Re: why routing the sides?
I have built my first ariels without routing, just glued everything piece by piece, using MDF. the second pair I made (www.orfeas.8m.com) was done using the routing method (with the help of CAM) I supose they become much more rigid and air tight. Also you can get pretty Identical sections for left and right speaker
zdr said:
I have built my first ariels without routing, just glued everything piece by piece, using MDF. the second pair I made (www.orfeas.8m.com) was done using the routing method (with the help of CAM) I supose they become much more rigid and air tight. Also you can get pretty Identical sections for left and right speaker
Ariel sound
I would also like to ask Chrish and Gazzafloyd about the performance of their Ariels. After having built Ariel Mk5, my brother decided to go for Mk6, just like the one Chris Built. However, we are pretty desapointed from the sound... We have lost that sparkle from the excellent Scanspeak tweeter and music, including vocals, has no presence...
I guess we have to keep playing around with stuffing materials, and x-over, but I would like your opinions on this
Orfeas
I would also like to ask Chrish and Gazzafloyd about the performance of their Ariels. After having built Ariel Mk5, my brother decided to go for Mk6, just like the one Chris Built. However, we are pretty desapointed from the sound... We have lost that sparkle from the excellent Scanspeak tweeter and music, including vocals, has no presence...
I guess we have to keep playing around with stuffing materials, and x-over, but I would like your opinions on this
Orfeas
Just to add my own tuppence worth.
I like the Ariel. I used to own a V4, which I sold a few months back when the number of speakers I owned was threatening to swamp the house. However... I think that the drivers and crossover are better than the cabinet is. Ariel was designed before there was any complete modelling theory regarding quarter-wave resonators, and the cab is also both extremely complex and relatively expensive to construct.
Here's an MLTL I modelled in the latest version of Martin King's MathCad worksheets (due for release around March). It uses the same Seas / Vifa drivers and Lynn's excellent crossover in an easier to build cabinet. I first posted this in the Full-range Forum in the thread on Martin's new sheets, but it might be of some use here.
Straight MLTL assuming 3/4" build material.
Internal line-length: 45" (external cabinet height 47 1/4" as best to double the top-panel).
Internal width: 6 1/2" (so preserving the external 8" width and avoiding any need for crossover modifications)
Internal depth: 12"
Circular port, 4"x2" (WxL) 3" up from the internal base on the front panel.
MTM driver layout, same as existing Ariel versions. Tweeter centre 11 1/4" down from the internal top or So.
0.25lbs ft^3 of Dacron (or similar) stuffing from the top to 4" below the lower driver.
Keep the large radii on the edges and top if you wish -no reason why not. You could also add a couple of internal braces too. Predicted anechoic response is shown below:
I like the Ariel. I used to own a V4, which I sold a few months back when the number of speakers I owned was threatening to swamp the house. However... I think that the drivers and crossover are better than the cabinet is. Ariel was designed before there was any complete modelling theory regarding quarter-wave resonators, and the cab is also both extremely complex and relatively expensive to construct.
Here's an MLTL I modelled in the latest version of Martin King's MathCad worksheets (due for release around March). It uses the same Seas / Vifa drivers and Lynn's excellent crossover in an easier to build cabinet. I first posted this in the Full-range Forum in the thread on Martin's new sheets, but it might be of some use here.
Straight MLTL assuming 3/4" build material.
Internal line-length: 45" (external cabinet height 47 1/4" as best to double the top-panel).
Internal width: 6 1/2" (so preserving the external 8" width and avoiding any need for crossover modifications)
Internal depth: 12"
Circular port, 4"x2" (WxL) 3" up from the internal base on the front panel.
MTM driver layout, same as existing Ariel versions. Tweeter centre 11 1/4" down from the internal top or So.
0.25lbs ft^3 of Dacron (or similar) stuffing from the top to 4" below the lower driver.
Keep the large radii on the edges and top if you wish -no reason why not. You could also add a couple of internal braces too. Predicted anechoic response is shown below:
Attachments
Not yet. I will be at some point later this year, but I have a couple of other things to finish first. I used to own the V4, and I loved the drivers and crossover, but the cabinet... no.
Ariel was designed using traditional rule of thumb methods by Lynn before anyone really understood how quarter-wave resonators actually behaved. Don't get me wrong, I'm not knocking it or him, I'm just saying that we've made some major advances in the field, thanks to Martin, and now we know how they work and how to model and predict their responses. The old closed end with driver, open far end, tapered or untapered line is sub-optimal in 99% of conditions. Mass-Loaded Transmission Lines almost invariably give better performance than the 'classic' geometry, and I long ago learned to trust Martin's MathCad worksheets. This new one I'm playing with at the moment is a stunner. Based on my own experience with the V4, and with numerous different MLTL and ML TQWT enclosures designed in these worksheets, I'm pretty confident this MLTL should easily better any of the existing enclosure designs in terms of its frequency response, while also being a heck of a lot easier to build. I know that many point out that a flat response isn't everything: fair enough, and I agree. But I like to get as close to it as possible. Those concerned over materials could build the whole thing from birch ply (my suggestion), or, like the V6, MDF outer panels with doubled front and rear baffles, the inner pieces being ply, with some ply bracing if you feel the need, so you can blend the art of the originals into it.
OK, so lots of Ariel owners will probably want to crucify me for speaking such heresy, but it's not actually that different: it's still a quarter-wave resonator, albeit with a wider and shorter line, and most Ariels actually employ a form of mass-loading anyway due to the restrictions on their floor-firing terminus imposed by the spacers, which can be raised and lowered according to taste. Consider it another option in a world of infinite possibilities.
Best
Scott
Ariel was designed using traditional rule of thumb methods by Lynn before anyone really understood how quarter-wave resonators actually behaved. Don't get me wrong, I'm not knocking it or him, I'm just saying that we've made some major advances in the field, thanks to Martin, and now we know how they work and how to model and predict their responses. The old closed end with driver, open far end, tapered or untapered line is sub-optimal in 99% of conditions. Mass-Loaded Transmission Lines almost invariably give better performance than the 'classic' geometry, and I long ago learned to trust Martin's MathCad worksheets. This new one I'm playing with at the moment is a stunner. Based on my own experience with the V4, and with numerous different MLTL and ML TQWT enclosures designed in these worksheets, I'm pretty confident this MLTL should easily better any of the existing enclosure designs in terms of its frequency response, while also being a heck of a lot easier to build. I know that many point out that a flat response isn't everything: fair enough, and I agree. But I like to get as close to it as possible. Those concerned over materials could build the whole thing from birch ply (my suggestion), or, like the V6, MDF outer panels with doubled front and rear baffles, the inner pieces being ply, with some ply bracing if you feel the need, so you can blend the art of the originals into it.
OK, so lots of Ariel owners will probably want to crucify me for speaking such heresy, but it's not actually that different: it's still a quarter-wave resonator, albeit with a wider and shorter line, and most Ariels actually employ a form of mass-loading anyway due to the restrictions on their floor-firing terminus imposed by the spacers, which can be raised and lowered according to taste. Consider it another option in a world of infinite possibilities.
Best
Scott
Scott, this is something I have been thinking of for a long time, so it is great that you have gone ahead and started simulating it. Thank you! I have felt too inexperienced in the speaker designing business to try out Martin Kings calculations with the Ariel drivers, but am very interested in DIY and good sound. If this works out it could be a killer DIY speaker concept... So I for one would love for the MLTL box to perform better than the Ariel box - as you say, it is so much easier to build. Would save a lot of work and (potential) headache.
However, I have no experience in the differences between MLTL boxes and labyrinth TL:s. There must be though - bass performance, for one, ought to be a bit different. What differences in sound character is there to expect?
Then there is the double "stagger-tuned" TL that Lynn O designed, with each driver getting its own line. I don't know how this could be pulled off in a straight box. Is it necessary? What do you think?
However, I have no experience in the differences between MLTL boxes and labyrinth TL:s. There must be though - bass performance, for one, ought to be a bit different. What differences in sound character is there to expect?
Then there is the double "stagger-tuned" TL that Lynn O designed, with each driver getting its own line. I don't know how this could be pulled off in a straight box. Is it necessary? What do you think?
Good questions all.
Hmm. Bass differences. Well, the MLTLs tend to go deeper, smoother than the 'classic' (wince) tapered geometry terminating in free-space. They are very neutral and well damped, and have an extremely pure bass and midrange lacking the ripple common in most other enclosures, including traditional, unstuffed TLs, or the conjestion in traditional, over-stuffed TLs.
A quick bit of theory for you which should help explain things. 'Transmission Line' is an inaccurate description created by Bailey some 30 years ago for a heavily stuffed, straight line. It's an electrical term, not an accoustic one at all. The correct name is Quarter Wave Resonator, and they are designed to resonante strongly at a specific frequency. That resonance is called the Fundamental. Now, the problem is that you also create unwanted higher harmonic resonances in the process. (In a straight pipe, these occur at multiples of the Fundamental frequency, in a tapered line, they occur at different points). You can supress some of these by altering the position of the driver along the line-length, and try to damp the others with stuffing, but stuffing brings its own problems: too much, and you start to absorb the Fundamental resonance you were aiming for in the first place. Not so good. Fortunately, we can model these enclosures accurately now, using Martin's software, so we know in advance where issues are likely to occur, but the basic enclosure isn't ideal.
Enter Mass Loading. The air-mass in the restricted terminus provides an additional load to the quarter-wave resonances in the enclosure. This in practical terms does two things. Firstly, it allows a much shorter pipe. Let's assume we have a pipe, open at one end resonating at 60Hz. By correctly mass-loading it, you can decrease this pipe resonance to around 35Hz. The second benefit is that it frequently reduces ripple over 100Hz, which can only be a good thing.
Is the twin line necessary? In my view -no. There have been reported subjective improvements in the later versions of Ariel once they abandoned the original geometry, which joined the two driver pathways into a single labyrinth, I believe regarding sorting out a very slight touch of mid-bass 'muddle'. I haven't heard the early versions, so I can't speak from experience. However, I can note a couple of things. Firstly, the initial path-lengths were considerably different -effectively, you were joining two quarter-wave resonators of different lengths, and hense, different tuning frequencies, together. Not such a good idea. Secondly, the enclosure was designed under the old rule of thumb methods, and wasn't optimised in any case. In an MLTL, both drivers 'see' an identical line-length and resonant mode, which they simply excite at slightly different points, so 'muddle' isn't going to be an issue. Ripple might be fractionally greater than for an equivelant setup with one driver, but this will be damped out by the light stuffing and mass-loading. In this case, they are pretty small drivers anyway, and so close together that differences will be very minimal.
Hope some of this helps
Regards
Scott
Hmm. Bass differences. Well, the MLTLs tend to go deeper, smoother than the 'classic' (wince) tapered geometry terminating in free-space. They are very neutral and well damped, and have an extremely pure bass and midrange lacking the ripple common in most other enclosures, including traditional, unstuffed TLs, or the conjestion in traditional, over-stuffed TLs.
A quick bit of theory for you which should help explain things. 'Transmission Line' is an inaccurate description created by Bailey some 30 years ago for a heavily stuffed, straight line. It's an electrical term, not an accoustic one at all. The correct name is Quarter Wave Resonator, and they are designed to resonante strongly at a specific frequency. That resonance is called the Fundamental. Now, the problem is that you also create unwanted higher harmonic resonances in the process. (In a straight pipe, these occur at multiples of the Fundamental frequency, in a tapered line, they occur at different points). You can supress some of these by altering the position of the driver along the line-length, and try to damp the others with stuffing, but stuffing brings its own problems: too much, and you start to absorb the Fundamental resonance you were aiming for in the first place. Not so good. Fortunately, we can model these enclosures accurately now, using Martin's software, so we know in advance where issues are likely to occur, but the basic enclosure isn't ideal.
Enter Mass Loading. The air-mass in the restricted terminus provides an additional load to the quarter-wave resonances in the enclosure. This in practical terms does two things. Firstly, it allows a much shorter pipe. Let's assume we have a pipe, open at one end resonating at 60Hz. By correctly mass-loading it, you can decrease this pipe resonance to around 35Hz. The second benefit is that it frequently reduces ripple over 100Hz, which can only be a good thing.
Is the twin line necessary? In my view -no. There have been reported subjective improvements in the later versions of Ariel once they abandoned the original geometry, which joined the two driver pathways into a single labyrinth, I believe regarding sorting out a very slight touch of mid-bass 'muddle'. I haven't heard the early versions, so I can't speak from experience. However, I can note a couple of things. Firstly, the initial path-lengths were considerably different -effectively, you were joining two quarter-wave resonators of different lengths, and hense, different tuning frequencies, together. Not such a good idea. Secondly, the enclosure was designed under the old rule of thumb methods, and wasn't optimised in any case. In an MLTL, both drivers 'see' an identical line-length and resonant mode, which they simply excite at slightly different points, so 'muddle' isn't going to be an issue. Ripple might be fractionally greater than for an equivelant setup with one driver, but this will be damped out by the light stuffing and mass-loading. In this case, they are pretty small drivers anyway, and so close together that differences will be very minimal.
Hope some of this helps
Regards
Scott
Thanks for taking the time to explain, Scott. Good answers! Time for me to do a serious read-up on MK:s web pages - or perhaps just to start building according to your suggestions...
A while ago a DIY friend invited me to listen to his ME2:s, driven by a 211PSE of his own design. It sounded so darn good that I knew, then and there, that I simply had to get the Ariels ready. Good for me I hadn't started building the cabinets! To me this "MLTL Ariel" is the most interesting development of the Ariel concept so far.
I will draw something up according to your measures - I can't seem to get the hang of things before I see an image - and come back here in a day or two, maybe with some more questions. Thanks again for explaining!
A while ago a DIY friend invited me to listen to his ME2:s, driven by a 211PSE of his own design. It sounded so darn good that I knew, then and there, that I simply had to get the Ariels ready. Good for me I hadn't started building the cabinets! To me this "MLTL Ariel" is the most interesting development of the Ariel concept so far.
I will draw something up according to your measures - I can't seem to get the hang of things before I see an image - and come back here in a day or two, maybe with some more questions. Thanks again for explaining!
Here goes nothing
OK, I did a drawing.
Image should be in this message if everything works out. Actually, guess what? there are some questions...
The 3 inches distance from the inside bottom of the speaker to the hole - is that
1. the distance between the center of the hole and the bottom, or,
2. as I drew it, 3 inches of clearance between the hole and the inside bottom?
I don't know anything about MLTL ports. Are they constructed something like a BR port? So I can use whatever I want with the correct diameter? And the length of it, is that
1. the total length from the baffle front, including whatever depth the baffle has, as I did it, or is it
2. the length of port that protrudes from the inside of the baffle into the speaker?
For now I plan to use birch ply for all sides, spruce for the front baffle since it is a nice and soft wood, and birch or spruce for the top "lid".
OK, I did a drawing.
Image should be in this message if everything works out. Actually, guess what? there are some questions...
The 3 inches distance from the inside bottom of the speaker to the hole - is that
1. the distance between the center of the hole and the bottom, or,
2. as I drew it, 3 inches of clearance between the hole and the inside bottom?
I don't know anything about MLTL ports. Are they constructed something like a BR port? So I can use whatever I want with the correct diameter? And the length of it, is that
1. the total length from the baffle front, including whatever depth the baffle has, as I did it, or is it
2. the length of port that protrudes from the inside of the baffle into the speaker?
For now I plan to use birch ply for all sides, spruce for the front baffle since it is a nice and soft wood, and birch or spruce for the top "lid".
Attachments
You're welcome. I like the drawing by the way!
Sorry, I should have been clearer about the port position -force of habit. You're frst guess is the correct one: 3" from the internal base of the cabinet to the centre of the port, which puts it as close to the floor as possible. As for port materials: same as you'd use for a reflex enclosure, which is helpful. MLTLs don't require anything special in this regard. Port length includes the depth of the front baffle, so again, you're first surmise is the correct one.
Material choice looks good to me. One suggestion: double the top-panel. The top of a QWR has the greatest pressure appled on it, so doubling it should help prevent resonances. As I say, I haven't built it myself (yet -give me a few months!), but I haven't gone wrong so far with anything I've designed and modelled before building in Martin's worksheets. I reckon this one could be very impressive. It's well worth reading the articles and papers on Martin's site: the amount I've learned from those, and from chatting with him is truly frightening, and I owe him a great debt of thanks for all his help, which has had the side benefit of stimulating my own interest in many new fields I might never have looked at.
I wonder what Lynn would make of it. I might fire an email his way and get his views if he has the time in between his editing and designing etc.
Cheers
Scott
Sorry, I should have been clearer about the port position -force of habit. You're frst guess is the correct one: 3" from the internal base of the cabinet to the centre of the port, which puts it as close to the floor as possible. As for port materials: same as you'd use for a reflex enclosure, which is helpful. MLTLs don't require anything special in this regard. Port length includes the depth of the front baffle, so again, you're first surmise is the correct one.
Material choice looks good to me. One suggestion: double the top-panel. The top of a QWR has the greatest pressure appled on it, so doubling it should help prevent resonances. As I say, I haven't built it myself (yet -give me a few months!), but I haven't gone wrong so far with anything I've designed and modelled before building in Martin's worksheets. I reckon this one could be very impressive. It's well worth reading the articles and papers on Martin's site: the amount I've learned from those, and from chatting with him is truly frightening, and I owe him a great debt of thanks for all his help, which has had the side benefit of stimulating my own interest in many new fields I might never have looked at.
I wonder what Lynn would make of it. I might fire an email his way and get his views if he has the time in between his editing and designing etc.
Cheers
Scott
drawings, v2
Alrightie!
Port moved further down, and lengt kept as it was. The large radii weren´t quite right so they are redrawn.
I added a top view and made another drawing with some colours and shades to get a better idea of the appearance. I thought perhaps it would be fun to try putting felt around the drivers. Or leather, or something. This is the light grey area you see in the coloured/shaded part.
This is drawn assuming 21mm birch ply for the side + bottom panels and 40mm spruce, maybe made with solid wood "square profiles" (sorry, I am swedish and lack some proper english/american words regarding wood working, just ask me to explain and/or correct me if I don't make myself understood) glued together with vibration damping glue.
The driver cutouts that Lynn O suggests should be easy to do. Maybe the Scan-Speak would like its own chamber?
Some bracing could be needed also, I guess. But perhaps it´s best to try and find the "nodes" when the boxes are built?
Alrightie!
Port moved further down, and lengt kept as it was. The large radii weren´t quite right so they are redrawn.
I added a top view and made another drawing with some colours and shades to get a better idea of the appearance. I thought perhaps it would be fun to try putting felt around the drivers. Or leather, or something. This is the light grey area you see in the coloured/shaded part.
This is drawn assuming 21mm birch ply for the side + bottom panels and 40mm spruce, maybe made with solid wood "square profiles" (sorry, I am swedish and lack some proper english/american words regarding wood working, just ask me to explain and/or correct me if I don't make myself understood) glued together with vibration damping glue.
The driver cutouts that Lynn O suggests should be easy to do. Maybe the Scan-Speak would like its own chamber?
Some bracing could be needed also, I guess. But perhaps it´s best to try and find the "nodes" when the boxes are built?
Attachments
Hallo,
here you can see my Ariel speakers
http://www.majerovi.com/richard/Ariel/Ariel.htm
I used them with Mosfet amplifier. I was not satisfy with their sound and for two years they stey in my bedroom :-(
here you can see my Ariel speakers
http://www.majerovi.com/richard/Ariel/Ariel.htm
I used them with Mosfet amplifier. I was not satisfy with their sound and for two years they stey in my bedroom :-(
Well, In the past I had Dynaudio Focus kit speakers. Than I had chance to hear Dynaudio Gemini. It was "fantastic" for me and project Ariel is using similar size of speakers and D'Appolito construction. I was mad and I concentrated all my resources to build them. (mean Ariel)
What I don't like? It is difficult for me to explain in English. Today I have Dynaudio Contour 2,8. Esotar is doing much better job than SS. Ariel's sound is big. But on other side, something is completely missing. :-( Mainly lowest bass frequency.
What I don't like? It is difficult for me to explain in English. Today I have Dynaudio Contour 2,8. Esotar is doing much better job than SS. Ariel's sound is big. But on other side, something is completely missing. :-( Mainly lowest bass frequency.
It is sad to hear, considering you did so much work on the boxes. And the filter parts aren't exactly for free...
Regarding bass, this hopefully won't be a problem as I plan to build dipole bass units - and if this doesn't work I´ll build TL subs.
My friend with the ME2:s has had fantastic results combining them with bass units (BR boxes) of his own design. I have good hope that Scotts MLTL alternative will have somewhat better bass and lower reaching bass. We´ll see! I will keep you informed of any progress in this thread.
Regarding bass, this hopefully won't be a problem as I plan to build dipole bass units - and if this doesn't work I´ll build TL subs.
My friend with the ME2:s has had fantastic results combining them with bass units (BR boxes) of his own design. I have good hope that Scotts MLTL alternative will have somewhat better bass and lower reaching bass. We´ll see! I will keep you informed of any progress in this thread.
Well, the bass should hold up to 50Hz better than the originals, and the overall response should also be at least as good, and probably flatter to 1Khz when the enclosure ceases to influence things, other than providing a solid mounting point for the drivers.
That said, Lynn does state on his site that he was not aiming for low bass with these enclosures. Nor am I -these are really midrange drivers, not midbass, and you can't squeeze that much out of them, whatever you do: laws of physics and all that. Oh yes -one adjustment: try stuffing the upper 2/3 of the cabinet first: reduce if necessary. I don't think giving the tweeter its own chamber will improve matters.
Best
Scott
That said, Lynn does state on his site that he was not aiming for low bass with these enclosures. Nor am I -these are really midrange drivers, not midbass, and you can't squeeze that much out of them, whatever you do: laws of physics and all that. Oh yes -one adjustment: try stuffing the upper 2/3 of the cabinet first: reduce if necessary. I don't think giving the tweeter its own chamber will improve matters.
Best
Scott
Perhaps this time I will get an audio project finished.
Found some wood to use: 21mm birch ply for the side, bottom and back panels. For the front baffle I found some spruce, a kind of construction spruce "board" made from solid wood profiles glued together. This particular "board" is sometimes used instead of concrete or steel for girders when building houses and stuff. Don't know what the english word is. Will have both spruce and birch cut to size this week.
Dacron - well, to be honest I have longhair wool... is that OK scottmoose? - speakon contacts, wire and drivers will all be in the house by the end of the week.
If I am really, really lucky, the boxes will be glued and ready by next weekend or the week after that. With three small kids to take care of, you never know. Wish me luck!
Found some wood to use: 21mm birch ply for the side, bottom and back panels. For the front baffle I found some spruce, a kind of construction spruce "board" made from solid wood profiles glued together. This particular "board" is sometimes used instead of concrete or steel for girders when building houses and stuff. Don't know what the english word is. Will have both spruce and birch cut to size this week.
Dacron - well, to be honest I have longhair wool... is that OK scottmoose? - speakon contacts, wire and drivers will all be in the house by the end of the week.
If I am really, really lucky, the boxes will be glued and ready by next weekend or the week after that. With three small kids to take care of, you never know. Wish me luck!
Good luck, and do let us know how you get on -you are embarking upon the first Ariel MLTL built, and your experiences will be invaluable. Does that sound ominous? Don't worry, it should be an absolute gem once competed, according to the worksheets, and I long ago learned to trust them. Your material choices also sound excellent.
No idea about the spruce stuff you're using for th efront panel, other than to remark that spruce is actually very well regarded by those in the know. One question out of interest -the profiles that make it up; do they all go in the same direction or in different directions? My full-time profession is a naval historian, and I've done some research upon the armour used in early ironclads (which often had up to 24" of solid wood backing the iron armour.) One of the interesting things I found is that the backing was always more resiliant when the grains travelled in the same direction. Not much in it, but could be of use. Don't know why -I'd have to ask a woodworker.
Long hair wool is fine -I suggested Dacron simply because it's cheaper to buy, and there's nothing in it, in terms of sound quality. What wire are you going to use?
Best
Scott
No idea about the spruce stuff you're using for th efront panel, other than to remark that spruce is actually very well regarded by those in the know. One question out of interest -the profiles that make it up; do they all go in the same direction or in different directions? My full-time profession is a naval historian, and I've done some research upon the armour used in early ironclads (which often had up to 24" of solid wood backing the iron armour.) One of the interesting things I found is that the backing was always more resiliant when the grains travelled in the same direction. Not much in it, but could be of use. Don't know why -I'd have to ask a woodworker.
Long hair wool is fine -I suggested Dacron simply because it's cheaper to buy, and there's nothing in it, in terms of sound quality. What wire are you going to use?
Best
Scott