Still just your subjective opinion.
Another case of "builder´s pride"? 😉
Edit: answer applies to post #39
Another case of "builder´s pride"? 😉
Edit: answer applies to post #39
The directivity of the ROAR is a problem for me. That is nothing to be proud of. It makes it quite difficult to use in a normal livingroom. Why would I "brag" about that?
I can "create" any measurements I want to. How can anyone believe what I claim based on what I post, since there is no independent observer that can verify my measurements??
Since the stated directivity of the ROAR defies the laws of physics then I recall my claim. The ROAR does not have a narrow directivity.
Can we go back to the original discussion now?
Maybe that is inadvertently getting us closer one of the points, the measurements, we have so far concluded that bass reflex is economical and easy to get right, or right enough at least, when compared to the other more challenging design principles, and given that the products undergo testing and measurements in publications such as production partner (example only) I assume that one of the design and decisioncriterias is how well a design measures, which is correct as long as subjective listening is applied in equal meassure.
So instead of spending 2000 man hours on R&D getting an eight order qw design in shape you can reach a similar result in 500 man hours getting a BR right, similar that is to cost of twice or even four times as many drivers and twice the number of enclosures to reach the same output capability, so what does the economics really look like, is driver and material cost not that big of a deal for the big names in the industry (big volumes = lower product cost)?
So instead of spending 2000 man hours on R&D getting an eight order qw design in shape you can reach a similar result in 500 man hours getting a BR right, similar that is to cost of twice or even four times as many drivers and twice the number of enclosures to reach the same output capability, so what does the economics really look like, is driver and material cost not that big of a deal for the big names in the industry (big volumes = lower product cost)?
Your measurements show 45"x26.5" TH has better directivity than 26"x22.5" FLH - but they are not "similar dimensions". How about 45" wide FLH?
All radiation from FLH is in phase in all operation range, too.
I will believe, please share with us your outdoor measurements of ROAR directivity.
In general, what matters in directivity, is the frequency vs. frontal and radiating area of the enclosure. Im not so sure how the actual design affects the directivity.. maybe there is something more in it I dont know. I'm just thinking how sound waves behave. Something to do with pressure/velocity modes of the sound waves?
Lets think about two different subs with 100% same frontal area. First one is FLH and the second one is huge cone driver in a sealed enclosure. Which one is more directional and why? Howabout if you put a large passive radiator in front of FLH, which fills the whole mouth, it suddenly becomes omnidirectional? Proper FLH is very very efficient tho, compared to bass reflex.
Even closed subwoofer enclosures can be directional down low, if there is enough combined cone area.
Same acoustic laws affect all sound frequencies, so one can think how midrange or tweeters are behaving in different setups. Low frequency speakers are just scaled up versions.
Lets think about two different subs with 100% same frontal area. First one is FLH and the second one is huge cone driver in a sealed enclosure. Which one is more directional and why? Howabout if you put a large passive radiator in front of FLH, which fills the whole mouth, it suddenly becomes omnidirectional? Proper FLH is very very efficient tho, compared to bass reflex.
Even closed subwoofer enclosures can be directional down low, if there is enough combined cone area.
Same acoustic laws affect all sound frequencies, so one can think how midrange or tweeters are behaving in different setups. Low frequency speakers are just scaled up versions.
Last edited:
I can't comment on the directive because I didn't measure it but I can comment on the SQ. I have build and compared basreflex, 6th order bandpass, a THAM style tapped and a ROAR style cab. Get the cabs to combine with tops isn't that difficult anymore with modern DSP, but the difference in sound is very obvious. 3th spot combined: bassreflex and the bp6, the bp6 is louder but the sound isn't that different. 2th spot THAM, very musical sounding, I would say you can use it without a problem in a living room. It sounds 'easier' an more natural to my ears then bassreflex and or BP. The winner and with a distance is the ROAR, I think the description given by Circlomanen to describe how it sounds is very accurate. For all you none believers, build and listen to it and then come back here. You all really don't know what you are talking about because you've never listened to the design. Thanks Martinsson and Circlomanen to share these wonderful designs!
Mynningsbeteende ROAR12 - YouTube
Here is an quite interesting video of the mouth of a ROAR at work.
From about 4 seconds in, the video is a slow motion burst (480 pictures a second but no sound).
It is easy to see that the sound pressure (towel amplitude) is larger along the edges of the front resonator. The signal is a friend playing bass guitar through a bass amp coupled to the ROAR12. It is quite rich in harmonics - hence all the fun patterns in the towel.
He could not use the ROAR indoors since everything rattled and buzzed almost regardless of sound pressure level. We discussed what he was experiencing with the ROAR and he made this short video to show the mouth behaviour.
The towel has about 3 cm amplitude at most. This is with a 100 watt amp and no high pass filter.
This shows me that the ROAR is a plane wave device. A plane wavefront will have a very different dispersion pattern once it leaves the straight non-expanding resonator compared to the typical spherical horn mouth bubble from a normal expanding horn.
Thanks for building and sharing your results and impressions from our designs! It is a very valuable feedback! 🙂
Here is an quite interesting video of the mouth of a ROAR at work.
From about 4 seconds in, the video is a slow motion burst (480 pictures a second but no sound).
It is easy to see that the sound pressure (towel amplitude) is larger along the edges of the front resonator. The signal is a friend playing bass guitar through a bass amp coupled to the ROAR12. It is quite rich in harmonics - hence all the fun patterns in the towel.
He could not use the ROAR indoors since everything rattled and buzzed almost regardless of sound pressure level. We discussed what he was experiencing with the ROAR and he made this short video to show the mouth behaviour.
The towel has about 3 cm amplitude at most. This is with a 100 watt amp and no high pass filter.
This shows me that the ROAR is a plane wave device. A plane wavefront will have a very different dispersion pattern once it leaves the straight non-expanding resonator compared to the typical spherical horn mouth bubble from a normal expanding horn.
Thanks for building and sharing your results and impressions from our designs! It is a very valuable feedback! 🙂
Some day I will give a try for Roar. I have many reflex, tapped horns and true bass horns but ROAR/other paraflex designs is new to me. Very interesting design indeed.
So far, my favourite true subwooferhas been HOG scoop when it comes to bass reproduction under 60Hz. Scary things when cranked up, they always makes me smile while listening to! But indeed they are big and heavy and I hate moving them.
Tham 15, I like them too being such a compact design and giving good amount of bass beetween 50-100Hz range.
Yet the bass reflex is the best compromise beetween spl, sq and size. I have modified version of RCF 1018 enclosures and I have had lots of fun with them.
So far, my favourite true subwooferhas been HOG scoop when it comes to bass reproduction under 60Hz. Scary things when cranked up, they always makes me smile while listening to! But indeed they are big and heavy and I hate moving them.
Tham 15, I like them too being such a compact design and giving good amount of bass beetween 50-100Hz range.
Yet the bass reflex is the best compromise beetween spl, sq and size. I have modified version of RCF 1018 enclosures and I have had lots of fun with them.
Ways to increase directivity....increase the size of the horn mouth...increase depth of horn throat...and profile shape?....when I tried to sim my PPSL in one of those particle simulator things....I got increased "directivity" (more like directivity anomalies) with a 922 sq cm slot...increasing depth and or CSA would increase directivity further but for some reason I doubted its significance in playback....as significant directivity creeps in with increase of horn size....it would do so from the top side of the pass band....so if anyone had a formula to factor in horn depth to directivity it would be GM but I think I've got it allllll figured out (lol!)....just like baffle width looses directivity near a certain point....throat depth = D.......
(13550/D) / 3 (or 4?) should give us some information (same with width/height)....Directivity/Polar is just a study of velocity per area.....velocity is tied to SPL (and phase) so where ever you see an increase in spl in the passband there is an increase in velocity (per area) somewhere in the system....thinking in a more tangle way....where the system falls off (cutoff) one can assume that velocity (per area) has fallen off....so the resonate note of the line in a ROAR is where its directivity falls off.....we have several resonate notes (width/ height / depth / circumference?) and the polar is an accumulation of those notes....you could get complicated and talk about harmonics even....the roar probably has directivity fall of around 130hz is my guess.....but the answer is not unfathomable if one cared to know.....and you don't have to model in AKABAK to get the gist.
(13550/D) / 3 (or 4?) should give us some information (same with width/height)....Directivity/Polar is just a study of velocity per area.....velocity is tied to SPL (and phase) so where ever you see an increase in spl in the passband there is an increase in velocity (per area) somewhere in the system....thinking in a more tangle way....where the system falls off (cutoff) one can assume that velocity (per area) has fallen off....so the resonate note of the line in a ROAR is where its directivity falls off.....we have several resonate notes (width/ height / depth / circumference?) and the polar is an accumulation of those notes....you could get complicated and talk about harmonics even....the roar probably has directivity fall of around 130hz is my guess.....but the answer is not unfathomable if one cared to know.....and you don't have to model in AKABAK to get the gist.
Last edited:
I said that based off of seeing one random ROAR picture in this thread...its literally tied to the dimensions.
Hey martinsson, one thing i'd like to say whether it fits this thread or not ...
thanks a lot for your blog and all the detailed info it provides...looked at some pages there for first time....nice work and very generous sharing! 🙂
thanks a lot for your blog and all the detailed info it provides...looked at some pages there for first time....nice work and very generous sharing! 🙂
Thank you Mark100, I have not been very active lately though, my two kids (2,5 and 1 years) old keeps me busy when I'm not working, but eventually I will get back to building stuff and posting odd stuff 🙂
Regarding directivity I understand that the pressure distribution in the aperture area is important as well, and I assume that a long-ish (scientific term) constant area segment prior to the aperture could potentially be beneficial in equalizing the pressure distribution in the area prior to the aperture.
I assume any directivity, regardless of where in the range it happens, is beneficial,how does directivity work in the case of BR?, do they couple together according to the wavelengths they reproduce or how does that work?
I assume any directivity, regardless of where in the range it happens, is beneficial,how does directivity work in the case of BR?, do they couple together according to the wavelengths they reproduce or how does that work?
Good question, especially at lower frequencies where most the spl is coming from the ports.
For bare cone drivers, more cones coupled closely togerher means ofcourse narrower dispersion at given frequency range as the total cone area increases.
As we know how vertical line array sources are behaving.. they have quite narrow vertical dispersion but wide horizonal dispersion. Scaled up version does the same for bass frequencies.
But yes. there are many more factors. I'm just trying to think the high/low perssure wave front excecuted by cone movement and how it behaves in the surrounding air. Low level physics 🙂
For bare cone drivers, more cones coupled closely togerher means ofcourse narrower dispersion at given frequency range as the total cone area increases.
As we know how vertical line array sources are behaving.. they have quite narrow vertical dispersion but wide horizonal dispersion. Scaled up version does the same for bass frequencies.
But yes. there are many more factors. I'm just trying to think the high/low perssure wave front excecuted by cone movement and how it behaves in the surrounding air. Low level physics 🙂
I tend to regard a typical 218 box as having three drivers, one is the port, dominating around fb, and above that it is the actual drivers them selves, very different emission sources with regards to their characteristics, and to make things even more interesting their behaviour also changes with output level.
Unwrapped phase in ROAR vs paraflex is interesting(classic paraflex, not alternate). i have both, and for the same driver so in designing them i see (or get a feel for)clear ‘trends’ in how they are incredibly different. How the wave must not be tossed out but instead pushed while held slightly(?) like teaching a kid too to ride a bike.
Thenc, paraflex must be forced into a ~60 degree (impedAnce phase exit?) that it obviously must be coerced into while roar can happily be seperated into folds and shapes if you use the harmonic intervals as any qw pipe(offset driver for example). Series vs parallel is very interesting, but both is totally another beast to tame. Its really a challenge. Paraflex is complicated and the effort pays off if you want it.
Never met anyone who didnt like a THAM. thats a household name, synonymous with ‘punch’!
No doubt about the sound. i havent loaded the ‘roar’ with the 12” drivers yet, but if its anything like parAflex itll be hell for plaster, and kitchen utensils, garage, tool boxes, etc!! Tham with a turbo of extended hz and zing! These things are a cluster bomb of harmonics and from intervals and angles that sure are unique. too bad they cant get the attention of the people who would enjoy them.
Sad, the ‘public’ will never know what they are missing. Those are unique speaker designs and ‘fUN/Exciting’ is an understatement 🙁
Last edited:
Does anyone notice the ‘trends’ occurring inside these boxes very much exist outside of them? example in (lengths) harmonics and also if designing from scratch, fold at these intervals and proportions? Even in sim i keep hitting these marks and they are then ongoing and excellent places to put a null? Or to agressivekly tune with as bias, low, high or even a midband bump/boost or rise. And most of all, if used within a sim to create more cabinet without loosing the wave form in segmented shapes(CSA)... (beyond the basic stuff)? My impedances dont lie to me? somethings working here if we notice. But is this chase/push/pull of harmonics and pressure nulls ‘forward’ creating the ‘sound’ signature we notice yet have to insist upon others who are skeptical, until they hear it? I dunno, but the sound is also easily flawed. And thats issue that can create apprehension or biased opinions against anything but ‘Simple’?
a Huge amount of thanks and respect for THAM, ROAR, HROAR and Pataflex. You people have created very special things that create smiles and laughter, excitement and impactful statements. its different. its not perfect’ but its definitely amazing and a lot of fun to be able to experience . Once bitten, its like a ‘drug’? Its addictive. I doubt ill ever have a bass reflex (by design) again, (except) it could be an MLTL for the very bottom, hard to reach 20 hz or so...
a Huge amount of thanks and respect for THAM, ROAR, HROAR and Pataflex. You people have created very special things that create smiles and laughter, excitement and impactful statements. its different. its not perfect’ but its definitely amazing and a lot of fun to be able to experience . Once bitten, its like a ‘drug’? Its addictive. I doubt ill ever have a bass reflex (by design) again, (except) it could be an MLTL for the very bottom, hard to reach 20 hz or so...
Attachments
Last edited:
Thanks for all the kind words, and upon hear both that THAM back in 2009, and more recently ROAR I remember asking myself if has anybody else in the DIY community tried this? And if so why have I not seen more of this?
I find it hard to believe that the brilliant minds over at EAW, Kenton Forsythe for instance, or the corresponding R&D crews at JBL, NEXO, l'acoutics, Turbosound,TWA, etc. has not tried all of this, and then decided against it, this is what puzzles me, why did they not take it further? I believe we have some of the answers here in this thread, but perhaps not all.
I find it hard to believe that the brilliant minds over at EAW, Kenton Forsythe for instance, or the corresponding R&D crews at JBL, NEXO, l'acoutics, Turbosound,TWA, etc. has not tried all of this, and then decided against it, this is what puzzles me, why did they not take it further? I believe we have some of the answers here in this thread, but perhaps not all.
-Maybe- If we look at the ‘bad’ we can deduct what would need to be resolved in order to allow a ‘vent’ to do it all and no ‘direct radiator’ putting in effort regardless?
but in order to consider ‘bad’ we must create it. Nobody tries to create ‘bad’ so-its overlooked, never, or rarely experienced, and certainly overshadowed by the typical response and observed results in typical testing.
A paraflex will stir up room modes like no other qw pipe(TL) i have ever experienced. As will a ROAR, but in another way. If you sim or RTA and measure a paraflex impedance plot, you find absolute artifacts of everything. Each fold and chamber is there, and put out in the room its stuck and can be for awhile ring, decay, harmonic, etc. these all have a great potential for inversely squared bad sounds in logarithmically smaller sizes.. (bad math but the issue renaibs if it exists). it does. But we have addresed it in many ways. it needs helmholtz and qw absorbers to fix its own harmonics. each fold is a standing wave created, but its also an opportunity. One that can exist at a pressure max for a harmonic. easily shown in many layouts of 3 folds(4th order or tapped entry at exit), or 4/5 folds as a conmpounded paraflex or roar.
In horn response we can create this. just get rid of a flare rate and make 3 segments of 80 cm. Offset driver TL. or use TH as 120/120/80 and 0.01cm segments.
The same can be said for a 0.01cm offset paraflex. and a ROAR...or any compound horn. And if folded as such the HARmonics are wiped out tremendously, not just as seen in a HR sim which doesnt show the ‘folds’.
folding a ROAR at or creating a THAM which folds at phase 90/270, or 180/-180 for each length of the existing higher than Fb @320cm, 240,120 and 80,40,20 is ‘almost’ how they seem to exist by design? paraflex too.. the offset or flare rate used is very confusing so its all clouded. But RTA and impedance sweeps are telling us where we ‘should’ have folded. (We, as in myself, and not the original artists who created the original design).
If ‘we’ can support a clean and easy to create yet ferociously powerful delivery of sounds without the added issues we may or may not want to admit can and do exist if not looked at, maybe the world stops to see what there is in Arizona, Sweden and not just ‘bass reflex toots from vents?
i might be alone in thinking any of this, but fact check me, if im mistaken, then i should be held accountable and correct my thinking. Test the idea on known cabinets or altered in room modes or designs to exaggerate it to observe it? then if recognized, it can be fixed. (Please note the pic. Intentionally designed with matching mini chambers(rear stuffed center folds) as absorbers with harmonic intervals of folded ROAR/Paraflex layout. It works, it wasnt even needed. it could have just been folded correctly....because the folding lengths corrected its impedance landmarks. the result was a flat and phase aligned RTA(much improved and currently getting beta versions to shift to the 120/120/80/40 or 80/80/80 segment lengths instead of the scattered variety of this one).
