Hi Tony! What amount of bass is going to be enough for this application?
I haven't modeled many of these drivers but all I have modeled seem to be able to give 105dB without straining.
While perhaps not "Reference " level it is still fairly loud, especially as it is going to be added to the SPL of the mid and tweeter; although I was just thinking that would only add 3dB.
I like navins comments "I just happened on this thread. These are my thoughts. A dual woofer system can be made to work in either 3 way or 3.5 way.
Dual 10" woofers in 80-90 liters, mated to a 6" midbass and tweeter should work for most (some might find dual 6" woofers adequate, others might not be satisfied with even dual 12" woofers)."
Although my opinion is to use a smaller midrange I know of people who would use a 6inch in this application
I haven't modeled many of these drivers but all I have modeled seem to be able to give 105dB without straining.
While perhaps not "Reference " level it is still fairly loud, especially as it is going to be added to the SPL of the mid and tweeter; although I was just thinking that would only add 3dB.
I like navins comments "I just happened on this thread. These are my thoughts. A dual woofer system can be made to work in either 3 way or 3.5 way.
Dual 10" woofers in 80-90 liters, mated to a 6" midbass and tweeter should work for most (some might find dual 6" woofers adequate, others might not be satisfied with even dual 12" woofers)."
Although my opinion is to use a smaller midrange I know of people who would use a 6inch in this application
Thoughts from the peanut gallery.....
OK.... this is going to be long.... not the best speller.. so please be kind....
A lot of interesting ... quality info in this thread... I too am working a Classic 3 Way...
Some back ground.... have been designing speakers for almost 50 years... a lot of junk along the way... some periods of inactivity... but some fine successes too. I only bring up the 50 year thing to provide some support... maybe a bit of credibility to my comments. As my dad often said.. there are always different ways around the barn.....
IMO... a projects success is directly realated to the amount of sensible detail in what one wants for an outcome. Home work if you will ... plus a firm grasp of costs, ability and performance goals. Knowing where compromise is possible and where it is not. This is where an engineer really earns their money.
What I have concluded with my project is (in terms of initial questions)... enclosure... Monkey Coffin on a stand ala JBL L100 (1.6-2.3 cu. ft).... or floor stander (2.3-4 cu. ft.)
LF extension, max expected system output and efficiency. And the usuall WAF, Room and realistic financial considerations.
Once you have answers on these issues... then we need to be able to acknowledge the following...
1) Design integration... oh so important. Evey part has an effect on the other parts.... so easy to get into trouble if you do not think the design all the way through.
2) Crossover... IMO... 80% of the work is here... and that assumes the system design, drivers selected and performance goals are do-able and realistic.
3) Personal ability to actually build the system ... and more importantly... design and implement the all important crossover.
I do reallize all of this has been said god knows how many times... the above is just common sense... not glamorous (sp?) to be sure.
Bottom line is it all comes down to how realistic and serious you are. Not throwing stones here... just know one can loose focus on a project like this ... the anticipation...excitement... complexity and all the well intended advise. (Including this submission)
As for me.. I am doing the floor stander... it is just easier to execute and meets my performance goals. I have done all I can to keep the project as simple as possible.... working with drivers I have learned through experience work well together. This is key... because it make a tremendous difference in the crossover work. As in way easier...
Other things I am doing to keep things as easy as possible with out trading away first class performance...
Seal box. Just plain simpler... easy... predictable... Easy to build... stuff and measure. Part of this is I can live with a 90 dbw system efficiency. And the correct sealed woofer can do this.
A woofer that is flat to 2K (so I can run it up to 500 hz with out cross over heroics) and big enough to really deliver 105 db at or below 40 hz. Easier to work out the bass step issues too. That system integration thing....
A high quality dome mid.... no midrange enclosure.. predictable performance... easy implementation construction wise and crossover wise.
And a easy to integrate dome Tweeter.
All the above can be done with a 8-12 part count crossover. Basic 2nd order electrical... 4th order accoustic. Some frequency spreads to correct for horizontal demensional off sets... some pading of the midrange and a simple inductor - zobel combo on the woofer. I'm not out chasing peaks and dips with this trap or that. Don't have to do that with drivers that complement each other and that are operating in their intended band pass.
As for the drivers I am using... the Peerless 830669 12". I have used it in several projects and at less than $50. US wholesale it is a bargain and very consistant.
The mid is the Morel CAM 558... $60 US or so ... excellent... flexable performer and consistant.
The tweeted is the Seas 27TDFC... the giant killer. Again... flexable and easy to work with.. consistant. $35 US last I checked.
So there it is... in my mock ups this system will compete with just about any commercial system below $5K... and offers a nice balance of size, performance, simplicity and value. Very musical. Very dynamic... a bit large for small rooms and in certain WAF situations. Simple rectangular box tall enough to get the 4 cu ft net and a 42" tweeter axis height.
That's it from the Peanut gallery... let the slings and arrows begin!!
OK.... this is going to be long.... not the best speller.. so please be kind....
A lot of interesting ... quality info in this thread... I too am working a Classic 3 Way...
Some back ground.... have been designing speakers for almost 50 years... a lot of junk along the way... some periods of inactivity... but some fine successes too. I only bring up the 50 year thing to provide some support... maybe a bit of credibility to my comments. As my dad often said.. there are always different ways around the barn.....
IMO... a projects success is directly realated to the amount of sensible detail in what one wants for an outcome. Home work if you will ... plus a firm grasp of costs, ability and performance goals. Knowing where compromise is possible and where it is not. This is where an engineer really earns their money.
What I have concluded with my project is (in terms of initial questions)... enclosure... Monkey Coffin on a stand ala JBL L100 (1.6-2.3 cu. ft).... or floor stander (2.3-4 cu. ft.)
LF extension, max expected system output and efficiency. And the usuall WAF, Room and realistic financial considerations.
Once you have answers on these issues... then we need to be able to acknowledge the following...
1) Design integration... oh so important. Evey part has an effect on the other parts.... so easy to get into trouble if you do not think the design all the way through.
2) Crossover... IMO... 80% of the work is here... and that assumes the system design, drivers selected and performance goals are do-able and realistic.
3) Personal ability to actually build the system ... and more importantly... design and implement the all important crossover.
I do reallize all of this has been said god knows how many times... the above is just common sense... not glamorous (sp?) to be sure.
Bottom line is it all comes down to how realistic and serious you are. Not throwing stones here... just know one can loose focus on a project like this ... the anticipation...excitement... complexity and all the well intended advise. (Including this submission)
As for me.. I am doing the floor stander... it is just easier to execute and meets my performance goals. I have done all I can to keep the project as simple as possible.... working with drivers I have learned through experience work well together. This is key... because it make a tremendous difference in the crossover work. As in way easier...
Other things I am doing to keep things as easy as possible with out trading away first class performance...
Seal box. Just plain simpler... easy... predictable... Easy to build... stuff and measure. Part of this is I can live with a 90 dbw system efficiency. And the correct sealed woofer can do this.
A woofer that is flat to 2K (so I can run it up to 500 hz with out cross over heroics) and big enough to really deliver 105 db at or below 40 hz. Easier to work out the bass step issues too. That system integration thing....
A high quality dome mid.... no midrange enclosure.. predictable performance... easy implementation construction wise and crossover wise.
And a easy to integrate dome Tweeter.
All the above can be done with a 8-12 part count crossover. Basic 2nd order electrical... 4th order accoustic. Some frequency spreads to correct for horizontal demensional off sets... some pading of the midrange and a simple inductor - zobel combo on the woofer. I'm not out chasing peaks and dips with this trap or that. Don't have to do that with drivers that complement each other and that are operating in their intended band pass.
As for the drivers I am using... the Peerless 830669 12". I have used it in several projects and at less than $50. US wholesale it is a bargain and very consistant.
The mid is the Morel CAM 558... $60 US or so ... excellent... flexable performer and consistant.
The tweeted is the Seas 27TDFC... the giant killer. Again... flexable and easy to work with.. consistant. $35 US last I checked.
So there it is... in my mock ups this system will compete with just about any commercial system below $5K... and offers a nice balance of size, performance, simplicity and value. Very musical. Very dynamic... a bit large for small rooms and in certain WAF situations. Simple rectangular box tall enough to get the 4 cu ft net and a 42" tweeter axis height.
That's it from the Peanut gallery... let the slings and arrows begin!!
MoonDog since you seem to be partly on the same page I'll add some more detail to my earlier posts.
For the purpose of a loudspeaker that can be built by most I propose we steer away from the exotic ideas (wideband/fullrange/ceramic cones/AMTs etc..) unless we can settle on a tried and tested design like this
Strassacker: Lautsprecher - Boxen - Selbstbau
"Exotics" like Lowthers, Fostex, Thiel/Accuton, etc.. require far more skill than many of us have and I think your idea is to have enough people in on this as possible.
The choice of using a 6" for the mid was because this system can easily be expanded for home theater (with the 6" mid and tweeter doing duty for the other 3-5 channels). Also post the 18W8545/8546 (early 90s) I have come to realise that many driver manufacturers have finally been able to control cone breakup as high as 3kHz which is plenty for even a simple 12db/oct crossover. Using a smaller driver should allow you to crossover higher (about an octave higher) but I find fewer 5" mids that can go from 200-5K as cleanly than I find 6" mids that can go from 100-2.5kHz. Some time ago John Krutke tested a 6" (18Sound 6ND430-16) that would rival many a 5" for midrange reproduction (however this driver is not easily available across the world for it to be considered).
Since the mid is where our ears are most sensitive I suggest that we first agree on this driver (this alone can be a rather long and fiercely debated topic). Once this driver is chosen the choices for matching bass and hf drivers will flow easier. Atleast this way we tackle only one variable at a time.
If budget, cabinet skills, and XO design was not a problem SEAS/ScanSpeak systems like those proposed below are an option.
ScanSpeak:
Woofer 2 x 23W4557 + optional 23W passive radiators (35 liters each - 70 liters total)
Midbass 18W8531
Tweeter 3004-6600 (for the soft domeies) or 6640 for the hard domeies
SEAS:
Woofer 2 x W26FX002 (55 liters sealed x 2 = 110 liters total)
Midbass W16 (NX for the soft conies, EX for those who prefer Mg cones)
Tweeter T29CF002
a system that is very similar has already been done by Troels here
W26 Classic
for those on a budget something along these lines...
2 x Peerless 830452 10" woofers (can work in 20 liters but 30 liters works better) or Dayton RS270 (why do they not make a non-shileded version)
1 x SB Acoustics SB17NRX35 or SEAS L18/CA18
1 x SB Acoustics SB29RDC or SEAS T27TFFC (with chamber) or Peerless 810921
"John K" has already got a L18 based design here...
Zaph|Audio
Perhaps some would prefer considering Pro sound drivers from PHL/Beyma/18Sound etc. I wond er if these drivers are as easily available though.
Personally I would love to see the SEAS Excel based system be done if only becuase at that price point it would not only require REFERENCE crossover components and design and excellent cabinetery but also becuase for those who want to expand this system into Home Theater duty, SEAS makes a coax driver (C16N001) and the new U series (U16RCY).
Can I ask why not the Seas MCA15RCY as already suggested by others I believe does not make it in to your list. It seems to have a nice broad useable range, with good distortion figures and being a 5" used as a midrange it should be capable of decent volume levels.
Hi Tony
I tend to be in the vented camp to get more bass extension. The volumes for those drivers I got from Madisound's recommended size as with a new computer I haven't installed the modelling software. I think they have struck a good compromise between extension and power handling.
I don't know how low builders would want to go but 35Hz is not a bad goal with a 10" driver.
I agree that the MCA15RCY is a good mid choice and can also be the mid woofer variant if needs to be crossed over lower.
What ever the outcome of this thread it is providing a lot of ideas, a good resource and driver choices for those who want to build a 3-way. Researching drivers can take a lot of time but there are several great options that have popped up here. Keep it up.... good stuff.
One choice we haven't discussed seriously yet, although it has been mentioned; is a driver and cross-over to convert the 2-way speaker to a 3-way or 3.5way, for those who have the Vifa P-13/D27 combo or who can still get it.
Speakerbits in North Melbourne does have some new stock on their website, although a year ago I was told to buy now as they had stopped production and were to become the proverbial "Unobtanium" as does Madisound
Speakerbits in North Melbourne does have some new stock on their website, although a year ago I was told to buy now as they had stopped production and were to become the proverbial "Unobtanium" as does Madisound
Oh my! the list was just an example and by no means comprehensive. It was only suggestive. My proposal was that we
a) lay down the definition of reference - is this going to speaker that many of us can afford to build (cost, skills, etc..) or it is going to be a true reference project (with Excel/Revelator/Illuminator quality drivers)?
b) we bringour collective expereince together to first choose the Mid driver as this is where our ears are most critical - if the MCA15RCY is the choice of the majority hey I am cool with it.
BTW as far as the MCA15RCY is concerned has it not been been superceededby the U16RCY? Is there a significant cost difference between these two?
In room 35Hz with a 10" in say 40 liters sealed is not diffcult. Vented does give more bass efficiency but I suspect (and I can be wrong) that most of us will place these speakers close (within 2m) of atleast the rear wall, besides since these speakers will be floor standers they will get gain from the floor. Hence maybe sealed (sealed boxes are also more tolerant of small deviations in either driver specs or cabinet contruction) might be the way to go.
there are a lot of options for 35Hz from 2 x 10" in 60-80 liters sealed - the Peerless 10" (830452) is one such driver.
Rabbitz, only since you are familar with Madisound's catalog (I am in India and dont have regular access) what do you think of SB Acoustics drivers (specifically the SB29NRX, SB17NRX, and SB29RDC).
1. The P13/D27 drivers are not available for many of us.
2. Coverting a 2 way to a 3 or 3.5 way may require a complete XO redesign.
3. Are the P13/D27 still "reference"? I thought the game has moved on.
But why the suggestion of obscure drivers? I thought we had decided several pages ago to choose drivers that are easily obtainable. Which was what led to the suggestion of the Seas 27TDFC/TFFC and the MCA15RCY they are easily obtainable with proven performance and are relatively cheap. I'm not trying to be difficult, it just has me confused and/or interested as to why people keep suggesting obscure or unusual drivers that don't get used very often, is the difficulty in obtaining them worth the time and effort, is it worth creating a design that in 12months time may be outdated purely because driver sourcing becomes impossible.
Last edited:
Sorry I did not mean to come across as rude or abrubt, I am genuinely interested in what makes the drivers you have suggested more appropriate, as I don't recall seeing them used in other designs so I have very little knowledge of them.
John thanks for the input. Nothing will be thrown from this camp 
eminently sensible! I Like the idea of dome mids, but always consider them for a four way application... The MDM55's were always on my list, I assume the ones you mention are the newer generation.
Ted, I don't know how low is low enough I look at the models for sealed boxes and see them (gently) rolling off starting at around the 100Hz mark... whereas the BR tend to be flat down to maybe 40Hz with the same driver.. Sealed has other advantages such as superior transient response (one of the reasons I wanted to try a TL, but my woofers aren't suitable for TL's). However I'm not experienced in the difference in sound from a Sealed speaker compared to a bass reflex with lower extention... My speakers do dual duty as Stereo and for Movies (still only stereo) so I like them to be able to do realistic sound effects without a separate sub... They are currently leaky and not reaching optimal performance, I did some measurements ages ago, with a dodgy computer mike which showed the -3db at around 57Hz (modeled at 36Hz) but I didn't trust it.... so maybe I'm worrying about nothing... I plan to seal these boxes properly sometime over the next couple of months and I'll do some before and after nearfiled (and maybe 1M) measurements. So maybe after that I will be in a better position to comment
Rabbitz, I might have been having a bad day with unibox today I was sure the first time I modelled that driver that I got quite good results with it both vented and sealed (but maybe it was a peerless)... I'll have another go and post the results I think I was confused.. Just looked at the CA26RE4X (as opposed to the A26RE4 which I modelled today)... will do the sim on that, as it looks a lot more appropriate
Tony.
eminently sensible! I Like the idea of dome mids, but always consider them for a four way application... The MDM55's were always on my list, I assume the ones you mention are the newer generation. Ted, I don't know how low is low enough
I look at the models for sealed boxes and see them (gently) rolling off starting at around the 100Hz mark... whereas the BR tend to be flat down to maybe 40Hz with the same driver.. Sealed has other advantages such as superior transient response (one of the reasons I wanted to try a TL, but my woofers aren't suitable for TL's). However I'm not experienced in the difference in sound from a Sealed speaker compared to a bass reflex with lower extention... My speakers do dual duty as Stereo and for Movies (still only stereo) so I like them to be able to do realistic sound effects without a separate sub... They are currently leaky and not reaching optimal performance, I did some measurements ages ago, with a dodgy computer mike which showed the -3db at around 57Hz (modeled at 36Hz) but I didn't trust it.... so maybe I'm worrying about nothing... I plan to seal these boxes properly sometime over the next couple of months and I'll do some before and after nearfiled (and maybe 1M) measurements. So maybe after that I will be in a better position to comment Rabbitz, I might have been having a bad day with unibox today
I was sure the first time I modelled that driver that I got quite good results with it both vented and sealed (but maybe it was a peerless)... I'll have another go and post the results I think I was confused.. Just looked at the CA26RE4X (as opposed to the A26RE4 which I modelled today)... will do the sim on that, as it looks a lot more appropriate Tony.
OK here are some models of the Seas CA26RE4X (basically the specs on the madisound site).
The sealed on has NO stuffing at all (that was the only way to get something to match with the specified 30L
The vented is with "walls covered".
The first graph compares the sealed vs vented performance for the two models..
The next two are peak excursion graphs for sealed vs vented... The sealed is the safer of the two in that it doesn't matter what freq is sent to the woofer it will never exceed it's maximum mechanical range of +- 10mm with all powers up to its rated 80W.
The vented however will have problems if fed high power signals less than 25Hz (with 17mm peak excursion at 20Hz and 80W).
I've posted the excursion plots at 50W power.. at 20W power the Bass reflex stays under the 4mm xmax, the Sealed still exceeds it slightly. Overall the vented box I think wins hands down for this woofer (provided it is not fed with high power at less that 25Hz), with much better bass extension and with what should be quite a bit lower distortion at moderate levels.
I haven't tried any models for this woofer other than those posted on the madisound site
edit: that comment about the sealed vs vented excursion is a bit exagerated... plotting together on the same scale there isn't a lot in it. the BR has the edge at 30W where it remains under the Xmax whereas the Sealed goes slightly over for freqs below about 55Hz... added the other graph to show... I don't know of a way to remove the other enclosure types...
Tony.
The sealed on has NO stuffing at all (that was the only way to get something to match with the specified 30L
The vented is with "walls covered".
The first graph compares the sealed vs vented performance for the two models..
The next two are peak excursion graphs for sealed vs vented... The sealed is the safer of the two in that it doesn't matter what freq is sent to the woofer it will never exceed it's maximum mechanical range of +- 10mm with all powers up to its rated 80W.
The vented however will have problems if fed high power signals less than 25Hz (with 17mm peak excursion at 20Hz and 80W).
I've posted the excursion plots at 50W power.. at 20W power the Bass reflex stays under the 4mm xmax, the Sealed still exceeds it slightly. Overall the vented box I think wins hands down for this woofer (provided it is not fed with high power at less that 25Hz), with much better bass extension and with what should be quite a bit lower distortion at moderate levels.
I haven't tried any models for this woofer other than those posted on the madisound site
edit: that comment about the sealed vs vented excursion is a bit exagerated... plotting together on the same scale there isn't a lot in it. the BR has the edge at 30W where it remains under the Xmax whereas the Sealed goes slightly over for freqs below about 55Hz... added the other graph to show... I don't know of a way to remove the other enclosure types...
Tony.
Attachments
Last edited:
Hi navin
The only SB Acoustics I've played with is the SB15NRX. The build quality is exceptional and close to Scan Speak quality, great motor with copper in the right places and very low inductance voice coil. I compared it to the 18W8531G00 (not really fair) and certainly came close in a lot of areas. I still think they need a bit of work on their cone material as there was a hint of hollowness in the mids (I thought so anyway and Andy G also picked it up on a listen). Experimenting with damping coatings did improve it. I think the cone material is probably the same as on their woofers but once they sort it out on the mid woofers, look out.... early days yet.
They are easy to work with and are one of only a few drivers that don't have that bathtub shape plot in the lower frequencies for power handling modelling.... very impressive. It went lower and better controlled than I expected and the strange blips higher on the FR didn't show up in listening even when using 2nd order electrical crossovers but I don't listen directly on axis.
Have a look at the nude SB15NRX.
Attachments
I think there is a discrepancy between real life use and the modelled excursion on drivers. There was a discussion about this (I think in the FR forum) and something about music waves vs sine waves etc. All I know I have several drivers that should clap out with only a few watts but my ears clap out before they do. I tend to use it now as a rough guide only and for comparisons. I usually use respected manufacturer's power handling figures as a guide to see if it falls in with the intended use. Of couse I could be completely off track on this subject.
I've only used an old version of MJK software, which allow for length (73"), Driver position ratio (24"), So = 2.4Sd, SL= 0.8Sd and stuffing density = 0.1lb.
It was done in a hurry, so I'm sure it can be improved.
Well I do not consider the combination obscure, old>? yes, but I was thinking more of those people who may have bought the combination and not yet made the 2-way
( Holding my hand in the air here )
Drivers may have got better but perhaps not all drivers are so well documented and easy to work with and we still have not DEFINITELY shortlisted any thing.
I would still consider this post at the camp-fire discussion stage.
Could well be Rabbitz (that there is in general a bad corelation between the models and real life, not that you are off track
) , I got scolded a few years ago (here on diyAudio) for not taking cone excursion into consideration when modelling stuff (in particular BR) so now do it as a matter of course...It could be that most people are never reaching more than say 10W input power
My first amp (well first I made) was a 10W/channel one, and It used to be plenty loud enough for me in my 15 saqure meter bed sit I do recall when I first fired up my prototype MTM and cranked it that I was rather surprised at the excursion of the 5" drivers.. In that case they seemed to be extending more than I expected... I'd have to revisit the model though to be sure..
I can see the argument for music signals vs sine waves though... A pure sine wave will max out the driver but music is full of varying transients so in real life will put quite differing demands on the driver.
Tony.
Many drivers have progressive suspensions, which limits the efficacy of your typical model for excursion. There's also the Xmax vs. Xlim thing. Many drivers with very high efficiency have very low Xmax to maximize the coil in the gap without adding lots of extra mass. They run out of Xmax right away but Xlim is usually quite a bit higher.
Not that you don't know all this, but might be useful to some others watching
Points taken, but I considered the point of a project like this is one that anyone can build. For people who already own particular drivers it may be fine to recommend older drivers but you've limited it to yourself and others who may happen to have those drivers. If you choose drivers that are easily obtainable around the world and are part of a current lineup that is not likley to be removed any time soon, then you open the project up to everyone plus you will most likely get more input because people are interested in contributing to a project that they can actually build because they can get the drivers, if they can't it won't hold much interest for them. I guess there is also no reason why the community couldn't come up with a basic design with a 2 or 3 xover options for other drivers.
I hope you haven't taken anything i've said as argumentive, I don't mean to come across like that, I guess I'm just trying to put forward the opinion of someone who is still a learner in this area, and a project like this sparks my interest greatly. I feel the same about mulitiple 10" or 12" drivers, is it really necessary to use multiple 10" drivers in the base design to produce a system with a nice smooth response at normal listening levels plus a bit extra. Maybe I misinterpreted but I thought the original idea was to produce a relatively simple design which has a reasonable level of performance for a reasonable budget.