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

[dualazmak]

Frequency (Fq) Responses in the Completed System Measured by using “Cumulative White Noise Averaging Method” under the Present Standard Crossover Configurations and Relative Gains:

Part-3: Fq Responses in Amplifiers’ SP Output level before Protection Capacitors

Hello friends,

The present total system and configurations, of course, remain unchanged as can be seen like this;

This post shares the results in "3. Responses in Amplifiers’ SP Output level before Protection Capacitors";

For this purpose, I need reliable SP-level to RCA line-level converter, i.e. High-to-Low converter, for the recording of the "white noise" in the SP-level signals by using USB ADC Tascam US-1x2HR and Adobe Audition 3.0.1.

Even though we can find many reasonably priced High-to-Low converters mainly in car audio market, it is very difficult getting the guaranteed specification descriptions for those converters.

Audio-Technica's AT-HLC series, however, are rare exceptions in this category, and I purchased and fully validated AT-HLC150 which seems to be not available outside of Japan;

https://www.audio-technica.co.jp/product/AT-HLC150
For my present Fq response measurements, I do not need and do not use the two ACC delay power supply cables of AT-HLC150.

The small attenuator in AT-HLC150 was kept at MAX (bypass) position al the way during the measurements.

Here I will not go into the details of my tests and validation on AT-HLC150, but I could fully confirm the specification by using my tentative amplifier Yamaha A-S301 and Klipsch ProMedia 2.1 THX in my office upstairs. AT-HLC150 is really flat and low noise (almost no noise) throughout 20 Hz to 40 kHz tested subjectively in listening sessions and also objectively using the white noise averaging Fq measurements.

I found we need to connect the two ground cables to the ground level of amplifier (ground of phono stage or grounded chassis) for low noise and stable utilization of AT-HLC150;

The below photo shows the setup for SP output Level Fq measurements in Accuphase E-460;

Instead of the Be-SQ drivers, I connected the SP lines to 8 Ohm 100 W dummy speakers (resistors) for Fq measurements in complete silence.

I applied the same setup, using 8 Ohm 100W dummy speakers, for YAMAHA A-S3000 (to drive woofers) and Sony Ta-A1ES (to drive Be-tweeters and horn super-tweeters), with all the volumes and gains unchanged, and the JRiver's volume (as master volume) kept at -10 dB all the way in playing the white noise.


The recorded white noise by Adobe Audition 3.0.1 is like this;

One of the very nice features of my primitive "Cumulative White Noise Averaging Method" is that we can easily and flexibly prepare virtually mixed sound by using Adobe Audition's nice copy, paste and mix-paste functionalities like this;

Using this recorded (and virtually prepared) sound track, I analyzed the Fq responses in SP output levels before the protection capacitors in my system.

Firstly, analyzed by MusicScope 2.1.0;

And,

Then loaded into Adobe Audition 3.0.1, and the total response curve and the High Fq channels;

The entire Fq responses;

As well expected, the EKIO's target configurations are quite nicely reflected in the amplifiers' SP output level before the protection capacitors;

In my coming next post, I will share the Fq response measurements in amplifiers' SP output level after the protection capacitors, and one high-pass (low-cut) 1.5 microF capacitor in ST line;

I will, of course, share the comparisons in Fq responses "before and after" the capacitors in amplifiers' SP output level signals for Be-SQ, Be-TW and Horn ST;

 

[dualazmak]

Frequency (Fq) Responses in the Completed System Measured by using “Cumulative White Noise Averaging Method” under the Present Standard Crossover Configurations and Relative Gains:

Part-4: Fq Responses in Amplifiers’ SP Output level after Protection Capacitors

Hello friends,

The present total system and configurations, of course, remain unchanged as can be seen like this;

This post shares the results in "4. Responses in Amplifiers’ SP Output level after Protection Capacitors";

The setup for Fq response measurements by using my primitive "Cumulative White Noise Averaging Method" was like this;

Please refer to my previous posts #396 and #401 for the details of the audio interface TASCAM US-1x2HR and the high-to-low converter Audio-Technica AT-HLC150, respectively.

I applied the same setup, using 8 Ohm 100W dummy speakers, for Sony Ta-A1ES (to drive Be-tweeters and horn super-tweeters), with all the volumes and gains unchanged, and the JRiver's volume (as master volume) kept at -10 dB all the way in playing the white noise.

The recorded white noise by Adobe Audition 3.0.1 is like this;

I have been much interested in what would be the difference in Fq response before and after the protection capacitors and the high-pass (low-cut) 1.5 microF capacitor;

I could easily copy-paste the "before CAP sound" from my recorded data shared in my previous post #401 for before and after comparison, shown as follows.

Fq response before and after the 68 microF protection CAP in CH5+6 to drive Be-SQs (Be-squawkers);

Fq response before and after the 10 microF protection CAP in CH7+8 to drive Be-TWs (Be-tweeters);

And, Fq response before and after the 10 microF protection CAP plus 1.5 microF high-pass (low-cut) CAP in CH9+10 to drive horn STs (super-tweeters);

For the further analyses of Fq responses in amplifiers SP output signals in total, just like described in my previous post #401, I virtually added and prepared other five amplitude-matched sound portions;

Using this recorded (and virtually prepared) sound track, I analyzed the Fq responses in SP output levels before the protection capacitors, in my system.

Firstly, analyzed by MusicScope 2.1.0;

And,

Then loaded into Adobe Audition 3.0.1, and the total response curve and the High Fq channels;

The entire Fq responses;

As well expected, the EKIO's target configurations are quite nicely reflected in the amplifiers' SP output level after the protection and high-pass capacitors;

In my coming next post, I will share the Fq response measurements in actual room sound at listening position using one measurement microphone (Behringer ECM8000) with the new audio interface Tascam US-1x2HR which has 48 V phantom supply and low-noise mic preamp.

Having and referring to so far obtained Fq response results in the prior four stages in the multiway audio system, I will apply the same "Cumulative White Noise Averaging Method" of Fq response measurement in actual room sound in detail; all drivers singing together, each of the drivers, in L channel only, in R channel only, and of course, L+R together.

After that, I will also have another summary and discussion post on my Fq response measurements by using the "Cumulative White Noise Averaging Method" in the present multichannel multi-driver (multi-way) multi-amplifier system.

Edit to add at 21:41 Japan time on July 2:
As for the "shape of total Fq response curves", I can summarize the so far measured ones as shown like this;

 

[dualazmak]

Frequency (Fq) Responses in the Completed System Measured by using “Cumulative White Noise Averaging Method” under the Present Standard Crossover Configurations and Relative Gains:

Part-5: Fq Responses in Actual SP Room Sound at Listening Posistion Using One Meaesurement Microphone

Abbreviations in this post;

SW: Sub-Woofers Yamaha YST-SW1000 (L & R)
WO: Woofers of Yamaha NS-1000 (L & R)
Be-SQ: Beryllium Mid-Range Squawkers of Yamaha NS-1000 (L & R)
Be-TW: Beryllium High-Range Tweeters of Yamaha NS-1000 (L & R)
ST: Metal Horn Super-Tweeter Fostex T925A (L & R)

Edit:
As for the present frequency response of my BEHRINGER ECM8000 measurement microphone (specially selected unit in 2008), please refer to my post #831.

Hello friends,

Last weekend, during the really quiet past-midnight occasion, I did the microphone measurement sessions as Part-5 of the post series on Fq response analyses.

The present total system and configurations, of course, remain unchanged as can be seen like this;

This post shares the results in "5. Fq Responses in Actual SP Room Sound at Listening Position Using One Measurement Microphone";

The setup of the microphone Behringer ECM8000 was just same as I shared in my post #388;

Firstly, I rather intensively recorded and analyzed the white noise sound from the Be-TWs and horn STs, since they are currently driven by same amplifier Sony TA-A1ES with the EKIO crossover channels CH7+8+9+10 all together;

The Fq response curves in 2 kHz to 20 kHz were analyzed by Adobe Audition 3.0.1 with rather intense smoothing of FFT size 256 as shown below;

The reasons for I put the gray masks in 18 kHz to 21 kHz are that the measurement microphone Behringer ECM8000 cannot hear over 20 kHz, the white noise applied is up to 20 kHz, and also the rather intense FFT 256 smoothing starts at around 18 kHz to reflect/include the sharp cut-off over 20 kHz; the curves over 18 kHz, therefore, have little value for any discussion in this case.

As you can see in the above, Yamaha NS-1000's Be-TW starts to slowly but considerably decline over ca. 14 kHz which was also mentioned by @mikessi in his post #350 in this thread, and this is one of the main reasons for my addition of horn STs in my system (as mentioned in my post #351) referring to the unique alignment/positioning of STs, Fostex T925A.

We can see that the horn STs (Fostex T925A) has rather high efficiency compared to the Be-TWs; it has about "12 dB to 15 dB higher efficiency" than Be-TW, judging from these figures plus the fact of -4 dB gain setting in EKIO for CH9 and CH10.

These findings would suggest a feasibility of having variable and reliable attenuators for horn STs just before the ST terminal; this consideration would be practically not so important, however, since the sound over 16 kHz hardly to be heard, and also I have age-dependent slight decline in hearing over ca. 7.5 kHz. I will further discuss these aspects later in this post and maybe in the next summary post.


Then, as usual, I moved on to intensive Fq measurements of the real room sound at my listening position using Behringer ECM8000 microphone and the "cumulative white noise averaging method".

Even though I also carefully recorded the sound from L-channel drivers only, and also the R-channel drivers only, the following analyses and discussions are limited to the L+R sound at listening position, for better simplicity of your understandings in this post.

The actually recorded "white noise" sound track was like this;

Then, firstly analyzed by MusicScope 2.0.1;

And then, fully analyzed by Adobe Audition 3.0.1. In order to have "at a glance" view of Fq responses over 25 Hz to 20 kHz in common analysis parameters, I first analyzed using the FFT size of 4096;

As shown in the figure, the slight upward total Fq response above ca. 6 kHz has been intentionally configured by EKIO crossover and downstream gains (including the volume of Sony TA-A1ES) in order to compensate my age dependent slight hearing decline in that Fq range.

The high range portion, 1 kHz to 20 kHz, can be better shown in this figure analyzed by using FFT size of 512;

For better viewing and understandings on the "total" Fq response given by all the SP drivers, I could prepare this figure;

You may agree that the volume/amplitude in the super-low sound area given by SWs, 25 Hz to 60 Hz, wound be a little bit too high judging from the curve. In my real room sound listening, however, I just like the current setting of the "bass portion" for low-rich music files like pipe organ recordings; it would be just my "personal preferences" ( e.g., please refer to my posts here, here and here). Of course, the level can be easily and flexibly adjusted up-and-down by using IR remote controller for L&R SWs.

The EKIO's target Fq response curves and the real room sound analyzed by Adobe 3.0.1 and MusicScope 2.0.1 by using the primitive "White Noise Averaging Method" can be seen like this;

Now, I can also summarize the "total Fq response curve" at each of the five stages/levels of my multichannel multi-driver (multi-way) multi-amplifier system in this figure;

And, I can also summarize in this figure;

Although I know well and fully understand that Fq response is just one of the various factors (as we have been intensively discussing throughout this thread) determining the "total sound quality" of our audio system, I believe it would be worthwhile to measure and see the Fq responses in each of the stages/levels in this kind of multi-channel multi-way audio system.

 

[dualazmak]

Frequency (Fq) Responses in the Completed System Measured by using “Cumulative White Noise Averaging Method” under the Present Standard Crossover Configurations and Relative Gains:

Part-6: Summary, Discussions, and a Little Step Forward


Abbreviations in this post;

SW: Sub-Woofers Yamaha YST-SW1000 (L & R)
WO: Woofers of Yamaha NS-1000 (L & R)
Be-SQ: Beryllium Mid-Range Squawkers of Yamaha NS-1000 (L & R)
Be-TW: Beryllium High-Range Tweeters of Yamaha NS-1000 (L & R)
ST: Metal Horn Super-Tweeter Fostex T925A (L & R)

Hello friends,

In my posts #393 through #403, I could measure the Fq (frequency) responses in each of the five stages/levels of my multichannel multi-driver (multi-way) multi-amplifier audio system by using the primitive "cumulative (recorded) white noise averaging method";

The results can be summarized in the following two images/figures;

Throughout these careful measurements, I confirmed and validated that primitive "cumulative (recorded) white noise averaging method" is really powerful and reliable in terms of;

1. the method is universally applicable in the stages of digital out of crossover software (EKIO), DAC's analog out, amplifier SP out, and of course in the actual room SP sound,

2. the method is accurate, sensitive and reproducible, having little or no statistical fluctuation, because of the FFT averaging analysis on the "accumulated rich data" of the recorded sound,

3. the recorded "white noise tracks" can be re-analyzed any way, anytime, afterwards,

4. flexible mix-paste (sound mixing) can be done to virtually simulate any combination of the channels, especially in amplifiers' SP out signals before going into SP drivers,

5. if needed, the environmental "continuous room back ground noise" can be reduced/removed by the Adobe Audition's "noise capture - noise reduction" function,

6. if needed, suitable gain/level adjustment can be applied for "level matched comparison" of Fq response shapes between the different series of the recorded data,

7. flexible and suitable FFT size (as smoothing intensity) can be selected depending on the frequency zone of interest.


In any way, it is very nice getting this "reliable over 20 Hz to 20 kHz" total Fq response curve at the listening position, as reference for possible further flexible fine tuning of the system;

Now, looking at this figure, some of you may feel that the elevating/upward Fq response at higher region (above ca. 4 kHz) would be a little bit "too much" for your ears.

I also feel somewhat the same especially while listening to the music tracks of rich "high-frequency-transient-sound". One typical example of such music track is nice recording of harpsichord/cembalo by Scott Ross;

Since the Be-TWs and STs are presently driven by EKIO's CH7+8+9+10 signals by one amplifier Sony TA-A1ES through DAC8PRO's CH7 (L) and CH 8(R), I have currently no functionality to flexibly/variably adjust the relative gain for ST against Be-TW.

As shared in my post #403, the metal horn ST (Fostex T925A) has considerably higher efficiency than the Be-SQ, now I really would like to have variable "relative gain control" for ST in downstream at amplifier level or afterwards, for flexible selection of "upward Fq response slopes" in high Fq range over ca. 4 kHz.

In this perspective, I carefully considered the three options;

The Option-1 is the simplest way using one reliable and durable attenuator for ST. This means, however, going back to the "attenuator plus capacitor (1.5 microF low-cut/high-pass) passive network" sequence for ST which I really would like to exclude and avoid in my project from the very start. I really would like to avoid any of this kind of variable attenuators, sliding resistor type or fixed resistors rotary selection type, as these attenuators give variable impedance when looking from the amplifier.

I also always a little bit suspicious about the stability and durability of these attenuators having the physically moving contact points. Also even the "fixed resistors rotary selection type" attenuators would always have non-uniform burn-in issue for the individual resistor in it.

In my post #248, I intensively studied and discussed how to eliminate attenuators in this kind of very sensitive HiFi multichannel system.

Consequently, the Option-1 is clearly against my policy in this thread, and it is not feasible for implementation, even though I still have several really nice and fully overhauled attenuators.

The Option-2 should be the ideal solution in this system with DAC8PRO which offers up to 8 channels, if I would not care about the further considerable budget and physical space needed. With this option, I may control the gains for Be-TWs and STs completely independently, even though they receive the same signal of CH7+8+9+10 from EKIO through DAC8PROs CH7+8 while STs have 1.5 microF low-cut (high-pass) capacitor just before the drivers.

In this case of Option-2, however, I should select fairly nice high-quality "2-in/4-out XLR balanced active distributor" in professional audio gear market (rather expensive though), and I also need to purchase another XLR capable HiFi integrated stereo amplifier to enable variable gain control with the volume dial for STs.

I assume you feel the Option-3 would be rather eccentric or weird since in this option the SP level signal would be once depressed into RCA line level, and then put into another HiFi stereo integrated amplifier to drive STs dedicatedly. Especially, we should concern about the quality and specs of the "high-to-low converter" involved. I know well that I can eliminate such "high-to-low converter", if Sony TA-A1ES has pre-out RCA and/or rec-out RCA, but unfortunately this is not the case with the simply designed TA-A1ES.

As for "high-to-low converters", I shared in my posts #401 and #402 that Audio-Technica's AT-HLC-150 (seems not available outside of Japan though) is a kind of exception in this product category which has guaranteed spec sheet;

https://www.audio-technica.co.jp/product/AT-HLC150

For my present utilization in driving STs, I do not need and do not use the two ACC delay power supply cables of AT-HLC150, and the small attenuator in AT-HLC150 is always kept at MAX (bypass) position.

Again, here I will not go into the details of my tests and validation on AT-HLC150, but I could fully confirm the specification by using my tentative amplifier Yamaha A-S301 and Klipsch ProMedia 2.1 THX in my office upstairs. AT-HLC150 is really flat and low noise (almost no noise) throughout 20 Hz to 40 kHz tested subjectively in listening sessions and also objectively using the white noise averaging Fq measurements as shared in my posts #401 and #402 .

As for the possible additional HiFi stereo amplifier in Option-3, I am intensively testing the "rather budget side" but still very nice Yamaha A-S301, which I already have, for dedicated drive of STs;

You may find about Yamaha A-S301 at;
Overview:
https://usa.yamaha.com/products/audio_visual/hifi_components/a-s301/index.html
Features:
https://usa.yamaha.com/products/audio_visual/hifi_components/a-s301/features.html#product-tabs
Specs:
https://usa.yamaha.com/products/audio_visual/hifi_components/a-s301/specs.html#product-tabs

We also discussed on A-S301 at here and here. As @TheBatsEar mentioned there, and as people discussed in the thread entitled "YAMAHA A-S301 or A-S501 for KEF Q100?", the measurement data for A-S301 would be quite similar to A-S501 reported here.

During my long journey of amplifier exploration , I intensively tested and tentatively used Yamaha A-S301 at here through here, and found that A-S301 is quite nice at least for driving only Be-TWs and STs; please refer to Case-3 configuration in my post #291 where AS-301 was driving Be-TWs plus STs.

The total system diagram incorporating the Option-3 is shown like this;

And the total actual physical connections can be represented by this diagram;

In this total system configuration, I first applied the "cumulative white noise averaging method" for the Fq response measurement of Be-TWs and STs at listening position with three different gain sets (volume dial positions) in A-S301 while keeping the gain (volume) of TA-A1ES at - 17 dB. Please note that in this configuration, TA-A1ES acts as master gain controller for Be-TWs plus STs, and A-S301 acts as relative gain controller for STs;

Then the high Fq range response curves were given by Adobe Audition 3.0.1 as follows;

Having these variable high range Fq responses, I measured the Fq responses of the total sound at listening position;

And, finally I could get these total Fq response curves;

As well as;

As you can see, now we can flexibly and variably select the upward Fq response slope in high range, i.e. higher than ca. 4 kHz, by using the volume (gain) dial of A-S301 depending on the "high Fq ear listening capability" of the audience and/or on the Fq nature/composition of the music track. It is also very nice that we can control it by the IR remote controller for A-S301 while sitting on the sofa at listening position.

For the physical layout of the four amplifiers, fortunately the width of A-S301 is exactly the same as TA-A1ES, so that I could place A-S301 underneath of TA-A1ES; we should be careful enough for the stacking layout since the quasi class-A TA-A1ES is considerably warmer than class-AB A-S301;

I will continue my careful and intensive listening sessions using the above slightly modified four amplifier configuration during the coming one month or two before fully deciding it as my new standard setup.

 

[gene_stl]

How about an active two way crossover between the Be TW and the Fostex Super Tweeter amplifiers, and the Okto DAC?

You already have the amplifier. If you are already using a car audio component (some of which are very excellent) that would make it much easier to find a two way stereo system.

There are also many such crossovers in pro lines and they usually have XLR connections.

https://www.amazon.com/dbx-223xs-Stereo-Crossover-Connectors/dp/B004S2FVSM

https://artproaudio.com/product/cx311-2way-crossover-with-subwoofer-out/

Or a sony Esprit with custom plug ins.

or , or ........

 

[dualazmak]

How about an active two way crossover between the Be TW and the Fostex Super Tweeter amplifiers, and the Okto DAC?
You already have the amplifier. If you are already using a car audio component (some of which are very excellent) that would make it much easier to find a two way stereo system.
There are also many such crossovers in pro lines and they usually have XLR connections.
https://www.amazon.com/dbx-223xs-Stereo-Crossover-Connectors/dp/B004S2FVSM
https://artproaudio.com/product/cx311-2way-crossover-with-subwoofer-out/
Or a sony Esprit with custom plug ins.
or , or ........

Hello gene_stl,

Thank you so much for your continuing attention on this thread, and your above kind suggestions.

Yes, I also carefully considered about the possible addition of active crossover in downstream of DAC8PRO.

I am rather reluctant in that option, however, since these active crossover gears include internal ADC-digital XO processing-DAC sequences which again would raise the possible asynchlonization issue with the DAC8PRO in upstream also feeds signals to SWs, WOs and Be-SQs.

Throughout this thread, I have been always taking much care in avoiding any asynchronization issues and/or any possible drift out of sync over time between the DACs in different components.

The only one validated exception in this perspective is the AES/UBU digital out of DAC8PRO (which is digital through out of CH1+CH2) into second DAC device, if the second DAC is capable of AES/EBU digital-in and receiving sync clock info from DAC8PRO through AES/EBU and not using its own internal clock. I could confirm this feature by using Onkyo DAC-1000 with AES/EBU digital connection from DAC8PRO, as shared in my post #190.

In any way, we cannot exclude the possible asynchronizaiton issue unless otherwise synchronize the multiple DAC gears by external sync clock signal universally supplied/distributed by one reliable outer precision clock device. Unfortunately, however, DAC8PRO does not have outer sync signal input.

As you may agree, the ultimate best and ideal solution would be to replace DAC8PRO with similar 10-channel or more channel "pure DAC device", hopefully OKTO's DAC10PRO or DAC12PRO or DAC16PRO in which all the channels would be fully in sync by using single ES9038 PRO or fully synchronized two ES9038PROs. I once asked to Pavel of OKTO about the possibilities of such advanced DACXXPRO series, but his response at that time was just negative.

This is the one of the reasons for that I am much interested in Danville Signals possible one box solution dspNexus 2/16 which is not yet available, as we discussed before;
https://danvillesignal.com/dspnexus-dsp-audio-processor

At present, I will continue sticking to the really wonderful DAC8PRO in my project, and I do need separate 8 channels (stereo 4-way) for SWs, WOs, Be-TWs and Be-TWs+STs, where I would like to have simple variable relative gain control "in analog level" for STs against Be-TWs; this is just enough and feasible for the surprisingly wonderful total sound quality I have already achieved...

 

[gene_stl]

I was not proposing a digital crossover. An analog , would disturb the phase alignments probably less than a high level passive method.
You already have the requisite amplifier in place, and are operating the Be TW and ST in a fashion with it, that I don't think you would do, at lower frequencies.

 

[dualazmak]

I was not proposing a digital crossover. An analog , would disturb the phase alignments probably less than a high level passive method.
You already have the requisite amplifier in place, and are operating the Be TW and ST in a fashion with it, that I don't think you would do, at lower frequencies.

Hello gene_stl, hello friends,

Thank you for your kind follow-up. OK, now I fully understand your point; you are suggesting an analog crossover at downstream of DAC8PRO for Be-TWs and STs.

By the way, just this morning, I suddenly noticed a better alternative way for independent gain/volume control for Be-TWs and STs.

The CH1+CH2 of DAC8PRO provides "simultaneous" output of balanced XLR out, unbalanced RCA headphone out, and digital through AES/EBU out.
Both of the XLR out and RCA headphone out are under the DAC8PRO's preamp control, i.e. under the gain/volume control which I always set at -4 dB. (AES/EBU digital out is just through out of digital input and not under the gain/volume control.)

Consequently, I can use CH1+CH2 of DAC8PRO for BE-TWs and STs! Where the XLR out of CH1+CH2 should go into Sony TA-A1ES to drive Be-TWs, and the nice headphone out of CH1+CH2 with TRS-RCA adaptor should go into Yamaha A-S301 to drive STs. In this manner, I can control gain/volume for Be-TWs and STs independently.

As for the signal for SWs, I can use CH7+CH8 of DAC8PRO into RCA input of SWs Yamaha YST-SW1000. For this, I need to use XLR-to-RCA cable of No3 pin floating (disconnected); I believe this is OK since I will use it for SWs only, and the SWs have its built-in power amplifier for gain/volume control.

The "XLR-RCA with No3 floating (cold)" is not highly recommended, but Pavel of OKTO suggested it can be used even we will lose 6 dB, in his post here. I already have 4 of "XLR-to-RCA No3-cold" adaptors, so everything is ready for test and confirmation.

I will test and confirm above option very soon by the end of this week, and will revise my diagrams accordingly...

 

[dualazmak]

Hello gene_stl and friends,

Yesterday evening, I quickly tested the new four amplifiers configuration, and it works perfectly fine!

I revised the system diagrams as follows;

It is really nice that we can use the simultaneous headphone out (CH1+CH2) of DAC8PRO together with CH1+CH2 XLR out, both under DAC8PRO's preamp gain control. In the manual of DAC8PRO, we can read;

Analog connectivity
Main outputs (back side)
8 x Neutrik gold-plated XLR connector
Output level: 4.1 V RMS differential full-scale
output impedance: 200 Ohms
Headphone output (front side)
6.3 mm (1/4”) TRS connector
Output level: 4.1 V RMS full-scale
120 mA linear output current
fixed to output channels 1 and 2
output impedance <100 mOhm

Just for our future reference, as for the use of XLR-to-RCA (No3 floated/cold) cable adaptor at CH7+CH8 of DAC8PRO for SWs, let me repeat as follows;

The XLR-to-RCA (No3 pin floated/cold/disconnected) is not always recommended because of 6 dB loss. It should be quite feasible in my system here, however, since using it only for low Fq signal (ca. 15 Hz - 80 Hz), and the SW (sub-woofer) Yamaha YST-SW1000 has built-in powerful power amplifier. The nice reference post supporting this was given by Pavel of OKTO Research in his post here.

I am using two of Emotiva's "Balanced XLR to Unbalanced RCA Adapter Interconnect" for which I carefully confirmed "No3 pin floated/cold/disconnected" using my tester.

In my new revised EKIO configuration, I set +6 dB in Panel-1 and Panel-2 for SWs (which go into CH7 and CH8 of DAC8PRO) in order to compensate the 6 dB loss due to the XLR-to-RCA (No3 floated/cold) adaptors.

Let me also emphasize again that the utilization of HiFi grade high quality "integrated amplifiers" is really nice in this kind of multichannel multi-way system for flexible fine tuning of the total sound "on the fly", I mean flexible tuning while listening to the music.

 

[dualazmak]

Frequency Response in the Latest Configuration with Four Amplifiers plus Active Sub-Woofers

Hello friends,

As shared in above recent post #409, all the SP drivers, i.e. sub-woofers (SWs), woofers (WOs), Be-squawkers (Be-SQs), Be-tweeters (Be-TWs) and metal horn super-tweeters (STs), are now driven by fully independent amplifiers with volume/gain controller in each of them, for flexible and on-the-fly (during music listening) fine tuning of the total frequency response.

(Please find the larger diagrams in my previous post #409.)

Yesterday evening, using the established and validated "Cumulative White Noise Averaging Method" as shared in my posts #392 through #404, I briefly measured at my usual listening position and confirmed the variable fine-tuning control especially the relative gain for the STs by the dedicated amplifier Yamaha A-S301.

The high Fq range (2 kHz - 20 kHz) responses with three different ST gain sets are shown in this diagram;

This time, I set the SW-WO-SQ Fq range (20 Hz - 6 kHz) in rather flat shape by the volume/gain of SWs YST-SW1000, Yamaha A-S3000 driving WOs and Accuphase E-460 driving Be-SQs; Sony TA-A1ES driving Be-TWs was kept at usual -17 dB, and the Yamaha A-S301 driving STs was in three variable relative volume/gain sets as shown above.

The entire Fq responses (20 Hz - 20 kHz) with all the SP drivers can be shown in this diagram;

As shared in my previous posts #403 and #404, the elevating Fq response over 7 kHz is intending to compensate my slight age dependent hearing decline; now the upward slope over 7 kHz can be easily, safely and flexibly controlled by A-S301 driving STs on-the-fly, while listening to music, using the IR remote controller for A-S301 from the sofa at my listening position.

 

[dualazmak]

Fq Response Measurement under the Subjectively Best Fine-Tuned Gain/Volume Settings with the Four Amplifiers plus Active Sub-Woofers

Hello friends,

Yesterday evening, while repeatedly listening to my "Audio Sampler Tracks", I carefully fine-tuned each of the relative gain/volume of the four integrated amplifiers and the sub-woofers so that the music sound would best fit for my ears and brain. Then I measured the total Fq response by the "cumulative white noise averaging method" as usual, under the best fine-tuned relative gain/volume setting;

Please note that this measurement was done with one microphone set at my usual listening position on the sofa (3.6 m from the surface of the SPs) while all of the L & R SP drivers are singing.

As shared in my several previous posts, the faint upward Fq response over ca. 7 kHz (variably controllable with A-S301 driving super-tweeters) is intending to compensate my age dependent slight hearing decline.

 

[dualazmak]

Hello friends,

Just for your reference, very interesting and informative discussions are going-on on the following two threads where I am partly participating there having my experiences in this multichannel multi-driver (multi-way) multi-amplifier project;

What makes speakers "disappear " and can it be measured?

Midrange dome drivers banned ?

 

[dualazmak]

Intensive comparison of (high-range) sound cleanliness/clearness between Option-3 and Option-4 for driving super-tweeters (STs)

Hello friends,

I shared in my post #409 that my latest DAC8PRO-amplifier-driver setup includes;

the Be-SQs are driven by SONY TA-A1ES receiving CH1+CH2 XLR balanced out of DAC8PRO
the STs (Fostex T925A) are driven by YAMAHA A-S301 receiving CH1+CH2 unbalanced RCA headphone out of DAC8PRO

And the entire configurations can be summarized in this Option-4 diagram;

After intensively listening to various music tracks with this latest configuration for the past a few weeks, I now feel very very subtle deteriorations in cleanliness or clearness of the total sound (or I can say a very subtle cloudiness) in high-range sound, in comparison (by my memory) with the former sound even though the total Fq response has been always kept in this currently best shape;

Consequently, yesterday and today, I did rather intensive comparative listening sessions with above Option-4 and the former Option-3 where STs are driven by A-S301 receiving high-to-low converted input from the SP terminals of TA-A1ES driving Be-TWs;

I very carefully fine-tuned the volumes/gains of each of the amplifiers so that the total Fq response shape, as shown above, should be exactly identical for Option-3 and Option-4 configurations.

As the result, I found that the Option-3 configuration gives slightly but reproducible better cleanliness and clearness in sound compared to Option-4.

The reason(s) behind the finding would be, in Option-3 Option-4, Be-TWs are driven by XLR balanced CH1+CH2 output of DAC8PRO and the STs are driven by RCA unbalanced headphone CH1+CH2 output of DAC8PRO; even though the DAC8PRO's specification for headphone-out is really nice, but there would be slight difference/inconsistency in delay (synchronization) and/or phase rotation between the XLR-out and headphone-out, I assume, and these slight difference would result in some distortion causing a little bit of uncleanliness/unclearness in high-range sound when covered simultaneously by Be-TWs plus STs.

Furthermore, the unique physical alignment of the STs, i.e. positioned under the woofers as described in my post #27, is really one of the critical factors in my system in terms of the 3D sound perspective (3D sound stage); and hence the gain balance, delay/synchronization as well as phase consistency for the Be-TWs and the STs would be essentially important for the cleanliness and clearness of the total sound.

Now my wife also agree that the Option-3 gives very slightly better cleanliness and clearness in sound compared to Option-4, especially when listening to female vocal and violin tracks.

Having the above results and consideration, I rolled back my standard configuration into Option-3 setup where DAC8PRO's CH1+CH2 headphone RCA unbalanced out is used for L&R active sub-woofers. You would please note the entire Fq response shape remains unchanged as shared in my post #411.

(As for the unexpectedly nice performance of the high-to-low converter AT-HLC150, you would please refer to my post #401 and #404.)

 

[Doodski]

Super duper! You always make such nice graphics explaining your system. Well done.

 

[dualazmak]

Super duper! You always make such nice graphics explaining your system. Well done.

Thank you for your encouraging comment, as always.

Yes, I agree with you that graphical and/or schematic representations of our audio system greatly help many people easily understanding what we are doing!

 

[dualazmak]

Updated system diagrams and photos

**************************************************
Edit on July 7, 2024
Just for your reference,,,
The latest system setup of my DSP-based multichannel multi-SP-driver multi-amplifier fully active audio rig, including updated startup/ignition sequences and shutdown sequences: as of June 26, 2024: #931
**************************************************


According to my above post #413, I fully updated the system diagrams and photos as follows;

Current best tuned total Fq response;

As shared in my previous posts #403 and #404, the elevating Fq response over 7 kHz is intending to compensate my slight age dependent hearing decline; now the upward slope over 7 kHz can be easily, safely and flexibly controlled by A-S301 driving STs on-the-fly, while listening to music, using the IR remote controller for A-S301 from the sofa at my listening position.

Photos of the listening room;

And the distances;

The unique physical alignment of the STs, i.e. positioned under the woofers as described in my post #27, is really one of the critical factors in my system in terms of the 3D sound perspective (3D sound stage).

Please refer to my post #311 for details of the listening room environments.

Present physical layout of PCs, DACs and amplifiers;

 

[Doodski]

Trials with Hypex NC400 based DENTEC DP-NC400-4-EXP Multichannel Amplifiers

Quick and Brief Conclusion:
The total sound quality with DENTEC DP-NC400-EXP was really wonderful and very nice fit for my multi-channel project, and it is definitely the top-ranked candidate for introduction into my system, even though it is really expensive.

Hello friends,

As I pre-announced here in this thread, Mr. Fujimoto, President of SOUND DEN Sound Pro Shop, kindly visited my home yesterday with two of DENTEC DP-NC400-4-EXP on his car. DP-NC400-4-EXP has 4 (four) Hypex NC400 amp modules in it, and the amp and the wiring are very carefully fine tuned-up by Mr. Fujimoto and DENTEC people including cryogenic treatments and some of the thick pure silver wiring (I do not know the effectiveness, though).

DENTEC DP-NC400-4-EXP (still not yet shown at their web site) is the new version/model of their DP-NC400-4, and the new DP-NC400-4-EXP has its "heavy-duty transformer plus large capacitors type" power supply (PS) in external PS Units. The one PS Unit supplies DC power for two NC400 modules, so, therefore, two PS Units are needed for one DP-NC400-4-EXP main instrument. This new concept and design of external PS Units will enable future reasonable(?) upgrade with possible new Hypex amp modules keeping the PS Units would remain unchanged, he said.
View attachment 74266

View attachment 74268

Before to try and listen to the sound of DP-NC400-4-EXP in my stereo 5-way (10-channel) multi-channel system, Mr. Fujimoto and I carefully listened to my reference sound with DAC8PRO in single DAC mode and ACCUPHASE E-460 integrated amp plus the renovated LC-network as show below. As described before, the RCA input into sub-woofers YAMAHA YST-SW1000 was achieved by ONKYO DAC-1000(S) through sync AES/EBU (CH1+CH2) connection from DAC8PRO.
View attachment 74272

We very carefully listened to the reference sound of some of my Audio Sampler music tracks, and to his reference CD tracks recorded at their DENTEC Recording Studio. Then, he became much interested in the performance and sound quality of DAC8PRO (still only one in Japan, I believe) which he tried for the first time.

Next, I changed the connections at the SP Cabling Board into mutli-channel, multi-amplifier routes;
View attachment 74275

And we carefully connected 8 XLR cables from DAC8RPO into two of DP-NC400-4-EXP, and also connected the 16 SP cables to the binding posts of two of DP-NC400-4-EXP. He preferred that one to drive L-channel SPs, and another one to drive R-channel SPs, as shown below;
View attachment 74277

Now I set the software EKIO's configurations as follows which mimics the original LC crossover network of YAMAHA NS-1000;
View attachment 74278 and;
View attachment 74280

After careful listening sessions to the multi-channel sound of some of my Audio Sampler music tracks, and to his reference CD tracks, we also tried the second EKIO configuration using 48 dB/Oct LR filters for the crossover of woofers to Be-squawkers at rather low ca. 475 Hz as follows;
View attachment 74281
and;
View attachment 74282

Using above described procedures, we carefully compared the reference sound and the DP-NC400-4-EXP multi-channel sound in two different EKIO configurations.

Very regretfully, Mr. Fujimoto and the two DP-NC400-4-EXPs could stay at my home only for half a day; even though we could have the intensive ear-listening sessions, no measurement session with REW and ECM8000 microphone was performed this time.

OK, now what would be my (our) impressions on the sound quality with two DP-NC400-4-EXPs in my multi-channel system?

Here, I should not describe too much about our subjective impressions, but let me say that the total sound quality with DP-NC400-4-EXPs is just amazingly nice and wonderful in comparison with my reference sound of single amp system.

We agreed that my reference sound with ACCUPHASAE E-460 is still also very nice, and two DP-NC400-4-EXPs in multi-channel system gives audibly better responses of the SP Units (thanks to the elimination of the LC-network) and the S/N and distortion features are below audible level even in pianissimo small volume and also in rather large sound volume, just the same as in the reference sound.

We also found and confirmed that EKIO's crossover functionality is really wonderful and accurate with no deterioration of sound quality played by JRiver in 192 kHz 24 bit.

Also, thanks to the fine-tuned DP-NC400-4-EXPs, together with the still very nice cabinet and SP units of YAMAHA NS-1000, the audible 3D sound perspectives and phase features are just perfect in any way; even when we moved, left and right, up and down, front and back, around my usual listening position, the nice 3D and phase features of the sound remained unchanged.

This is my first experience in listening to Hypex or Purifi Class-D amplifier(s), and I was really impressed by the sound quality of Hypex NC400, fine-tuned by DENTEC, especially the smooth, very clear and clean but not "dry", I may say a kind of "rich", sound which is my happy surprise, as I have been thinking of somewhat dry or crispy sound with Class-D amps including NC400.

I heard that the two of DP-NC400-4-EXPs came to my home were already under the purchase order from one Japanese audio enthusiast, and this was the one of the reasons that they could stay at my home only half a day yesterday with careful handling of Mr. Fujimoto.

In conclusion, the total sound quality with DENTEC DP-NC400-EXP was really wonderful and very nice fit for my multi-channel project, and it is definitely the top-ranked candidate for introduction into my system, even though it is really expensive.

BTW, Mr. Fujimoto looks to be much interested in DAC8PRO; I also told him that DAC8 Stereo is now also available, and I will soon send him the links of OKTO RESEARCH as well as amirm's nice reviews on DAC8PRO and on DAC8 Stereo.

As per the usual you are a classy act @dualazmak (slang for classy) Love that system!

 

[Doodski]

@dualazmak if I may make suggestion can you please not use the black font or whatever color that is. I use the black background for ASR and the black text is very difficult to read.

 

[Doodski]

@dualazmak here is a screenshot of the text that is difficult to read.>

 

[dualazmak]

@dualazmak here is a screenshot of the text that is difficult to read.>
View attachment 147359

Hi, I understand your concern. But, I believe (almost) white background and black font are the default standard for viewing ASR Forum, isn't it? If possible, I would like to hear other people's standard font color for their posting...