Introducing Directiva - An ASR open source platform speaker project

[fluid]

So best practice would be to use a wide baffle or bigger midrange driver combined with a low frequency supercardioid to minimize problems? Can you do this without increasing the distance of the two (or more) sources?

I don't know about best practice per se it depends on the response and directivity you want to achieve.

A 15" woofer in the right shaped box has pretty good cardioid/supercardioid behaviour by itself and doesn't suffer loss of output in the same way a passive or active cardioid does, BEM directivity simulations below, at 400Hz the rear output is more than 15dB down and at 160 degrees it is over 20dB down. The D&D, Kii and the Directiva R2 have more even directivity over a wider bandwidth.

Did you get a lower R60 with all cardioids or dipoles and frequencies due to the cancelation at some points in the room if you linearize the frequency response below schroeder frequency at the listening position?

I don't have any in room cardioid measurements. This thread might help
https://www.audiosciencereview.com/...ity-compared-with-in-room-measurements.21508/

 

[TimVG]

Hi

The design goal wasn't really to improve boundary interference effects (but as always there can be happy side-effects), but rather to follow as close as we could the basic principle that the off-axis sounds should be spectrally similar to the direct sound, in our case down to the the transition region, and if you look at the radiation behaviour, we've succeeded very well in doing so.

So due dilligence is still needed when placing this speaker in a room. I happen to have a very small room, I can't position them any further than just below half a meter from the side walls, and this still works well!

 

[abdo123]

@ctrl

I have a question that often goes unoticed in crossover design, Genelecs offer a setting for people that place their speakers on a surface (a table or shelf .etc) that changes the response of the speaker in an interesting way.

Could you please explain how the radiation pattern of a speaker changes when it's on a surface instead of floating in the air (typical anechoic free-standing measurements)

Is it possible to design a filter / crossover for the R2 module to use when it's placed on a surface? Is it possible to accommodate for this ourselves with our own (different) speakers?

Thank you!

 

[TheBatsEar]

Is there some kind of Wiki that condenses this thread into meaningful instructions for the interested layman?

 

[Rick Sykora]

Is there some kind of Wiki that condenses this thread into meaningful instructions for the interested layman?

I am pondering with the admins on how we might reorganize. Not a lot of good options unfortunately...

What info are you seeking?

 

[TheBatsEar]

Something like this:

1. Directiva V1
   1. Abstract
   2. Design
      1. Active vs Passive
      2. BR vs passive woofer
      3. Enclosure
      4. Measurements and Graphs
   3. How to replicate
      1. BOM, sources, alternatives
      2. Instructions
   4. Results
      1. Interpretation of measurements
      2. Graphs
   5. Buy me some pizza here

Could be a Wiki, could be a page at Gitlab/Github, something where one can fix mistakes and do revisions and possibly reuse for other projects. I suspect the forum software used here has a wiki module that could be used.

I for one would use this to make a decision if i want to build my own, or if it's even possible to get the parts.

 

[Rick Sykora]

Something like this:

1. Directiva V1
   1. Abstract
   2. Design
      1. Active vs Passive
      2. BR vs passive woofer
      3. Enclosure
      4. Measurements and Graphs
   3. How to replicate
      1. BOM, sources, alternatives
      2. Instructions
   4. Results
      1. Interpretation of measurements
      2. Graphs
   5. Buy me some pizza here

Could be a Wiki, could be a page at Gitlab/Github, something where one can fix mistakes and do revisions and possibly reuse for other projects. I suspect the forum software used here has a wiki module that could be used.

I for one would use this to make a decision if i want to build my own, or if it's even possible to get the parts.

Got it. Thanks!

Should have some time for this after r2

 

[Jinjuku]

Been looking for another build project as it's been like 5 years since I did a DIY. Should I wait for the R2 or still give serious consideration to the R1?

I'm figuring with subs, amps etc that I'll be at least $4K when it's all said and done for the R1.

 

[Rick Sykora]

Been looking for another build project as it's been like 5 years since I did a DIY. Should I wait for the R2 or still give serious consideration to the R1?

I'm figuring with subs, amps etc that I'll be at least $4K when it's all said and done for the R1.

What are you planning to do with the new setup? r1 is more targeted towards a stereo music system for small to medium venues. r2 will have a lot more flexibility with regard to room size and application. r2 is also a considerable upgrade in directivity. Until I build a couple and compare, am not sure whether the improved directivity is a major improvement...

We have one study that says it is!

 

[Jinjuku]

What are you planning to do with the new setup?

Satisfy my curiosity Really 2.0. I love my JBL 308MK II with matching subs. Just that taken to the next level. I still pinch my self over what I was able to get for $1000 in that setup.

 

[TheBatsEar]

Should have some time for this after r2

Cheers

 

[Savi]

Ofc, the Purifi woofers were considered for r2 and mainly avoided due to cost. The team is still optimizing the r2 slot loading, but depending on results, may revisit r1 and try the slot loading on it. Or maybe the new Purifi 5.25 will be available, and we can try it in the r2 cabinet! Lots of possibilities and not enough time or money to try them all, so we are doing our best.

I do not want to derail the thread on other driver but I was wondering if there are specific parameter (VAS, Qts, Xmax, etc...) your are looking for when building a cardioid for a given driver size (around 6inch to stay in the thread) ? I know how these paramaters play in woofer aligment and play in vituixcad when needed but for cardioid I am lost.

 

[abdo123]

I do not want to derail the thread on other driver but I was wondering if there are specific parameter (VAS, Qts, Xmax, etc...) your are looking for when building a cardioid for a given driver size (around 6inch to stay in the thread) ? I know how these paramaters play in woofer aligment and play in vituixcad when needed but for cardioid I am lost.

From what i understood max output and sensitivity is the most limiting factor.

 

[Rick Sykora]

I do not want to derail the thread on other driver but I was wondering if there are specific parameter (VAS, Qts, Xmax, etc...) your are looking for when building a cardioid for a given driver size (around 6inch to stay in the thread) ? I know how these paramaters play in woofer aligment and play in vituixcad when needed but for cardioid I am lost.

Frankly, the selection of the SB15 was very much an educated guess. We wanted a driver that would allow good overlap with the Seas DXT and needed to be able to extend low enough to match up with the bass module. Also looked for a driver with a solid distortion profile that would hold up well when pushed harder.

We had a targeted 2 potential cabinet alignments: the passive resistance one (like D&D 8c) and a bass reflex as potential option for those who might want to use r2 as a more standalone monitor. Another key consideration was cost as many members had expressed concern over the expense of r1’s Purifi woofer. So, this drove us toward a woofer that was more suited for BR alignment in a ~7 liter cabinet. To get more low end extension, the bass reflex option was allowed to have less low-end directivity too. If both of these alignments could be supported by one driver, this would help speed along the r2 schedule.

As sims were done, we found the stock cabinet was too restrictive and so allowed the shape to change for better directivity. The shape change improves the higher frequency directivity and the depth helped for the lower frequency directivity. At this stage, we are still working on the passive resistance aligmnment and have not revisited the bass reflex design. Quite a few sims were done to optimize the passive resistance design and more since the first prototype was built. These sims were needed to refine the cardioid directivity since it is a less well-defined science.

@TimVG might have some additional references and tips to get started, but r2 had some additional requirements beyond just being a cardioid that may not apply. I recall some original guidance towards a woofer that works well in a bass reflex design, but as you can gather, is just a starting point.

 

[TimVG]

I don't believe the scope of R2 involved enough testing to draw any definite conclusions. Of course, in the case R2 more testing isn't needed because we reached what we were after pretty much from the first attempt, but since we're all hobbyists of course we're a bit limited in time and money

I can say however that excursion is not the 'limiting' factor of the midrange driver. The combination of slots and damping add output to the midrange, both destructive and constructive - I honestly believe a more expensive midrange driver would add nothing in terms of subjective experience, perhaps not even objectively. But => (insert the first sentence).

 

[ctrl]

I have a question that often goes unoticed in crossover design, Genelecs offer a setting for people that place their speakers on a surface (a table or shelf .etc) that changes the response of the speaker in an interesting way.

I don't know the exact use case Genelec is basing on (mixer size, listening distance, ear height,...) - can certainly be simulated or, much easier, just make a measurement.
At 160Hz, the wavelength is 2.1m, perhaps there are edge diffraction effects through the mixing console, which leads to an exaggeration in this frequency range?
Where are the studio engineers, they should know something like that

A 15" woofer in the right shaped box has pretty good cardioid/supercardioid behaviour by itself and doesn't suffer loss of output in the same way a passive or active cardioid does, BEM directivity simulations below, at 400Hz the rear output is more than 15dB down and at 160 degrees it is over 20dB down. The D&D, Kii and the Directiva R2 have more even directivity over a wider bandwidth.

You raise an interesting point. The radiation pattern below 1kHz of a D&D, Kii or even the Directiva r2 is quite similar to that of a very large floorstanding speaker.

Compared to a 15'' driver, the Directiva r2's radiation below 400Hz is even significantly better.
Here for comparison sonogram and polar diagram of the 15'' woofer in the large cabinet, from post#1101, with the Directiva r2 at identical scaling:

Perhaps the r2 can (partially) deliver the "sound characteristics" of a large floorstanding loudspeaker (with a corresponding bass module for r2), with a significantly smaller footprint.


In the forum there is the "Subwoofers make all big speakers obsolete?" thread. For "normal" small 2-way speakers plus SW, the question can be clearly answered in the negative, because the radiation will differ too much and therefore a comparable sound characteristic will never be achieved (ratio of direct and reflected sound).
With passive or active control of radiation, small speakers can rival large speakers in terms of radiation.

How close this comes to the listening experience of large speakers, if the difference in max SPL is significant, I can not say.

 

[thewas]

I don't know the exact use case Genelec is basing on (mixer size, listening distance, ear height,...) - can certainly be simulated or, much easier, just make a measurement.
At 160Hz, the wavelength is 2.1m, perhaps there are edge diffraction effects through the mixing console, which leads to an exaggeration in this frequency range?
Where are the studio engineers, they should know something like that

See here:

Adam T5V Review (Studio Monitor)

Here are my Adam T5's. Not the best setup acoustically. I set the LF to 0 dB for the left speaker near the wall and at +2 dB for the right speaker with no wall behind it. No frequency response measurement done yet.

www.audiosciencereview.com

 

[D!sco]

Hey technical people. I have a passive cardioid question. A broader explanation of the science behind why cardioid works is definitely in order eventually, but for now, a thought. I’d really like to design a three way bookshelf version of this if possible, and I’ve been contemplating how exactly to do that. Seeing Rick build the monitor reminds me vaguely of some of the more triangular Troels Gravesen designs, like the SBA-741. 3.8 degrees may be shallow for my taste here, but it begs the question, does passive cardioid work at an angle? The goal being minimal expansion to the baffle width, additional depth being added for woofer enclosure and venting. How much of a redesign would that require, and would it even be possible to do cardioid at an angle like this? Crossover and enclosure dimensions notwithstanding of course.

A napkin sketch from my tablet

 

[TimVG]

Yes, it will still work in an angle, although it must be said changing any aspect of the original design will lead to a change in performance. Simply making the same slots in a different enclosure could lead to a very different end result. Any change in volume, internal dimensions and even thickness affects performance.

My advice would be to build the top module, but add a very small bass module so that the end result looks like a standmount.

 

[muad]

Is there any way to calculate the minimum distance from side and rear wall, before the cardiod effect ceases to function? I have a very tight fitting space...