package, but also directly supports emulating multi-way crossovers via your soundcard - both traditional analog topologies and linear-phase digital networks. It's not outrageous - $225 or so (you'll also need a measurement mic), and it'll be money well spend in the long run.
The best advice I can give is to seriously consider buying SoundEasy - it is a full speaker measurement/design package, but also directly supports emulating multi-way crossovers via your soundcard - both traditional analog topologies and linear-phase digital networks. It's not outrageous - $225 or so (you'll also need a measurement mic), and it'll be money well spend in the long run.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.