• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

ASR Directiva Open Source Speaker Review

Spocko

Major Contributor
Forum Donor
Joined
Sep 27, 2019
Messages
1,621
Likes
3,001
Location
Southern California
the directivity would be messier, and the maximum output would be lower. Getting a flat listening window would be trivial once you have an NFS and you're making an active design.
So it sounds like expensive drivers are unavoidable when it comes to certain minimum performance metrics after all?
 

ctrl

Major Contributor
Forum Donor
Joined
Jan 24, 2020
Messages
1,640
Likes
6,279
Location
.de, DE, DEU
I did a quick sim, and it looks like it's just about an okay fit. MaxSPL is limited though, so I can imagine that Amir heard some ominous noises when playing loud. Here is the MaxSPL curve with the PR, I uses about 20 liters here, so real numbers might be a bit different:
That about fits, but only if a high pass is used to protect against too much excursion at low frequencies.

The Purifi woofer has a linear excursion of +-10mm and the mechanical limit is +-14mm.
The passive radiator from SBAcoustics has a mechanical limit of +-11mm.

This fits quite well, because at frequencies below 50Hz the PR reaches the limit of maximum mechanical deflection before the woofer and thus protects the expensive woofer (by making itself heard with "agonizing" noises).

Max bass SPL - with HP BW fourth order at 30Hz
1633365171010.png
The simulation refers to the half-space (2pi), for the full-space we have to subtract 6dB.
This results in a maxSPL@50Hz of about 95dB free-field.

Max bass SPL - with HP BW fourth order at 40Hz
1633365200806.png
As above, we convert maxSPL to full-space and get about 98dB maxSPL@50Hz free-field.

As I said, this is about what two "normal" 6-6.5'' drivers can do at maxSPL.
But of course the limits set by physics apply here as well.

For such a small loudspeaker the result is very good and with some support of the listening room's boundary surfaces you can add another 3-8dB depending on the speaker position.


Im not saying it's not a successful design, I guess what I'm trying to say is that I'm disappointed Amir didn't have any particular reaction to the bass. I'm aware of what the purefi woofer can do on paper, but I wanted to read some personal impressions.
Well, the speaker's low bass behavior was tuned rather neutrally. Nothing should stand out there.
In addition, the final fine-tuning is still missing. This usually changes the impression in the low bass range too.
 

Spkrdctr

Major Contributor
Joined
Apr 22, 2021
Messages
2,230
Likes
2,957
Exchange Purify for Satori WM16P. That should probably do it.

I'm talking about a commercially available speaker pair. How much does it cost to get the same performance? What would those speakers be? Anyone got any great ideas?
 

voodooless

Grand Contributor
Forum Donor
Joined
Jun 16, 2020
Messages
10,432
Likes
18,441
Location
Netherlands
The Purifi woofer has a linear excursion of +-10mm and the mechanical limit is +-14mm.
The passive radiator from SBAcoustics has a mechanical limit of +-11mm.
Ah, I used the 6 mm Xmax version for my sim. That would mean a bit more output indeed, but still a bit of a compromise in the end. A good one though I think. In the end, you want a small design with a relatively big low end, which this will give. Add a few subs, and this combination will do massive dynamics if needed.
 

H-713

Senior Member
Joined
Jan 24, 2021
Messages
379
Likes
668
Well done guys, very impressive. Pretty much blows every other 6.5" 2-way out of the water.

One thing that I think might be really worthwhile, however, is to do this same process, but use a more economical woofer. Nothing really wrong with the tweeters, but the use of $370 woofers limits the market for this to some degree. It might be interesting to explore a similar project using a Seas or SB Acoustics woofer, sacrificing some bass distortion performance for cost. Perhaps the ER18RNX or the equivalent size SB / Satori woofer.

The Purifi woofers are probably the best on the market... but they're still really, really expensive for a 6.5" driver.
 

eboleyn

Member
Forum Donor
Joined
Sep 28, 2019
Messages
93
Likes
99
Location
Portland/OR/USA
That about fits, but only if a high pass is used to protect against too much excursion at low frequencies.

The Purifi woofer has a linear excursion of +-10mm and the mechanical limit is +-14mm.
The passive radiator from SBAcoustics has a mechanical limit of +-11mm.

This fits quite well, because at frequencies below 50Hz the PR reaches the limit of maximum mechanical deflection before the woofer and thus protects the expensive woofer (by making itself heard with "agonizing" noises).

...

Well, the speaker's low bass behavior was tuned rather neutrally. Nothing should stand out there.
In addition, the final fine-tuning is still missing. This usually changes the impression in the low bass range too.

As mentioned in a previous reply not long before this, I'd built a similar box and been running it for 8+ months now using the same Purifi 6.5" mid/woofer. Mine used a sealed design so it has somewhat less bass volume but had at least some response extending even lower.

I saw the numbers even before it was built and assumed that it would need a real subwoofer to get excellent low bass and sub-bass, and as the saying goes, I was Not Wrong. I have a 15" Rythmik sealed sub that is in my setup, and the whole stack sounds Glorious.

Frankly, even with EQ, unless you have a dedicated driver for the very low frequencies of less than 20 Hz up to about 100 Hz, you'll lose some of those low frequencies even if you don't miss them due to not being used to getting them.

From looking at the available high-quality drivers and the levels of distortion you get in each range, I wasn't able to make a formula which would work across the whole human hearing/experience range of < 20 Hz (Note I say this because we can perceive less than 20 Hz, just not as directly with our ears) up to ~20kHz without using at least 3 drivers:
  • < 20 - ~60-120Hz
  • ~60-120Hz - ~1-2kHz
  • ~1-2kHz - > 20kHz
Of course the advantage of using ~80Hz or less for crossover of the lower bass/sub-bass is that you can use a single subwoofer vs. needing 2 units in stereo.

EDIT: I'm not talking about "full-range" drivers here of course, but that's a wholly separate discussion...
 
Last edited:

Robbo99999

Master Contributor
Forum Donor
Joined
Jan 23, 2020
Messages
7,024
Likes
6,885
Location
UK
I'm talking about a commercially available speaker pair. How much does it cost to get the same performance? What would those speakers be? Anyone got any great ideas?
You could buy two JBL 308p Mkii speakers (£300 for a pair when on sale) and use my Anechoic EQ, I think it might have better directivity even but not sure, following is a link to my post showing the effect of my anechoic EQ on the Listening Window & On-Axis, along with directivity plots from Amir:
Distortion of the JBL 308p Mkii is a lot higher though as seen in following pics, but if you don't listen at loud levels then I think it's ok and when I measured my speakers in my room I got lower distortion than that pictured here.
JBL 308p distortion.png
EDIT: updated this post with my own distortion measurements in my room of my 308p speakers after anechoic EQ:
74dB for 2 speakers at 2 meters (my normal listening level):
JBL 308p distortion 74dB 2m 2speakers (dB).jpgJBL 308p distortion 74dB 2m 2speakers (percent).jpg
80dB for 2 speakers at 2 meters (my max listening level):
JBL 308p distortion 80dB 2m 2speakers (dB).jpgJBL 308p distortion 80dB 2m 2speakers (percent).jpg
Other louder measurements up to 88dB:
JBL 308p distortion 83dB 2m 2speakers (dB).jpgJBL 308p distortion 83dB 2m 2speakers (percent).jpgJBL 308p distortion 86dB 2m 2speakers (dB).jpgJBL 308p distortion 86dB 2m 2speakers (percent).jpgJBL 308p distortion 88dB 2m 2speakers (dB).jpgJBL 308p distortion 88dB 2m 2speakers (percent).jpg
 
Last edited:

PeteL

Major Contributor
Joined
Jun 1, 2020
Messages
3,303
Likes
3,849
I'm talking about a commercially available speaker pair. How much does it cost to get the same performance? What would those speakers be? Anyone got any great ideas?
It's mention in the review it'd be in the 4-5k range,
 

H-713

Senior Member
Joined
Jan 24, 2021
Messages
379
Likes
668
So it sounds like expensive drivers are unavoidable when it comes to certain minimum performance metrics after all?
Not entirely. Revel uses SB Acoustics drivers (~$70) for their woofers, and a tweeter that costs about $50. I'm sure they're getting them a bit cheaper than that too. You lose some performance dropping down from the Purifi, but you do NOT need a $370 woofer to get pretty great performance.
 

thewas

Master Contributor
Forum Donor
Joined
Jan 15, 2020
Messages
6,907
Likes
16,974

Spkrdctr

Major Contributor
Joined
Apr 22, 2021
Messages
2,230
Likes
2,957
You could buy two JBL 308p Mkii speakers (£300 for a pair when on sale) and use my Anechoic EQ, I think it might have better directivity even but not sure, following is a link to my post showing the effect of my anechoic EQ on the Listening Window & On-Axis, along with directivity plots from Amir:
Distortion of the JBL 308p Mkii is a lot higher though as seen in following pics, but if you don't listen at loud levels then I think it's ok and when I measured my speakers in my room I got lower distortion than that pictured here.
View attachment 157153
Thank you!
 

Rick Sykora

Major Contributor
Forum Donor
Joined
Jan 14, 2020
Messages
3,634
Likes
7,381
Location
Stow, Ohio USA

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,441
Likes
5,399
Location
Somerville, MA
Not entirely. Revel uses SB Acoustics drivers (~$70) for their woofers, and a tweeter that costs about $50. I'm sure they're getting them a bit cheaper than that too. You lose some performance dropping down from the Purifi, but you do NOT need a $370 woofer to get pretty great performance.
This is kind of my point. You're paying a lot for miniaturization. For me, it makes more sense to just use subwoofers, or build a larger speaker (traditionally this is where DIY becomes a good value)

After reviewing the measurements more carefully, I think I'm actually impressed by the DI more than anything. The DXT is doing a fantastic job.

Can someone put the bass performance in some kind of reference for me? Better than a Revel f206? Better than a 708p? If it beats either of those, that is a huge win for a speaker of this size.

Also, what is the limiting transducer here? What runs out of juice first, the tweeter (a conventional dome in an odd waveguide), the woofer (a monster) or the PR (which doesn't strike me as that big)?

I'd love to hear some comparitive descriptions of this speaker as it really is a unique design. Is it atmospheric or very focussed? Does the low bass distortion make it sound big or small? Etc.
 

temps

Active Member
Joined
Mar 9, 2020
Messages
199
Likes
347
feel like I'm taking crazy pills

"Genelec 8040s" and "cheaper woofer" throughout the thread... does nobody look at the frequency response? This thing's LF extension is absolutely exceptional, and that is largely because of the expensive woofer. The distortion measurements are absolutely spectacular. It is +/- 2dB, starting at the exact frequency where the 8040 is down 6dB!

These ideas to replace the woofer with a cheaper one to get the BOM down... okay, I guess??? It becomes a completely normal, unexceptional speaker in that case. Dozens if not hundreds of those designs have already been done.

If anything, I want to see this same design with the newer, presumably even more expensive, Purifi 6.5 woofer that has even more excursion... the only cheaper alternative woofer mentioned in the thread so far that stands any hope of maintaining this level of performance would be the XBL^2 woofer from Misco.
 

617

Major Contributor
Forum Donor
Joined
Mar 18, 2019
Messages
2,441
Likes
5,399
Location
Somerville, MA
feel like I'm taking crazy pills

"Genelec 8040s" and "cheaper woofer" throughout the thread... does nobody look at the frequency response? This thing's LF extension is absolutely exceptional, and that is largely because of the expensive woofer. The distortion measurements are absolutely spectacular. It is +/- 2dB, starting at the exact frequency where the 8040 is down 6dB!

These ideas to replace the woofer with a cheaper one to get the BOM down... okay, I guess??? It becomes a completely normal, unexceptional speaker in that case. Dozens if not hundreds of those designs have already been done.

If anything, I want to see this same design with the newer, presumably even more expensive, Purifi 6.5 woofer that has even more excursion... the only cheaper alternative woofer mentioned in the thread so far that stands any hope of maintaining this level of performance would be the XBL^2 woofer from Misco.

I would argue that absolute output level and extension are more important than distortion. I'm not sure how this speaker performs in that respect; it seems like the PR and the tweeter would run out of steam before the purefi, but I await comments from the designer.

Let's not compare this to a Genelec. This is a DIY project, and it is a unique value proposition.
 

eboleyn

Member
Forum Donor
Joined
Sep 28, 2019
Messages
93
Likes
99
Location
Portland/OR/USA
feel like I'm taking crazy pills

"Genelec 8040s" and "cheaper woofer" throughout the thread... does nobody look at the frequency response? This thing's LF extension is absolutely exceptional, and that is largely because of the expensive woofer. The distortion measurements are absolutely spectacular. It is +/- 2dB, starting at the exact frequency where the 8040 is down 6dB!

These ideas to replace the woofer with a cheaper one to get the BOM down... okay, I guess??? It becomes a completely normal, unexceptional speaker in that case. Dozens if not hundreds of those designs have already been done.

If anything, I want to see this same design with the newer, presumably even more expensive, Purifi 6.5 woofer that has even more excursion... the only cheaper alternative woofer mentioned in the thread so far that stands any hope of maintaining this level of performance would be the XBL^2 woofer from Misco.

FYI, I direct-ordered Purifi's newest sampling X-series aluminum mid/woofer for the similar reason of "why not". My existing build of using Purifi's 6.5" X-series mid/woofer plus a Raal 70-20XR is likely already at or beyond the limit of anything I can detect with my ears, but I'm doing another build with their newer mid/woofer plus a full Raal 140-15D, which also has greater frequency extension and lower distortion.

Again, I doubt I will be able to tell the difference, but it's about doing it for Fun at this point anyway.
 

temps

Active Member
Joined
Mar 9, 2020
Messages
199
Likes
347
I would argue that absolute output level and extension are more important than distortion. I'm not sure how this speaker performs in that respect; it seems like the PR and the tweeter would run out of steam before the purefi, but I await comments from the designer.

Let's not compare this to a Genelec. This is a DIY project, and it is a unique value proposition.

I'm not the one comparing it to a Genelec - there are many posts in the thread doing exactly that, though, and I agree with you. They are completely off base.

To be honest I myself, don't particularly care about headroom once you get up past 90dB or so. It has more than enough as is. This looks like it'd make for an absolutely exceptional studio monitor, vastly superior to anything else in its price range or even well above. Has extension almost equivalent to my KH310s, in a much smaller cabinet, with better directivity - although the vertical notch is a bit of a bummer. Not sure how that factors in nearfield but a monitor that doesn't need a ceiling cloud would be great.
 
Top Bottom