• 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!

Multi-Channel, Multi-Amplifier Audio System Using Software Crossover and Multichannel-DAC

OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
I just tested the delay compensation function of EKIO by measuring the real sound with REW-Wavelet analysis;
WS000396.JPG
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
I also confirmed that REW-Wavelet analysis is also very much useful to "see" the listening room acoustic environments;
WS000398.JPG
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
I also checked and confirmed that delay settings in EKIO do not affect the total Fq response;
WS000399.JPG
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Together with my ear-listening trials and evaluations, the results of above objective measurements are well proving the accuracy and reliability of EKIO's crossover and delay compensation capabilities, I believe.
 

maxxevv

Major Contributor
Joined
Apr 12, 2018
Messages
1,872
Likes
1,964
If I ever eventually do such a project, I would go at it with a 3-way speaker and a set of active 12" subs.

Each speaker driver in series with a Hypex amp of appropriate power. Meaning 6 separate amps.

Was wondering, are there 6-line / 8-line co-axial wires of appropriate gauge thickness for such an application? It would certainly make connections a lot cleaner in my opinion.
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Let me share with you about my "audio check" sound tracks, "Super Audio Check CD" by CBS/Sony released in 1983;
https://www.discogs.com/Various-Super-Audio-Check-CD/release/12859958
Since I do not know whether English version of this CD is available or not, here is my private English translation of the track titles;
WS000400.JPG


These tracks are really nice and useful for our ear-listening and microphone measurement tests and fine tuning of audio system as well as tuning of room acoustic environments.

Just for example, "Track-2: phase check" and "Track-3: Five Location/Positioning" are very much useful for easy ear listening check and tuning of phase, level/gain and L-R balance of each of the crossover stereo channels.

As shown above, of course we have 20 Hz - 20 kHz Sine Spot and Sine Sweep tracks, Flat White Noise and Flat Pink Noise.

The "Track-20 High Fq Linearity Check" is really extraordinal recording of huge European Carillon containing very much powerful, high-energy, high-speed, high frequency bell ring sound together with wide range 30 Hz - 20 kHz base sound. I usually use this track to check tweeters and super tweeters, as well as total Fq linearity. I have never heard recorded sound having such a high-energy and high-speed high Fq components.

edited to add: Sony just indicated that the sound source for Track-20 is "Bimmel Bolle Classic Orgel", which sounds like huge Carillon for my ears.
 
Last edited:
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
I assume many of you are curious about my speaker alignment especially the position of super-tweeters (STs), FOSTEX T925A, sitting under the woofers;

WS000401.JPG


I know many people (almost all the people) using STs place them above the main SP unit, i.e. just above tweeters. In my alignment, on the other hand, tweeter and ST are far separated and they are sitting above and under the woofer+squawker, respectively.

Several years ago, I did really intensive listening sessions to find the best position of STs while listening to various kind of music, and finally reached this best alignment. I found, at least in my acoustic environment, this alignment gives best 3D perspective of total sound.

When I put STs on top of the cabinet, i.e. just above the TWs, the higher sound was allocated rather in upper space which was not pleasant for me. Then I tried each of the 8 ST positions in 45 degree-step around the cabinet, and found that the position under the WO is the only the best and magical position.
WS001733.JPG


In the position #5, WO and SQ are "sandwiched" by TW and ST, and the higher sound is coming not from point sources but coming from a big face/surface covering all the "NS-1000+outer round space", I feel.

Or, I should say, a big round face/surface covering all the "NS-1000+outer space" is emitting high sound, which is very much well allocated together with the sound from WO+SQ. Consequently, the 3D perspectives of the total sound and phase is just perfect and very much comfortable in this alignment.

At my listening position, about 3.5 m from the SPs, the 3D perspective of the total sound is really amazing; I feel all of my SPs just disappear and wonderful 3D sound appears in front of me.

I hope above descriptions would properly tell you my experiences on this...

I do not think the ST sound is much bounced by floor and/or carpet; the position is still about 30 cm above the floor. The Horn ST T925A is placed with its head a little bit "up" as you can see, and it is just targeting your ears/face. I cannot fully exclude, however, the possibility of tiny bouncing effect by floor and/or carpet.

Just for your info,my horn type ST FOSTEX T925A is very much high-efficiency one, and I set the attenuators for them in rather low -15 dB.

I highly recommend you to just try this alignment placing ST under your woofer+squawker...

Please refer to my post #84 for the differences between YAMAHA NS-1000 and NS-1000M.

Edit:
You would also please refer to my post;
- Overhaul maintenance of super-tweeter FOSTEX T925A and further signal fine tuning thereafter: #485

Edit on November 4, 2022:
Today, I wrote in my post here on that remote thread as follows:
>Our ears are located left and right of our head, and hence we are rather sensitive for left-right sound image/allocation. On the other hand, our ears and brain would be relatively gullible with up-and-down sound staging, I assume. I think that my unique physical alignment (positioning) of super-tweeters at under the main SPs, together with the upper tweeters, would give pseud-coaxial sound staging if complete time alignment has been achieved like in my setup (please refer to my summary post here), where tweeter and super-tweeter sing together higher than about 8 kHz.
>
>In any way, I would like to suggest those who using super-tweeters to at least try/test the positioning so that super-tweeter and tweeter would sandwich the woofer and squawker from up and down.
 
Last edited:
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Was wondering, are there 6-line / 8-line co-axial wires of appropriate gauge thickness for such an application? It would certainly make connections a lot cleaner in my opinion.

EDIT:
Please also refer to:
- Semiannual complete cleaning of all the metal-to-metal connectors:
#671
- Semi-annual intensive cleaning of all the metal-to-metal connectors/contacts, and complete renewal of all the tin-electroplated copper terminals with heat-shrink insulators: #895
- You should never solder the crimped spade terminals: #904
- Renewal of SP cabling boards beside SP systems:
#906

Do you mean speaker cables? I used to tried various kinds of SP cables including rather expensive ones, but now I stick to thick 5.5 sqmm or 3.5 sqmm reasonable price 600V VCT (Vinyl Cabtyre) cables;
WS000406.JPG


The "VCT x8 3.5SQ" is my recommendation for our multi-amp projects. I am using it like;
WS000407.JPG


You may google by "VCT x4 5.5SQ", VCT x8 3.5SQ", "VCT x12 3.5SQ", and so on.

I decided never to use banana plugs but always to use Y-lugs R-lugs as shown above, and also for amp and SP terminals.

After my 30-year try-and-error journey, now I definitely believe that these very cheap tin plated pure copper Y-lugs and R-lugs are the best terminal-contact solution since rather soft metal tin plate effectively increases the contact surface area when tightly connect and squeeze;
WS000408.JPG


AWG (UL), section area (mm^2) and SQ (JIS: Japanese Industry Standard);
WS000409.JPG
 
Last edited:

Lbstyling

Addicted to Fun and Learning
Forum Donor
Joined
Jan 2, 2019
Messages
500
Likes
461
Interesting! You are heading in a similar direction to myself. I wonder if it is for similar reasons?

In 2010? I was running MC16 or 18? (cant remember which), but I had a Echo Audiofire 12 channel DAC running into 12 amps (connex 500w per channel), then into a pair of 3 way DIY speakers (SEAS Excel drivers) for the L+R, with Duel JBL 2242 Subs. The rest of the DAC channels were running a 5 channel surround system, all calibrated using REW and a CM140 Mic. On to of this, I ran RePhase

JRiver can automatically switch between 2.1 channel and 5.1, so it was a solution that I was sure is the future.

In many ways, it was awesome! But not perfect.

The front speakers were SPL limited in the treble to 99db (Hyqiphon OW1) The amps were a little too noisy, and the computer system (J River, ASIO4ALL, Rephase) kept giving me issues with audio sinc when watching movies. When it worked it was magical.

In the end, a simpler system using a pair of JMlabs 816s speakers was superior in many respects. Improved pattern control of the speakers made a bigger difference that all the electronics, and in my large room, the extra SPL and sonic signature of the speakers made the sound more exciting.

I can suggest some basic things you could change? The speakers themselves would have reduced diffraction with a round-over added to the edge of the cabinet (the largest one you can do.) The problem is that this will mean modifying a classic speaker, so perhaps building custom cabinets would be advantageous?

The choice of speaker is an excellent one otherwise! That BE mid is extremely revealing, however, I would want to see graphs of how it behaves off axis given the age of the design. Its a wide cabinet, so has a good chance of being fairly good down to 800hz or so.

As you are running digital active so correction for pattern control, this adjustment is easy of course.

Are you running phase correction software in this?
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Hello Lbstyling,

Thank you for joining and for your interests, and I highly appreciated your nice suggestions.

Last summer, I renovated my YAMAHA NS-1000; all the LC network was removed and the two attenuators were bypassed for not in use. I added two stereo SP terminals so that WO, SQ and TW are now directly connected to SP cables. As you can see in my previous posts, I built outer LC+attenuator box with brand new coils, capacitors and three new attenuators (for SQ, TW and ST). This renovation was really very much effective in improving the total sound quality even though still driven by single-DAC + single-amp. I was again very much impressed by the total design and capabilities of my rather old NS-1000 and its SP units, WO, Be-SQ and Be-TW; they are still very very nice, and should be top ranked even now, I believe. (Be=Beryllium)

Then in last October, I started to consider that these WO, Be-SQ and Be-TW should sound much more efficiently with "much better sound quality" if all of the "LC coils, capacitors and especially attenuators" would be removed/eliminated so that they would be directly connected to each of the three dedicated amplifiers; yes multi-channel, multi-amplifier system. I knew several people were/are working on such multi-amp projrect with NS-1000 and NS-1000M, but almost all of them are using digital crossover hardware (like DBX VENU360) in analog line level, i.e. they actually need processing of PC --> DAC --> Pre-Amp --> [AD (again!) -->crossover processing --> DA (again!)] --> amps -->SP units.

As you may fully agree, I really would like to avoid multiple DA-AD-DA processes, and started to search and investigate software crossover solutions which directly receive digital signal from JRiver, Roon, etc. inside the same PC, and send the multiple channels into multiple DACs. I encountered, however, just like you, the synchronization issue between the multiple DACs.

I was, I am, very much excited about the news of DAC8PRO from OKTO RESEARCH and its really wonderful specs and capabilities as fully reviewed in this ASR Forum;
https://www.audiosciencereview.com/...d-measurements-of-okto-dac8-8ch-dac-amp.7064/
DAC8PRO is 8-channel DAC using single DAC tip ES9028PRO (so essentially no sync issue!) with wonderful specs just perfect fit for my project of stereo 5-way 10-channel together with my active sub-woofers YAMAHA YST-SW1000.

Your suggestions for further SP renovation and even having new SPs are very much appreciated, but please allow me at present to "fix my SPs as they are now" and to go into multi-channel multi-amp system to know how the total sound quality will be improved. I may, of course, consider further SP renovations after the completion of my current project shown in my first post here.

For your kind understanding, my current project is just for pure stereo L&R Hi-Fi audio system, and I have currently little interest in A/V system of 2.1, 5.1, 7.1 or more surround channels.

In any way, your continuing participation in this thread regarding multi-amp selection, REW-Wavelet analysis, room acoustic tuning, etc., will be highly appreciated.

As for the important "phase" match and tuning in my multi-amp project, of course we can use "phase inversion" of each of the EKIO's BW or LR filters (as shown in my post #2), and sub-woofers YST-SW1000 have their own phase inversion switch in it controllable with IR remote controller. Also, as I posted #26, "Super Audio Check CD's Track -2" is very nice for ear listening confirmation/tuning of "normal or inverted" of each of the 5-way stereo channels. As you can see in post #2, EKIO can show total (and individual) theoretical phase overlapped with Fq response curves.
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Phase issues...-1-

Now, many people who kindly keep eyes on my thread easily assume and imagine that I did, I am doing, various trials and experiments using EKIO's reliable and accurate functionalities including cut-off filter configuration, group delay settings, etc.
Yes, I did, and I am doing/enjoying, trials, simulations and experiments...

It is not my intention here, however, to share all of the details of my current trials and experiments with EKIO...

In this post, I would like to share about my general ideas and current solutions regarding phase issue;

Use steeper slope crossover cut-off --> complex phase issue!
Combination of cut-off filters of different slopes --> phase issue!
combination of different filter type --> phase issue!
Group delay setting --> phase issue!
L-R gain (balance) difference around cross point and/or groups --> phase issue!

Many of you may agree with me that these are the inevitable but major issues with us throughout our endless journey on the road of crossover+ multi-channel audio system, and there should be no single perfect solution.

My solutions or philosophy, at least in my current project, on these phase issues from "total sound quality" point of view are;

"The simpler, the better."
and
"The simplest, the best."

Here, "total sound quality" should be evaluated and determined not only by objective measurements, but also, more importantly, by actual ear listening to music (not always to sound).

Nowadays, I frequently tell myself; "You should not be objective measurement addict, but do trust your ears and your brain while listening to sound and music through your audio system".

In the following a few posts, I will briefly describe the bases and background for my above solution or philosophy.

to be continued to "Phase issues...-2-".
 
Last edited:

Lbstyling

Addicted to Fun and Learning
Forum Donor
Joined
Jan 2, 2019
Messages
500
Likes
461
Hi duel,

Use Pos's RePhase software to fix most of this. You cannot fix phase off axis, as you are not using a point source speaker, but it will get you close.

Having a smooth transition to 4pi pattern control would also improve this. (Removing diffraction points on the cabinet)
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Phase issues...-2-

Since I have no further physical renovation plan for my SPs at present and I will just change the driving engine from current single-DAC+single-amp to multichanne-DAC+mutichanel-amp, my present really nice "total sound quality" should be my standard reference sound throughout the project.

Even though I have a lot of experiences of attending the listening sessions for modern very much expensive SPs, ex. B&W 800 D3, Sonas Faber IL Cremonese, etc. at nice audio showroom in Akihabara Electric Town in Tokyo and at several very nice small hall-theater in Japan, I still do believe YAMAHA NS-1000 (as well as NS-1000M) is one of the best SPs in the history of Hi-Fi audio.

Edited to add on May 3, 2020;
You would please refer to my post of May 3 for differences between NS-1000 and NS-1000M.

Let's look at carefully, therefore, the original YAMAHA's design and LC-network of NS-1000;
WS000411.JPG


As we can easily find so may reviews on NS-1000 and NS-1000M, here I will not go into the details of rigid well-designed cabinet, magical 3D alignment of WO, SQ and TW, and the wonderful specs and sound of WO, Be-SQ and Be-TW (Be = Beryllium).

For our discussion on phase issues, I would like to have closer look and understandings on its original LC crossover network;
WS000413.JPG

As you can see, this is a very simple and conventional 12 dB/Oct high-cut low-cut crossover network, and we should notice the connection to Be-TW is inverted at the TW's cable terminal. After my recent renovation of moving the network+attenuators into outer box and direct cabling to WO, Be-SQ and Be-TW, I can now very easily hear the stereo sound of WO only, Be-SQ only, Be-TW only, Be-SQ+Be-TW without WO, etc., and of course all together.

Whenever I hear the "Track-2 Phase Check", "Track-3 Five Positioning/Location" and "Track-9 Sine Sweep 20Hz to 20kHz" of the "Super Audio CD" (see my post #26) only with Be-TW or Be-SQ+Be-TW, I am always much impressed by YAMAHA's fine tuning of the network coils and capacitors as well as the "inverted" connection to the Be-TW; the "phase perspective" is really smooth, natural, almost perfect, all over 20Hz to 25kHz. I believe this should be one of the many reasons that NS-1000 and NS-1000M have been used also as monitor SPs for a long period in professional scenes including many broadcasting studios/centers.

I believe that any physical 3D modification, such as "round edges" of the cabinet, should affect (better or worse) the 3D phase perspective of my NS-1000 which should be definitely avoided in this project (at present). I decided to use my NS-1000, therefore, "as it is now" for current multi-channel multi-amp project.

In case if I would change several physical and/or network "parameters or factors" for SPs at one time, then I should be in confusion to identify the deciding factor for better or worse "total sound quality".

I added ST FOSTEX T925A (inverted connection, same as Be-TW), and SW YAMAHA YST-SW1000 which has phase inversion switch controllable by IR remote controller.

And in my recent (last summer) renovation to take the network+attanuators out of the cabinet into outer box with brand new coils, film capacitors and attenuators, I fully copied the YAMAHA parameters as shown below;
WS000329.JPG


Using this renovated network+attenuators in outer box, now I have significantly improved "total sound quality" which should be my reference sound throughout current project. The present Fq (frequency) response measured at 3.5 m listening position with ECM8000 mesurement microphone is;
WS000386.JPG


From here, I would like to move forward into this multichannel project expecting some (much) further improvements by eliminating coils, capacitors and attenuators; they are now consuming a lot of amp power before driving each of the SP units.

The first to try EKIO crossover configurations after the arrival of DAC8PRO, therefore, should be quite similar to above current physical LC crossover, I believe.

Please note that the low-cut for ST is only one 1.50 micro-F capacitor providing 6dB/Oct slope; only this low-cut slope for ST will be inevitably changed if driven by dedicated amp, since EKIO's cut-off slope is 12dB/Oct or higher (steeper).

As for the SW's own high-cut filter, technical info from YAMAHA indicated that it is 24 dB/Oct BW LC filter, but, in any way, the SW has its own phase inversion switch.

to be continued to next post of "Phase issues...-3-" coming tomorrow, hopefully...
 
Last edited:

Neddy

Addicted to Fun and Learning
Joined
Mar 22, 2019
Messages
754
Likes
1,018
Location
Wisconsin
Hi, I'm enjoying following along with your design!

One small point for you to test out when you get your OktoDAC: I've found that (running JRiver and ASIO4ALL) it doesn't work at 192KHz; I had to set JR to downrez everything 192k and above to 96K. Above that I get stuttering after very long load times. (I have not fully explored multi-channel vs. stereo DSD files yet). (PC is a Gen6i7NUC with 16GB RAM.)

PS. FWIW, I took a different approach, and did my crossovers and DSP downstream of the OktoDACs AES (stereo) outputs using a dBX Venu360 to provide for biamped mains and subwoofer (single, it supports 6 outputs, 3 inputs). Since the Venu has AES inputs, no 'loss' with more rounds of AD-DA there (confirmed by their tech support), and spec-wise it is VERY good.
This does have the virtue of keeping things VERY simple on the software front end (and the Venu software/app is remarkably well designed).

Your diagrams are similar (but far more detailed, and thought out!) to the ones I had to do for my layout!
Amazing how complex this stuff can get.

Can't wait to see your pogress reports after the DAC arrives!
 
Last edited:

Lbstyling

Addicted to Fun and Learning
Forum Donor
Joined
Jan 2, 2019
Messages
500
Likes
461
Changing over to digital active will change the phase and delay of the system dramatically from Yamaha's original design. When you have setup your crossovers, run a REW test and check out the phase.

Here is a speaker I am playing with at the moment post EQ for you to get an idea what I mean.

Phase is in green.
 

Attachments

  • 2452H EQed.jpg
    2452H EQed.jpg
    130.6 KB · Views: 243
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Phase issues...-3-

Please let me finish my post series on "phase issues" before to respond to your kind posts.
I hope the following my posts on "phase issues" would somewhat responding to your comments and suggestions...

Well,,, having above careful considerations (my post #31, #33) on my present status of SPs, I first configured EKIO, simulating the current physical LC network, as shown below;
WS000415.JPG

Please note that all the cut-off settings are LR (Linkwiz-Riley) 12dB/Oct, and the phase "Invert" is checked at SL (super-low) and MD (mid range) pannels. You should also notice, as described in my previous post #33, the SP cable connection to TW (tweeters) and ST (super tweeters) are inverted at the SP terminals. Consequently, this EKIO configuration is almost completely simulating my current physical LC (coils and capacitors) network (see post #33).

In EKIO's "Frequency response" graphic panel, the theoretical "total phase rotation" is overlapped in white dot line which accurately reflecting all the parameters of crossover filters, including "check or not" of "Invert". As we can see, the total phase feature with this configuration is just simple as expected, thanks to the conventional 12 dB/Oct filters. Just for simplicity, I indicated only the "total phase rotation" in the above image.

Of course EKIO can show individual phases like;
WS000416.JPG


Also, the total Fq response (white solid curve) is almost flat over 30 Hz - 20 kHz, and,

the theoretical group and total delay are minimal, i.e. below 10 mSec, over 30 Hz - 20 kHz;
WS000417.JPG


Now I can direct all of the 10 channels into single-DAC + single-Amp system, so that I can easily compare the EKIO crossover sound with non-EKIO sound, of course both still passing through the physical LC network;
WS000418.JPG


I was very much impressed that the EKIO sound with these crossover configurations is just completely identical (with my ears and brain) to the non-EKIO sound in terms of total sound quality including the 3D phase perspective; I did very careful and intensive listening comparison using various audio check tracks and many of my "audio sampler" music tracks. I have also compared the EKIO sound with non-EKIO sound while driving SWs only, WOs only, Be-SQs only, Be-TWs only and STs-only, and also various combination of these, ex. Be-SQs+Be-TWs, Be-TWs+STs.

Consequently, after the arrival of my DAC8PRO, I should firstly try this simple and conventional 12 dB/Oct EKIO crossover configuration with multichannel-DAC (DAC8PRO) and multi-amp driving my present SPs, without passing through the physical LC network, of course.

OK, then what would be the theoretical phase rotations and the real sound, if I would configure EKIO with steeper slopes, ex. 24 dB/Oct, 48 db/Oct?? I will show you in my next post "Phase issues...-4-"...

to be continued to next post of "Phase issues...-4-" coming today,hopefully..
 
Last edited:

rebbiputzmaker

Major Contributor
Joined
Jan 28, 2018
Messages
1,099
Likes
462
Phase issues...-2-

Since I have no further physical renovation plan for my SPs at present and I will just change the driving engine from current single-DAC+single-amp to multichanne-DAC+mutichanel-amp, my present really nice "total sound quality" should be my standard reference sound throughout the project.

Even though I have a lot of experiences of attending the listening sessions for modern very much expensive SPs, ex. B&W 800 D3, Sonas Faber IL Cremonese, etc. at nice audio showroom in Akihabara Electric Town in Tokyo and at several very nice small hall-theater in Japan, I still do believe YAMAHA NS-1000 (as well as NS-1000M) is one of the best SPs in the history of Hi-Fi audio.

Let's look at carefully, therefore, the original YAMAHA's design and LC-network of NS-1000;
View attachment 58440

As we can easily find so may reviews on NS-1000 and NS-1000M, here I will not go into the details of rigid well-designed cabinet, magical 3D alignment of WO, SQ and TW, and the wonderful specs and sound of WO, Be-SQ and Be-TW (Be = Beryllium).

For our discussion on phase issues, I would like to have closer look and understandings on its original LC crossover network;
View attachment 58441
As you can see, this is a very simple and conventional 12 dB/Oct high-cut low-cut crossover network, and we should notice the connection to Be-TW is inverted at the TW's cable terminal. After my recent renovation of moving the network+attenuators into outer box and direct cabling to WO, Be-SQ and Be-TW, I can now very easily hear the stereo sound of WO only, Be-SQ only, Be-TW only, Be-SQ+Be-TW without WO, etc., and of course all together.

Whenever I hear the "Track-2 Phase Check", "Track-3 Five Positioning/Location" and "Track-9 Sine Sweep 20Hz to 20kHz" of the "Super Audio CD" (see my post #26) only with Be-TW or Be-SQ+Be-TW, I am always much impressed by YAMAHA's fine tuning of the network coils and capacitors as well as the "inverted" connection to the Be-TW; the "phase perspective" is really smooth, natural, almost perfect, all over 20Hz to 25kHz. I believe this should be one of the many reasons that NS-1000 and NS-1000M have been used also as monitor SPs for a long period in professional scenes including many broadcasting studios/centers.

I believe that any physical 3D modification, such as "round edges" of the cabinet, should affect (better or worse) the 3D phase perspective of my NS-1000 which should be definitely avoided in this project (at present). I decided to use my NS-1000, therefore, "as it is now" for current multi-channel multi-amp project. In case if I would change several physical and/or network "parameters" for SPs, then I should be in confusion for the reasons of better or worse "total sound quality".

I added ST FOSTEX T925A (inverted connection, same as Be-TW), and SW YAMAHA YST-SW1000 which has phase inversion switch controllable by IR remote controller.

And in my recent (last summer) renovation to take the network+attanuators out of the cabinet into outer box with brand new coils, film capacitors and attenuators, I fully copied the YAMAHA parameters as shown below;
View attachment 58443

Using this renovated network+attenuators in outer box, now I have improved "total sound quality" which should be my reference sound for current project, expecting some (much) further improvements by eliminating coils, capacitors and attenuators; they are now consuming a lot of amp power before driving each of the SP units. The first to try EKIO crossover configurations after the arrival of DAC8PRO, therefore, should be quite similar to above current physical LC crossover, I believe.

Please note that the low-cut for ST is only one 1.50 micro-F capacitor providing 6dB/Oct slope; only this low-cut slope for ST will be inevitably changed if driven by dedicated amp, since EKIO's cut-off slope is 12dB/Oct or higher (steeper).

As for the SW's own high-cut filter, technical info from YAMAHA indicated that it is 24 dB/Oct BW LC filter, but, in any way, the SW has its own phase inversion switch.

to be continued to next post of "Phase issues...-3-" coming tomorrow, hopefully...
A super project! NS1000s are amazing speakers, have owned them for many years. The mid and tweeter have very low distortion.
 
OP
dualazmak

dualazmak

Major Contributor
Forum Donor
Joined
Feb 29, 2020
Messages
2,792
Likes
2,921
Location
Ichihara City, Chiba Prefecture, Japan
Phase issues...-4-

At the end of my previous post #37 “Phase issues...-3-, I wrote;
> OK, then what would be the theoretical phase rotations and the real sound,
> if I would configure EKIO with steeper slopes, ex. 24 dB/Oct, 48 db/Oct??


Yes, just for such trial experiment, I made another EKIO configuration using mixture of 24dB/Oct, 36dB/Oct and 48 dB/Oct LR (Linkwiz-Riley) filters as shown here;
WS000440.JPG


The total and individual phase rotation in this configuration are much more complicated than the previous 12dB/Oct conventional case, as you can see;
WS000441.JPG


Using this configuration, I again did intensive listening session while playing various audio check tracks and also my “audio sampler” music tracks.

My first impression was; “Well, this also sounds quite good, in terms of cleanliness, resolution, S/N, sonority, 3D topology, speed, 3D perspective…..”.

After about 10 minutes non-stop listening to my “audio sampler” music tracks, however, I started to feel; “Wait, I can feel something strange or wrong, what is this?? A little bit artificial-like sound, not natural, I have in my ears and brain…. It looks a little bit hard for me to listen/hear this sound for several hours or overnight. I can feel some strangeness in this sound.”

Then I switched the sound into usual non-EKIO reference, and I clearly recognized the difference; “Yes, this is my reference, authentic very excellent sound in terms of total sound quality! I can easily listen to music with this sound quality for overnight…”

Since I am not a native English speaker, it would be very difficult for me to find suitable English expression for the “strangeness” I felt with the above EKIO crossover configuration. If you have an experience of having “car sick”, I may say that the “strangeness” would be somewhat similar to the unpleasant feeling of very early phase of car sick, may be mainly due to the more complicated phase rotations in the sound.

At the end of my above trial session, I could easily conclude that the conventional 12dB/Oct rossover EKIO configuration should be much better in term of “total sound quality” including the 3D phase perspective.

I just tested further complications by adding 15 msec delay at LO, MD, HI and ST panels of EKIO;
WS000442.JPG


It made the total and individual theoretical phase rotation in awesome complexity;
WS000443.JPG


The theoretical total and group delay are also complicated now;
WS000444.JPG


Again I did intensive listening session with this complicated EKIO configuration, and I easily recognized and felt significantly enhanced “strangeness” in “total sound quality”.

In conclusion, even just with my “ear & brain” measurement, “the simple (simplest) and conventional 12 dB/Oct crossover configuration with no delay” should have first priority to apply to my multi-channel multi-amp system after the arrival of DAC8PRO.

As for the fine tuning of possible slight “delay” of the super-low sound by SWs (sub-woofers), I can easily move the SWs (heavy stuffs, though) back and forth within max. 40 cm range, and I believe such “physical 3D tuning” of SWs should be much better than software crossover’s delay setting which brings phase rotation complication to the total sound.

I know that several people are suggesting me to use software "RePhase", but I believe I should not go into that path at this start-up stage of my project. In this type of "audio research", we should move up step-by-step with modifying single or minimum numbers of factors (parameters, elements) in each of the steps...

To finish my post series on “Phase Issues…”, let me go back to the content of my post #31;
I wrote there;
At least in my current project, from "total sound quality" point of view, my solutions or philosophy on these phase issues are;
"The simpler, the better."
and
"The simplest, the best."
 
Last edited:

maxxevv

Major Contributor
Joined
Apr 12, 2018
Messages
1,872
Likes
1,964
I would suggest getting a calibrated microphone and use REW to record your various settings.
You will get more consistent results and better reference points as to exactly where you should adjust much more quickly and efficiently.
 
Top Bottom