Ascend Acoustics Horizon Center Speaker Review

[Dennis Murphy]

Amir's & Erin's plot scales plus ratios are seldom the same but in Amir share his raw spindata and it was possible trace Erin's spinorama plot in below we can have them compareable side by side and overlaid onto same scales and ratios ..
View attachment 92614

Those are pretty fancy pictures. I'm impressed. Thanks very much. I do wonder, however, whether the on'axis response should be highlighted so much. Given the near-impossibility--even with waveguides, and even more so with Coaxial configurations--to avoid on-axis diffraction effects, and given that for most people the first arrival will be the response at around 15 degrees off axis, I think it might be more informative to take the on-axis response out for at least one of the presentations so the listening window response is clearer. That's more problematic here, since one of the speakers is a center channel that will in fact be listened to directly on axis by one of the listeners. But even here it might paint a clearer picture (literally).

 

[Roen]

Those are pretty fancy pictures. I'm impressed. Thanks very much. I do wonder, however, whether the on'axis response should be highlighted so much. Given the near-impossibility--even with waveguides, and even more so with Coaxial configurations--to avoid on-axis diffraction effects, and given that for most people the first arrival will be the response at around 15 degrees off axis, I think it might be more informative to take the on-axis response out for at least one of the presentations so the listening window response is clearer. That's more problematic here, since one of the speakers is a center channel that will in fact be listened to directly on axis by one of the listeners. But even here it might paint a clearer picture (literally).

The graphs, to my amateur eyes, look like they will be a decent match?

 

[BYRTT]

Those are pretty fancy pictures. I'm impressed. Thanks very much. I do wonder, however, whether the on'axis response should be highlighted so much. Given the near-impossibility--even with waveguides, and even more so with Coaxial configurations--to avoid on-axis diffraction effects, and given that for most people the first arrival will be the response at around 15 degrees off axis, I think it might be more informative to take the on-axis response out for at least one of the presentations so the listening window response is clearer. That's more problematic here, since one of the speakers is a center channel that will in fact be listened to directly on axis by one of the listeners. But even here it might paint a clearer picture (literally).

Thanks you like presentation, the used colours had been tweaked lately to meet colours of ASR site good arguments there so tweaked a thinner line width for on axis curve as seen in below animation..

 

[Dennis Murphy]

Thanks you like presentation, the used colours had been tweaked lately to meet colours of ASR site good arguments there so tweaked a thinner line width for on axis curve as seen in below animation..

View attachment 92649

Thanks. I think that's more informative. You can see easily that the dip at 1 kHz is a systemic feature of the BMR, while the dip at 3 kHz is just a fleeting diffraction effect.. It took me awhile to figure out what's behind the 1 kHz dip. As it turns out (counter-intuitively) he little BMR midrange unit actually has broader dispersion above 1500 Hz than below. That's the region where the driver transitions from pure pistonic motion to a bending mode.

 

[BYRTT]

Thanks. I think that's more informative. You can see easily that the dip at 1 kHz is a systemic feature of the BMR, while the dip at 3 kHz is just a fleeting diffraction effect.. It took me awhile to figure out what's behind the 1 kHz dip. As it turns out (counter-intuitively) he little BMR midrange unit actually has broader dispersion above 1500 Hz than below. That's the region where the driver transitions from pure pistonic motion to a bending mode.

Agree and thanks BMR techtalk, should it happen be a TEBM46 BMR ask because plan some future diy fun using that transducer, below curve is on a dirty quick cardbord test baffle 120x90cm, it has some impressive high end reach so plan is midtweeter duty 550-580Hz and up..

 

[Dennis Murphy]

Agree and thanks BMR techtalk, should it happen be a TEBM46 BMR ask because plan some future diy fun using that transducer, below curve is on a dirty quick cardbord test baffle 120x90cm, it has some impressive high end reach so plan is midtweeter duty 550-580Hz and up..

View attachment 92667

it's the same driver. I think you'll find that the 7 kHz peak will come down substantially when you can fiddle around with baffle dimensions and have the driver properly loaded in a very small enclosure packed tightly with denim insulation. I sure hope so, because that peak would be a real screamer.

 

[BYRTT]

it's the same driver. I think you'll find that the 7 kHz peak will come down substantially when you can fiddle around with baffle dimensions and have the driver properly loaded in a very small enclosure packed tightly with denim insulation. I sure hope so, because that peak would be a real screamer.

Year very small enclosure volume with denim and/or also some toilet/kitchen roll paper can work wonder recorded in impedance sweeps, over at HiFiCompass site have traced their curves 0-60º (15º steps) of TEBM46, think it looks not bad as midtweeter plus my filtering is unlimited in a active Jriver DSP engine via dedicated power amp, crossover intension is 8th or 16th order linear phase within 1/8 wavelenght to midwooder so system dispersion is dictated by the BMR and kind of omni below 5-6kHz area or better, thanks as always interesting techtalk..

 

[Dennis Murphy]

Year very small enclosure volume with denim and/or also some toilet/kitchen roll paper can work wonder recorded in impedance sweeps, over at HiFiCompass site have traced their curves 0-60º (15º steps) of TEBM46, think it looks not bad as midtweeter plus my filtering is unlimited in a active Jriver DSP engine via dedicated power amp, crossover intension is 8th or 16th order linear phase within 1/8 wavelenght to midwooder so system dispersion is dictated by the BMR and kind of omni below 5-6kHz area or better, thanks as always interesting techtalk..
View attachment 92716

 

[Dennis Murphy]

You guys and your active dsp equalization. Next thing you know, you'll want to start putting automatic transmissions on cars.

 

[CoolHandDuke]

How would you think the tower version compares with the center, horizon version? I currently own the 2EX (and am very happy with them), but was considering jumping to the towers.

 

[Shazb0t]

How would you think the tower version compares with the center, horizon version? I currently own the 2EX (and am very happy with them), but was considering jumping to the towers.

They have the same driver assortment and internal cabinet volume. I imagine that they would measure fairly similarly. The Horizon Center was meant to be paired with the Sierra Towers after all.

 

[Shazb0t]

Would you be able to generate your EQ for this speaker @Maiky76? It would be appreciated!

 

[BYRTT]

Would you be able to generate your EQ for this speaker @Maiky76? It would be appreciated!

Think out of box it looks target voiced okay and most noise there is probably interference as port noise or so, until then have two EQ suggestions below that is based on Amir's anechoic analyze, upper one is my take on ideal for farfield target but it takes probably too many PEQ's for most so middle one is a 5 times PEQ compromise of upper one, and the lower one using 4 times PEQ is Amir's EQ trial from the listening test into post 1.

 

[Maiky76]

Would you be able to generate your EQ for this speaker @Maiky76? It would be appreciated!

Hi,

Here is my take on the EQ.

These EQ are anechoic EQ to get the speaker right before room integration. If you able to implement these EQs you must add EQ at LF for room integration, that usually not optional… see hints there: https://www.audiosciencereview.com/...helf-speaker-review.11144/page-26#post-800725

The raw data with corrected ER and PIR:

Score no EQ: 5.3
With Sub: 7.5


Spinorama with no EQ:

• Decent
• some resonances

Directivity:
Better stay at tweeter height
Horizontally, ON to +/-10deg

EQ design:

I have generated One EQ. The APO config file is attached.

• LW and Score are the same in this particular case

Score EQ Score: 6.1
with sub: 8.1

Ascend Acoustics Horizon Center APO EQ Score- LW 96000Hz
August032021-095552

Preamp: -3.4 dB

Filter 1: ON HPQ Fc 44.60,    0.00,    1.22
Filter 2: ON PK Fc 93.43,    -2.05,    1.42
Filter 3: ON PK Fc 601.62,    -2.38,    7.58
Filter 4: ON PK Fc 801.71,    -1.37,    5.40
Filter 5: ON PK Fc 1028.27,    1.32,    6.47
Filter 6: ON PK Fc 2336.67,    1.11,    0.68
Filter 7: ON PK Fc 3648.19,    -1.66,    10.30
Filter 8: ON PK Fc 18970.93,    3.36,    1.69

Spinorama EQ Score

Zoom PIR-LW-ON

Regression - Tonal

Radar no EQ vs EQ score
Minor improvements

The rest of the plots is attached.

 

[Shazb0t]

Wonder if 1kHz woofer or midwoofer.
As for 3-3.5kHz, I guess that shows again that RAAl tweeters like to be crossed at 4kHz.

You can't generalize about RAAL tweeters. The Horizon uses the OEM 70-20, which is actually designed to be crossed around 2 kHz, unlike the 64-10X used on the Sierra 2 series and my BMR, which needs a crossover point at least 1 kHz higher. From the vertical directivity plots, it looks like the 70-20 is being crossed around 2500 Hz (Dave would have to confirm), which is quite conservative for this unit. In any event, 1.5% maximum THD at 96 dB is an excellent figure for a ribbon and not of any audible consequence that I can imagine at this SPL.

Looking at the directivity index, I would have said ~3500Hz.
I doubt I could hear that distortion, but I would prefer no higher than 1%.

Could be, although that would be the highest I've seen for this tweeter, and the very broad beamwidth at 3 kHz seems inconsistent with the likely midrange dispersion, but Dave can educate us.

In CAD program its possible rotate horizontals and tilt verticals of Amir's spindata relative to the microphone at 2 meters distance, a tilt to minus 20º and plus 20º gives a clean deepest notch in area above 1kHz as seen below, +20º tilt land at 3500Hz and -20º tilt land at 3700Hz so around that area must be the XO frequency region

I was also curious about the midrange to tweeter crossover point and inquired with the speaker designer, David Fabrikant. His response is copied below:

... This is what makes the Horizon and our Towers quite magical.. There is a near perfect horizontal directivity match between the midrange we use and our version of the RAAL 70-20xram at where they cross. I discussed this in the polar response measurements here many many years ago.

http://www.ascendacoustics.com/pages...on%20Tower.pdf

Seems a bit silly to use a tweeter that is specifically designed to be crossed at ~1800Hz, costs about 5x more than the RAAL in the S2, and cross at 3.5kHz? Doesn't it? I mean, we could save a fortune and just go with the S2 RAAL ribbon in our towers and horizon if I wanted to cross that high.

Of course, if I did that, the horizontal directivity wouldn't be nearly as impressive.

The little ripple at 3.5kHz is just a minor reflection caused by the phase plug on the midrange driver.

There have been no crossover changes in the Horizon or our towers since their introduction.


[Second Response]

You got lucky today as we built a ribbon horizon for a customer that we didn't yet clean and pack for shipping.

It took some work on my end, I typically don't like moving fixed microphones around but this time I actually carefully measured its position so placing it back exactly where it belongs wasn't too bad

Here are the close mic response measurements for both the twt and mid. You can clearly see where they acoustically cross. I do apologize for poor quality, to save time, I took a pic of the monitor with my iPhone. Otherwise I would have had to find a USB thumb drive to transfer the files to this computer, and I have misplaced/lost every thumb drive I have ever owned

You can also see the reflection caused by the phase plug on the woofer in the woofer response, at ~3.5kHz.

It is also important to understand that since these are close mic response measurements and all drivers are connected, the tweeter response is somewhat influenced by the woofer. Which is why you see the tweeter roll off steeply and then start rising again at ~1600Hz. That rise is caused by the woofer...

There is no tweeter influence in the woofer response because I put a shield over the ribbon for that measurement.

Hope this makes sense.

So the actual midrange to tweeter crossover point is right around 2kHz. This helps explain why these speakers have wider than normal directivity in the upper midrange to the treble.

 

[MJT]

Aug 8, 2020
I couldn't put my finger on what was wrong with it other than my excitement level was not at max. Yes, that is a technical term. It is covered under US ISO standard, 23476-A (annex E).

Strangely ISO 23476 wasn't published until June 2021

 

[GabrielPhoto]

I do have a question about the measurements.
Why is the horizontal dispersion considered 70% even though some areas are clearly much less?