Sigberg Audio Saranna (fullrange, cardioid active floorstander) development thread

[poopy]

Mantas at Surround and Rear?

Thorbjorn has never shown what’s behind his sofa….
I know that Sigberg Audio has some stock B pairs of Mantas with reduced prices…. a good opportunity to buy a pair for the surrounds!

@sigbergaudio Have you tried Mantas in a L/C/R setup? Would it work? I believe the SBS1 could be used as surround/rear speakers.

 

[ryanosaur]

I believe the SBS1 could be used as surround/rear speakers.

Those are for your Atmos/Auro placements!

 

[sigbergaudio]

Thorbjorn has never shown what’s behind his sofa….
I know that Sigberg Audio has some stock B pairs of Mantas with reduced prices…. a good opportunity to buy a pair for the surrounds!

@sigbergaudio Have you tried Mantas in a L/C/R setup? Would it work? I believe the SBS1 could be used as surround/rear speakers.

I have not tried that, but I see no reason why it shouldn't work. I simply don't have the space to do something like that. I have also actually gone from 5.1 to 2.1 in my own main living room because the wide soundstage / imaging qualities of the speakers works so well with movies that I don't really feel I'm missing much. Sounds like a made up sales pitch I guess, but it's true. I have a friend with SBS.1 and dual Inkognitos here in Norway who did the same and had the same experience.

That being said, I encourage anyone to buy 5-7 SBS.1 or Mantas to test it out, I'm sure it's awesome! Having Mantas as LR or LCR and SBS.1 for the rest would likely work very well too.

 

[sigbergaudio]

On a related topic: One of my early memories with the SBS.1 prototype was watching "The accountant". That movie has a pretty well designed sound track and high dynamic range, and at least the version I watched was also mastered significantly louder than most movies. Early in the movie everything is pretty soft and low key, and I had the volume turned up very high because it was during testing of the prototype. Having never watched the movie either, the first time he fired a shot I both almost ducked behind the sofa because I believed *I* was the one taking fire, and also my ears literally hurt afterwards.

So even with the "entry level" SBS.1, the dynamic range and maximum SPL capacity is very well suited for movies.

 

[sigbergaudio]

Had a prolonged listening session today, and struggled to stop, just had to hear one more song..

I am going to Seas on Thursday this coming week for the first round of "proper" measurements. I suspect there's ever so slightly too much bass for a truly neutral speaker.. but it suits them so well! They sound so damn good right now on just about everything I throw on them, so I can't promise they won't stay that way! Luckily these are floorstanders and won't be marketed for the studio crowd.

 

[sigbergaudio]

Quick video showing excursion at relatively loud levels (averaging around 105dB@listening position). Healthy movement on the bass drivers, but perhaps more of note is the relative lack of moment of the coax, for those worried about distortion / problems related to what is essentially the waveguide of the tweeter moving back and forth.

The speaker is now essentially running in a 2.5-way configuration. The coax is allowed to roll off naturally, which it starts to do at around 200hz due to the small enclosure. To control the excursion there's also a 4th order high pass filter at 60hz.

 

[sigbergaudio]

And a video of the Seas anechoic chamber (we measured there today) for those who are curious. I suspect many will find it to be surprisingly large.

 

[sigbergaudio]

Distortion. Well under 1% at 96dB@1m basically everywhere except beyond 10khz and slightly over at 1500hz. The graph says 90dB at the Y-axis because this measurement is done at 2 meters.

 

[sigbergaudio]

And a few graphs. This is the first anechoic chamber visit for the Saranna, so these measurements are not final. They may get better, but they may also get worse, at least at any individual axis, since I prefer the sound of mild corrections combined with focus on the totality of the measurements rather than hammering on-axis flat. It also sounds like this speaker will sound most balanced when it's quite flat, so I may end up with slightly less bass than on these measurements. That can be customized by the customer as well of course.

This coax is somewhat untidy directly on-axis. Luckily on-axis will not be the reference/listening axis, and it also smooths off quickly off-axis.

Also note I had to measure the speaker upside down since the microphone didn't go high enough, and the bass port is then up in the air and on the rear, so the measurements don't seem to have captured the port output very well. Further, the chamber shows ~4dB less bass at 30hz than the Klippel NFS (thanks @AscendDF for that information). There's a measurement towards the end showing the bass measured from the front vs the rear. Simulations with VituixCAD indicate F6 of 26hz which is likely more accurate than these measurements.

All graphs are 1/24 smoothing, which I understand is what Amir typically use.

0 degrees:

15 degrees (listening/reference axis):

30 degrees:

60 degrees:

10 degrees vs 120 degrees to show an indication of the directivity

Bass measured from at 0deg and 180deg (Red), the latter showing more of the port output, with 7-8dB more at 20hz:

 

[sigbergaudio]

Have had a bit of time to analyze the findings at Seas now, and based on those I've made following adjustments:

Slightly decreased bass from 100-250hz
Slightly increased level from 4-8khz
Increased level above 14khz

I've also identified some resonances in the chamber for the coax, so that needs to be redesigned probably. In the meantime it has been reduced significantly with additional dampening inside the chamber.

The measurements below are from our own measurement corner that is reasonably accurate above 100hz. At around 6-10khz, Seas showed 2dB more, so the tendency of a dip in this area is probably less pronounced than it looks on these measurements. Orange is a measurement done on the same tuning that was used at Seas, and green is new.

This is at 10 degrees

 

[sigbergaudio]

"It looks like we have a resonance caused by the back wall."
"Hm. We could change the depth. Or we could remove it."
"Remove..remove what?
"The back wall?"
"Remove. The. Backwall."
"Nice! Let's go!"

The identified resonances in the output from the side ports have now been reduced significantly, and the port responses are in general much cleaner. The direct response also evened out a bit.

Nearfield measurement of one of the rearmost side ports before and after additional dampening + removing the backwall (blue is after):

Current 15 degrees (4-20khz was boosted by 2dB after this measurement)

 

[jhaider]

Interesting - so effectively the midrange is OB now?

Be curious to see how that affects the ability to keep close to the front wall.

 

[sigbergaudio]

Interesting - so effectively the midrange is OB now?

Be curious to see how that affects the ability to keep close to the front wall.

Yes it is.

I will see if I can do some in-room measurements soon.

 

[sigbergaudio]

Some in-room measurements after the latest modification of the enclosures. The speaker is now in practice open baffle from ~200hz and up. The response between 150-300hz obviously isn't ideal, but this is a room related dip, and it measures differently (no dip) in close vicinity to where this is measured, and doesn't sound as thin as it looks, so it has not been EQed when this was measured.

What I find interesting with these measurements are how tidy it looks from 3-400hz and up. This is probably some of the cleaner measurements I've done of any speakers in this room. There is zero EQ applied above 62hz in these measurements.

Left+Right+Average:

The bass section is still traditional / bass reflex, meaning we get a hybrid configuration where the bass is still full and tidy without having a massive speaker, while having an open baffle above 200hz.

In the measurement below there's two EQ points applied to one speaker (40 and 60hz), and none at all on the other.

20-200hz, Left+Right+Average, no smoothing: