Some photos of my diy projects, to whet you appetite ...

My PFM Flea (a diy clock) became a tick because:
-- a tick is an arachnid (8-legs)
-- the orig. Veroboard Flea I hacked in 2007 now has two 4-pin oscillators (2x4=8), each can be quickly switched in between.
-- The Dream Project [https://www.diyaudio.com/community/threads/the-dream-project.399007/ ] is based on an original, strange NOS Philips/Magnavox CDP that was made for 8-pin tda1543

The photos above are a continuation of another exterior thread [https://www.diyaudio.com/community/threads/revisiting-asrc-sample-rate-conversion-and-up-sampling.397868/] that was de-serviced by "those in charge" (on diyaudio.com).
Some of this the following content is also based on postings on PFM (pinkfishmedia forum). E.g. see:
https://pinkfishmedia.net/forum/threads/flea-to-saa7210-remove-22k-resistor-for-x_out-x_in.280206/

Some of what you're looking at in the photos above:

Orig. Magnavox CDB2000 (Dream Project, in situ) --> CS8421 + TDA1545a --> Rudolf B/Jocko Homo IV (Dream Project, in situ).

The "CS8421 + TDA1545a" is a "plug-n-play" feature I had wanted for a very long time. It allows one to keep as much of the upstream and downstream (in situ) stuff in place while still allowing me to rapidly switch out, say, dac chips. The plug-n-play is not a hot/live feature (to protect electronics), but does allow one to swap out DACs (or DF+DAC, ASRC+DAC, etc) in under a minute. This should please some objectivists.

In the exterior thread, you'll note I had experimented with several other plug+play modules, all built on Vero.

The CS8421 + TDA1545 was an important module experiment as it allows one convert from I2S to EIAJ with a single SSOP device. All prev methods that I'm aware of (including ones that I have prototyped) have been messy glue logic using multiple 74hc devices. Various CS841x receiver chips -- like classic CS8414 -- also allow for I2S to EIAJ conversion. However, the CS841x receiver chips do not allow for I2S input. Only SPDIF.

With the CS8421 + TDA1545, using quite a bit of resistor-based pin-hardware logic control/hacking of the CS8421, I've got both I2S-to-EIAJ conversion and the goodness of ASRC -- all on that small SSOP device. Cool. Alas, for NOS purists, the CS8421's minimal output is 2x. Ahh ... but sonics ....
As reported in ...
https://pinkfishmedia.net/forum/threads/flea-to-saa7210-remove-22k-resistor-for-x_out-x_in.280206/
... the sonics are very, very good.

And it was resourceful use of the orig Flea that sat retired since 2009.
And Martin Clark (of Acoustica) may recognize one of his LM317 tracking -pre-regs (it's a dual +/- 15v version built in 2007; currently using just +15v for the Rudolf IV)

Stereophile already has:
Home » Galleries » DIY Projects
So some topical has already been here for some time.

Please upgrade / expand "DIY Projects" into the Forum. And a dedicated Blog may be cool , too.

Stereophile can compete in DIY space because ...

-- It has a LONG, reputable track record as a print magazine (brand name recognition, over 60 yrs old) that adopted the Internet early

-- The Stereophile.com web site has traffic rankings that have been decent and have held for quite some time. Alexa traffic ratings engine -- the prev. standard for Website traffic -- unfortunately called it quits in 2022. So we lost some historical metrics. HOWEVER -- there are other traffic-ranking engines:

https://www.similarweb.com/website/stereophile.com/#overview
Global ranking: 86,395

https://www.similarweb.com/website/theabsolutesound.com/#overview
172,191

https://www.similarweb.com/website/audiosciencereview.com/#overview
34,275

https://www.similarweb.com/website/diyaudio.com/#overview
60,264

https://www.similarweb.com/website/audiokarma.org/#overview
57,814

https://www.similarweb.com/website/audiophilestyle.com/#overview
187,415

https://www.similarweb.com/website/head-fi.org/#overview
28,475

Further notes ...

Some questions one might ask about orig. post projects (and my attempts at answers!!) may be:

Q: What DAC chips or digital chips are you concentrating your efforts on for your messy "dream" projects?
A: Mostly DAC chips that are >20 yrs old. Multi-bits and hybrid multi are very interesting (like under-rated tda1305). And many are still cheap on Ali. But overall used prices have been skyrocketing on many classics and even newer chips, like the PMD100. I also like to mix things up, in combos never used in other models or DIY projects. Like the weird CS8421 + TDA1545 coupling. And Also: Found some relatively inexpensive classic Sony over-samplers and 1-bit dacs on Ali -- including the ones used in the coveted ES series -- so I'm curious about sticking them in the Philips dream machine, just for kicks.

Q: Low-cost Chinese PCB kits "diy" and modable-whole-box units are avail., many using classic dacs (tda1541, tda 1387, tda1305, etc). So why bother with the messy diy Vero's?
A: That's true! And that's where many of started -- myself included . But:
-- The Chinese kits don't incorporate ALL the tweaks and possibilities I'm now interested in. Like my the weird CS8421 + TDA1545 coupling. Or even weirder DF1704 + dual TDA1545 coupling (see the DIYA thread for that project, that had many doubters, but I got it to work, and now is one of my fave-sounding weird combos). Yeah ... many of these combos really get into the weeds of complexity, even absurdity. That's how it goes!
-- The Chinese kits may incorporate under-performing ancillary devices/sections. I like good Amanero for USB . Flea clocks. Tent or other HQ osc's or crystals. Discrete output or I/V stages. Tracking pre-regs. Etc. And, of course, use of boutique or good-brand parts sourced from reputable distributors like Digikey, Mouser, Newark.
-- my messy Veroboard "dead bug", direct-wire approach actually allows trace lengths to be tighter than PCB, using quality, thick, US-made wire and Kester silver solder.

...and ...

-- diy is more fun than ordering from a ready-to-go from Ali. I think that's why you are here ... in this thread, on this Forum.

The DREAM Project
[ for some reason, this was not appreciated at: https://www.diyaudio.com/community/threads/the-dream-project.399007/ ]

A 20 year-old fantasy, finally "a dream within a dream" ...

Bottom line:

Plug-n-chug myriad diy digital devices into a common battery-powered headphone-amp/IEM combo, with swap time < 1min. Take notes. Compare/contrast. Report as necessary. This thread as main reference.

Other DREAM features:

COMMON: Input (line) device -- such as a well-functioning, up-to-spec CD player (shown); and output stage (as shown: Rudolf Broertjes’ SS I/V Gain Stage, much-modded ).
In-BETWEEN: "Plug-n-chug myriad diy digital devices"

Other goals :

- Plug-n-play (but not hot swap, to protect electronics!)
- Occams' Razor
- All-things/else-held equal
- ABX and objectivitist-friendly
- Use of test equip. and software for metrics and data acq.: FLIR, DSO, SA, Cosmos ADC.
-
"Bottom line" is a combo of tools, techniques, developed over time, and its evolution.
- Update this thread as necessary.
- Vocabulary and grammar: Noun, verb, adjective, adverb, preposition, metaphor, etc.

More coming soon ...

====== Some photos below have been prev. posted in other recent threads or posts ======

above: wide / top view of Dream Project platform. A MAgnavox CD2000 cd player (1989) that did NOT strangely employ oversampling. The plug-n-play action-area is roughly middle right of photo.

above: one of plug-n-play modules -- DF1704 + 2x tda1545a (the 1545 is an 8-pin Philips stereo DAC, but the DF1704 is meant for 2x PCM1704 mono DACs; so I am using only the R channel of each 1545 !!!; channel separation is top-notch!!)

above: 7hz Timeless AE planar magnetic iem's resolve detail well -- ideal for testing

above: SAA7220 + 8x tda1387 module. The parallel tda1387 module is meant for direct replacement of tda1541(a)

above: SAA7220 + 4x tda1387 module. The parallel tda1387 module is meant for direct replacement of tda1541(a)

above: SAA7220 + 1 x tda1543 . Looking at schematic of an old Philips CDP using the same config, in another CDP.

above: 8x tda1387 module. Non-oversampling config.

above: misc. wide shot of testbed project; more info to be added here soon

above: misc. wide shot of testbed project; more info to be added here soon

above: misc. wide shot of testbed project; more info to be added here soon

above: misc. wide shot of testbed project; more info to be added here soon

More coming soon ...

I suppose an alternative name for the purposed child (sub-) forum may be "prototyping".
In many ways, the excitement and impetus for adding and experimenting with these strange modules is similar to the fun of playing with new or used with tradit components ... just not as "pretty" ...

Kal Rubinson wrote:

I have some parts that might interest you.

If you're serious, then interested!!! Thx for the offer. How do I PM you in Stereophle forum ?
Feel free to discuss parts non-privately -- as you see fit ;)

Superposition: a mathy, physicsy way of stating something like multiple states at once; and not either/or or neither/nor.
Such as the case CAN be with audio "philosophy" ... so not, Objectivism vs Subjectivism.
But Objectivism + Subjectivism. Etc.

The photos below may allude to something of the kind ...
Took delivery of some IC parts yesterday ... mostly, Sony and Philips "single-bit" DACs and some Sony oversampling IC, all from the early 1990s. JG Holt and other reviewers found the Sony ES CD players to be good-sounding units. And they measure well, too. So I will throw them into the experiment queue.

above: ... have a look at "Superposition". Yes, those are Sennheiser's flagship IE900's there. Essential for scientific eval. of physicsy phenomena.

above: Philips SAA7323 df+dac, SAA7350 df+dac; Burr-Brown PCM1750 (an 18-bit a/d); Sony CXD2561 Pulse DAC, CXD2560 df

above: same + different (Superposition)

To be cont. ....

Kal, LinkedIn request sent and ... I sent Jim Austin a note to pass on my email addy to you.

My latest chip acquisitions ... all handheld shots.
The DAC chip used in that 1993 Stereophile JG Holt review ...
https://www.stereophile.com/content/sony-cdp-x779es-cd-player

... is dual CXD2562Q Pulse dac preceded by a single CXD2561 digital filter.
Unlike Philips, Sony is a bit weird in not releasing datasheets for certain ICs they produced. One has to go into the service manual, or annual semiconductor manuals, but even those obfuscate info.
If you go to ...
https://www.dutchaudioclassics.nl/the_complete_d_a_dac_converter_list/

... and use the page search feature of your browser (Ctrl+F) to see how many hits one gets of certain ICs, you can get the feel of which chips (DAC's, DF's, ASRC's, etc) were popular. E.g., the Philips saa7220 + tda1541 combo was used in many players in the late 80s, (including some Sony's!!) .
From there, sites like hifiengine have service manuals and alldatasheet.com to dig deeper into device or product topology.
I use this ad hoc process to determine what combo of chips to experiment with.

Back in 2015, I posted some photos of other DIY projects.

https://www.stereophile.com/content/ancient-magnavox-cd-player-2015-why

Perhaps my greatest effort to date was the "Magnavox Mess" -- a Philips/Magnavox CD-650, with an outboard PS with DIN connectors.

Yes, I was emulating certain Naim CDPs that used saa7220/tda1541a combo (CDS). The sonics from the Magnavox Mess project is the best "digital" I've heard to date, and even exceeded the sonics of my VPI-19/Sumiko-BP Special analog rig.

The Radio Shack Veroboard project in these photos is a high-quality SPDIF device I built in 2010.

SPDIF coax RCA goes in and out, thru the device, via a very high-performance Newava pulse transformer, a 74HC logic buffer, and a good circuit with decent parts quality. A 5v pwr source is connected at the green terminal .
This device makes for nice sonic improvements, and is considerably less $$ than boutique RCA spdif cables. I use AudioQuest spdif cables, btw.
The silver QLS player was sometimes called the "digital turntable" in forums. It only plays WAV files -- via SD cards -- and was designed by a Chinese audiophile. It has an unusual dual-LiPO battery +/- power supply built in, which improves performance. And it can output SPDIF and I2S. I find the QLS useful in repeat-play as burning- and breaking-in tool for headphones, cables, etc.
A Schiit MODI MULTIBIT 2 DAC give the devices some scale. (More DIY tweaks to that Schiit coming soon)
DIY can very happily co-exist with the tradit. audiophile hobby -- some of us find it more interesting this way!

(Note: The Forum seems to cropping these images. Right click and open in new window or tab to see entire frame)

Note: I began this topical thread on Diyaudio.com (as now-banned member "philipsmarantz") . I am bringing all of my content from that and other forums onto the Stereophile forum, and expanding and clarifying the content concomitantly. Much of my work on the ASRC project (phase 1) has been completed. That is, I have successfully implemented the big three ASRC devices, popular over a decade ago, but still respected for their upsampling ability -- and used by Benchmark, Bel Canto, Musical Fidelity, Bryston and others. Most of the DACs and some CD players were reviewed by Stereophile, and measured by John Atkinson. I _THINK_ there was some correlation with improved subjective and objective performance when ASRC devices were used prior to DACs. All of these were of the delta-sigma variety commonly used by manufs.
In my experiments over the past few months, I have implemented these ASRC devices -- the big three, SRC4192, AD1896 and CS8421 -- in novel arrangements: Not with delta-sigma DACs. But with classic Philips multi-bit DACs, with and without intervening DFs (oversampling) , and as format converters (I2S to EIAJ, etc). All with ultimate success! And even excellent sonics in some cases!
So lets start at the beginning a few weeks ago ...
(see: https://www.diyaudio.com/community/threads/revisiting-asrc-sample-rate-conversion-and-up-sampling.397868/ for addit. comments) ...
==============
2023-04-04 8:00 am

Upsampling (and not oversampling) -- via a dedicated SRC chips like AD1896, TI's SRC4192 Cirrus CS8421 -- was somewhat popular in audiophile devices roughly beginning in the late 1990s and up to roughly 2018, in standalone D/A processors. dcs may have stated the trend in late 1990s.
And then the fad seemed to vanish. Not sure why the use of upsampling -- via ASRC -- went out of favor?
Perhaps some features became integrated into ESS and AKM asics; or FPGAs.
As a matter of fact ... I'm not really sure why it became popular in the first place ... maybe ASRCs kinda/sorta "re-clocked" the I2S lines in a clean way, all in the convenience of one IC device ?
Anyway ...
The ASRC chips are still around and stocked by Mouser and Digikey, at low volumes or only bulk qtys, and pretty at high $ for each chip. eBay has them for cheaper.

My parts bin has quite a few of them, many models and manufs, collected over 15 yrs ago, never used, but ready-to-go on micro PCBs. So I might experiment -- see what the old craze was about.
Looked on Ali for kit PCBs and found none.
Did find this French PCB module based on that 20-pin CS8421 device:

https://www.audiophonics.fr/en/interface-modules/asynchronous-sample-rate-converter-module-cs8421-32bit192khz-p-11456.html

======
Some of my "parts bin" collection ... the ones on the green boards are AD1896 and CS8421:

============

2023-04-04 10:29 pm

What I "meant" in the OP is that ASRC is acting like a buffer, or re-clocker. Similar to 74HC glue logic -- just getting the timing into better shape.
The various datasheets of the ASRC chips do go into quite a bit of theory and math as to how they work and their internal architecture.

One of the strangest uses of upsampling was in certain Musical Fidelity DACs. They offered selectable 96k and 192k via a rear button. Looking at the schematics of the MF, there seems to have been two SRCs used (????). CS8420 followed by a NPC SM5849AF, and then a PCM1738 dac. I'm guessing both ASRCs were necessary to achieve 192k.

=======

2023-04-08 2:06 am
#19

analog_sa said:
---------
As soon as async usb boards became available i felt no further need to use an asrc, to my ears it even became detrimental to SQ.
----------
This may be operating-system dependent.
Benchmark, way back in late 2007, experimented with both settled on synchronous USB -- asynch USB Amanero and XMOS may have been unavail back then:
https://www.stereophile.com/headphones/108bench/index.html
Also the quality of the Amanero USB board may only be so-so. Compared to, say, a TAS1020B + SRC4192.
. According to Benchmark's Elias Gwinn (footnote 1), the DAC1 USB runs in the USB protocol's "synchronous" mode, to allow the host PC to, at all times, send audio data at the original sample rate of the audio being played. If the PC (or, more specifically, Windows XP's kmixer DLL) is not forced to do sample-rate conversion, it can maintain bit-transparent operation. "There is a tradeoff, of course," said Gwinn: "significant amounts of jitter arrive at the DAC1. This is not a problem for the DAC1, however, because Benchmark's UltraLock clocking system makes it immune to jitter."
I'm still unclear as to whether jitter optimization is why an ASRC might improve sound (subjective). It might be due to the same dreamy magic that improves sonics when going thru a separate preamp + power amp. Rather than much simpler (shorter trace paths, etc) integrated. Or a DAC with volume control (straight into a pwr amp) -- but sonics improve if you first put it thru, say, a good Krell pre-amp.

=====

About jitter ... I don't recall any mention of "jitter" per se in any tech paper or even Red Book until about the end of the 1980s. Well into the 4th/5th gen players and early stand-alone D/A processors. And then there was another metric du jour gaining popularity: linearity.
But all this spec craze was years after some decent sounding TDA1540/1 machines -- especially the mod jobs, and Meridian and Misson high-enders -- began convincing the vinyl community that digital was pretty good.

==========

2023-04-12 7:42 pm
#35
Just acquired another Amanero ... this one is supposed to be the better device, as suggested elsewhen in this forum. (Both shown in photos below)
Recent success with the AD1896 asrc ** is prompting another potential experiment. Up-sampling w/o over-sampling.
[** Highly suggest others repeat that experiment -- easy to to do -- and report back.]
I do have two ASRCs ready to go: CS4121 and AD1896. Also have SRC4192 on hand (but that still needs to soldered onto green proto board).
Several DACs in the parts bin ... but many have built-in DFs already. AD1862 might be a good candidate.
Or how about ....

========

2023-04-12 9:24 pm
#36

A follow-up on the ASRC ad1896 experiment that should've been continued in this thread, not the RESET (with overbar) thread.
The high-school math exams I grew up with were prefaced by a note in BOLD: Show Your Work.

Hence, for the "exam" experiment, note:

Amanero feeding I2S plus MCK into the AD1896 w/ music [Not shown, no music on DATA. It is 0.0hz]: .

=========================

AD1896 feeding the TDA1305 dac w/, and w/o music [note that DATA line playing no music, has "background" 132.xx Khz signal --- dither??? Also note that up-sampling is an integer 2x, which is ideal ].

Further comparing sonics of crude up-sampled bread and protoboard set up against my much-modded Chinese dual (parallel) tda1305 kit, are very positive. But, to be fair, the Chinese kit is battling its internal inferior USB topology, as well as no ASRC.

======

2023-04-12 9:49 pm
#38
MarcelvdG said:
-----
Do you mean using an ASRC with the exact same input and output sample rates? The only advantage I could think of is that the AD1896 can properly round > 20 bit signals to the 20 bit word length of your DAC.
-----------
Not to be "mean" (= facetious), but I don't know what I mean ;)
That is, I don't know why the ASRC improves sonics? I definitely heard improvement when I added the AD1896 into the 1305 setup -- compared to no SRC.
Yes, as can be noted in the photos, the AD1896 will 2x "oversample". But it might be interesting to use the ASRC with, say, a TDA1543 or 1545 or 1541.
I suppose I should do some objective tests. But one can only go so far w/o, say, this: https://www.ap.com/analyzers-accessories/apx555/

===============

2023-04-12 11:18 pm
#40
-------------
bohrok2610 said:
According to TDA1305 datasheet tables 3 & 5 all filter characteristics are worse with double speed.
------------
Yeah, I saw that and the note on pg. 8. [datasheet]
Weirdly, table 5 (double speed) does not provide a "SAMPLE FREQUENCY" column as Table 3 does.
I suppose it's not surprising that the faster the internal circuits are asked to perform (2x vs 1x), the more demands there will be on power supply and regulation. So more hash in line. Delta-sigmas run so fast that they have to noise-shape their way out of that mess. The 1305 does do a little noise-shaping per its "hybrid" design.

Btw: from p. 2 of datasheet:
"The TDA1305T is a dual CMOS DAC with up-sampling filter and noise shaper."
Hmmm ... in any case, ... Naim liked it, in their Golden Years ;)

=========
... to be cont in Revisiting ASRC (sample rate conversion) and up-sampling -- 2

Content below continues and updates "Revisiting ASRC (sample rate conversion) and up-sampling"

Others' comments per orig. DIYA thread is between " ---" with member username noted for ref. As shown here

------
Others' comments per orig. DIYA thread
-------

All other content is mine (philipsmarantz on diya; hollowman on Stereophile and elsewhen).

Indiv posts segregated by " ==== "

______________________

==============

2023-04-13 1:02 pm
#41
philipsmarantz said:
Not to be "mean" (= facetious), but I don't know what I mean ;)
That is, I don't know why the ASRC improves sonics? I definitely heard improvement when I added the AD1896 into the tda1305 setup -- compared to no SRC.
Yes, as can be noted in the photos, the AD1896 will 2x "oversample". But it might be interesting to use the ASRC with, say, a TDA1543 or 1545 or 1541.
I suppose I should do some objective tests. But one can only go so far w/o, say, this:
https://www.ap.com/analyzers-accessories/apx555/
KN replied:
----------
If I recall correctly, the ratio-estimator block of the AD1896 co-incidentally enables it to very effectively suppress incoming data stream clock jitter. Much more so than is typically provided by a Digital Input Receiver chip's analog PLL. Perhaps, that is the reason you hear an improvement?
-----------
==========

2023-04-13 7:44 pm

-----------
#45
bohrok2610 said:
And when comparing Amanero+AD1896+TDA1305 to synchronous USB CM108+TDA1305 I would not expect possible jitter suppression of AD1896 to be the key for any audible differences.
-----
In the Chinese 1305 kit (see the Naim thread), there is the another device between the CM108 and TDA1305: CS8412. The CM108 has a 12Mhz osc. feeding it its XO. Also, the CM108 block diag. has something called a Clock Gen. What is this?
I don't recall exactly how I probed the kit dac, or mod it, but I couldn't get the 1305s to work w/o the CS8412. The CS8412 spits out MCK (256fz) at pin 19 -- in sync with its I2S lines - - and that feeds the 1305 MCK.
So is the Chinese kit using "synchronous" USB?
----------
bohrok2610 noted:

According to the datasheet CM108 has both SPDIF and I2S output. It seems that on your board CM108 sends SPDIF output to CS8412. That has nothing to do with synchronous USB which is used between host and CM108.

================

2023-04-13 10:37 pm
#47
In the Chinese 1305 kit, I went back and traced the I2S lines between the 1305 and CM108.
When a USB cable is plugged in, the CS8412 seems to be bypassed (via the Omron relays???) -- and the CS8412 (with no SPDIF but just USB plugged into Chinese kit) outputs "nonsense" at MCK and I2S outputs.
As far as CM108 and USB mode. The datasheet notes: "Isochronous Transfer uses Adaptive Mode with Internal PLL for Synchronization"
So neither async or sycn according to [1].
But Audiophilesytle seems to have varied opinion [4].
===============
(1)
http://bucarotechelp.com/networking/basics/88092302.asp
(2)
https://pdf1.alldatasheet.com/datasheet-pdf/view/87252/CIRRUS/CS8412.html
(3)
https://www.digchip.com/datasheets/parts/datasheet/574/CM108-pdf.php
(4)
https://audiophilestyle.com/forums/topic/31603-asynchronous-and-isochronous/#:~:text=USB%20audio%20always%20uses%20isochronous,from%20the%20USB%20packet%20timing
=================
2023-04-14 11:13 am
#57
All else held equal, the USB sync vs async is pretty confusing -- not the technology, per se, but the benefits of one over the other. And even after Gordon Rankin claims to have solved some problems with his Streamlength patent, see https://www.usbdacs.com/Concept/Concept.html , the sync/async issue was challenged by other major players, like

Benchmark:
---
The USB input is taken to a Texas Instruments TAS1020B chip, which extracts the audio data using a phase-locked loop and converts it to i2S. According to Benchmark's Elias Gwinn (footnote 1), the DAC1 USB runs in the USB protocol's "synchronous" mode, to allow the host PC to, at all times, send audio data at the original sample rate of the audio being played. If the PC (or, more specifically, Windows XP's kmixer DLL) is not forced to do sample-rate conversion, it can maintain bit-transparent operation. "There is a tradeoff, of course," said Gwinn: "significant amounts of jitter arrive at the DAC1. This is not a problem for the DAC1, however, because Benchmark's UltraLock clocking system makes it immune to jitter."

See:

https://www.stereophile.com/headphones/108bench/index.html
----------

It might be down to how much computation a USB receiver can do cleanly. Supposedly, the latest XMOS can do async fairly quietly. But even older devices, like CM108, might be just fine given they only do the work of 16/44.1 and are supplied with clean power on Vcc pins with plenty of decoupling.
Like the long-used Buick’s 3800 engine. Overbuilt for the job. Might take a bit more fuel, but burns cleanly and runs smoothly.
========================

2023-04-14 7:07 pm
#68
The size of the input buffer of the USB receiver may be a factor. And why Gordon Rankin and others chose the old (2001) Texas Instruments TAS1020 USB device. It's even older than the CM108 (2003). Rankin supposedly wrote his own async software that was stored in a separate EPROM chip, and that was somehow interfaced with the TAS1020.
AudioQuest Dragonfly and other popular, commercial USB dacs also use this device.
But if jitter performance is what you're after, there are other ways to get that than the async vs sync USB debate.

EDIT: In the Stereophile Wavelength Cosecant review (2009), JA noted simply switching to a battery-powered computer can improve jitter of the asynch USB dac.
https://www.stereophile.com/content/wavelength-cosecant-v3-usb-digitalanalog-converter-measurements
----
Finally, at first I didn't measure as good jitter rejection from the Cosecant as I had expected to from the graphs that Wavelength's Gordon Rankin had sent me, and from the fact that it operates in asynchronous USB mode with the DAC controlling the flow of data from the host computer. Using the AC-powered Mac mini as the source of the 16-bit Miller-Dunn J-Test signal, while data-related sidebands were absent from the FFT-derived, high-resolution spectrum of the Cosecant's output, some sidebands were present at AC power-supply–related frequencies and there was some spectral spreading of the central spike that suggested the presence of random low-frequency jitter (fig.12). I retested the Cosecant using the battery-powered MacBook. Again, no data-related sidebands can be seen (fig.13), and now the central spike is well-defined, with just one pair of sidebands visible ±60Hz. This is the best jitter performance I have measured from a DAC fed USB audio data.
-----------
=================

2023-04-18 8:42 pm
#73
---
jean-paul said:
AFAIK the more recent Dragonfly uses a PIC and some have a SA9023. Never understood the hype as dongles are a plain nuisance. They are also more expensive than a full fledged DAC
-------------

Actually, I use a low-$ dongle DAC with my Win 7 Dell laptop daily. It's this one:
https://hifimediy.com/product/s2-dac/

ESS Sabre ES9038Q2M DAC chip and SABRE9602 output stage driver. SA9023 USB receiver chip

I improved it a bit with a cheap Amazon USB isolator.

===============

2023-04-19 2:55 am
#77
------------
MarcelvdG said:
You can usually choose between different FIFO lengths (page 17, right column AD1896 datasheet). The longer the FIFO memory, the more irregular the input sample rate can be without the ASRC having to switch to a wider tracking loop bandwidth to ensure that the FIFO doesn't become empty or completely full, but the more the signal gets delayed.

--------------
The "default" (0) is long. Short is (1). I think I want short because a modern USB Amanero device -- what I'm feeding the ASRC with at the moment -- is likely spitting out some pretty clean (jitter-"free") I2S. Yes?

==========
2023-04-19 5:07 am
#78
-------------
MarcelvdG
I think it won't matter much when you only use it for music playback with a clean source; who cares whether the music starts 1.1 ms later?
================

2023-04-19 6:00 am
#79
Some experiments with an SRC4192 asrc. To quote Han Solo, "Sorry about the mess."
Anyway ...because the SRC4192 and AD1896 are pin compatible, it was easy to "plug-n-chug".
Weird things happen when one precedes the TDA1305T with such devices. For one, using that variable Tekpower PSU, you can play with voltage and see what happens.
With the AD1896, the max usable voltage for the TDA1305's Vcc pins (before sound distorts, and cuts out at 4.7v) is 4.4vdc. The min. TDA1305's Vcc pin is about 2.3v (yes, it will continue to work well below its datasheet min of 3.4v!!) . That max voltage is strange for the 1305 because its nominal is 5.0.
The SRC4192 is much more fussy. Upon initial installation into that "mess", I almost though something was defective or mis-wired. No ... simply reducing TDA1305 Vcc from 4.4 to 2.8v got music playing. Looking at the 4192's DS, the block diag. does specify the use of decoupling caps on the analog and digital Vcc pins. Adding those (and adding some more 100nF to the 1305 Vcc pins) allowed me stabilize the mess and use up to 4.2 V on the 1305.
Even with all that extra tweaking of the SRC4192, the AD1896 (w/o tweaks) sounds much better and cleaner. Who knows ... with another DAC following these ASRCs, results may be completely reversed. Lots of possibilities in the parts bin ... AD1853 and 1955. Maybe, TI1792 or 1794. Or even CS4398.

Content below continues and updates "Revisiting ASRC (sample rate conversion) and up-sampling"
THIS IS THE LAST in the "Revisiting ASRC (sample rate conversion) and up-sampling" series -- see end of this post.

Others' comments per orig. DIYA thread is between " ---" with member username noted for ref. As shown here

------
Others' comments per orig. DIYA thread
-------

All other content is mine (philipsmarantz on diya; hollowman on Stereophile and elsewhen).

Indiv posts segregated by " ==== "

+++++++++++++++++++++++

2023-04-19 8:06 am
#81

--------------

MarcelvdG said:
What IO supply voltage do you use for the AD1896? The DAC is only guaranteed to work when the high input level is at least 70 % of its digital supply, 80 % for the clock input.
--------------

Both the divide-by-two daughterboard (dividing Amanero MCK to 11.289) and the AD1896 (or SRC4192) are supplied 3.3v from the Amanero ( its gpio 3.3v supply lines). This is the easiest way for me to get 3.3v to the AD1896 or SRC4192 need. And then heavily decouple that 3.3v on at various places on the ASRC.
---------------
MarcelvdG said:
Besides, the long unterminated wires and poor or nonexistent supply decoupling won't help.
---------------
Not sure what you mean by "the long unterminated wires" . Yes, the breadboard connections are not tight. And yes: lots of flying wires. But I am using thick solid core, as short as I can possible manage. The messy, 3D "flying object" arrangement is meant to keep leads short.

=================

2023-04-20 2:37 am
#84
------------
weissi said:
Please check your logic voltage levels - this is what the TDA1305 expects:
---------------
Right!
Amanero out (no /2) ;SRC 4192 (/2)
BCK: 1.64vdc; 1.43
WS: 1.64vdc; 1.43
DATA: ~0.82 ; ~0.88
MCK: 1.64vdc ; 1.33

All above: with with music playing.
The 1305 DS does note V_DD min = 3.4 v. Testing with digital display variable PSU, there is some wiggling room.
The ASRCs are in no way to exceed 3.6 for VDD_CORE, especially the SRC4192.
=================

2023-04-20 8:14 am
#89
I've tried to keep the messy ASRC experiment as all-things-kept-equal as possible. The contenders: a now-highly-modded Chinese dual (parallel) TDA1305 kit vs. the ASRC mess you've seen in my photos.
But with even in the mess with the SRC4192, the inferior of the two ASRCs tried thus far, I keep preferring the ASRC mess. Sounds cleaner and more dynamic. Maybe the superior Amanero has something to do with it ???

About the " logic voltage levels" issue ... The 1305 runs just fine even down to 2.7vdc (I am feeding both analog and digital V pins with the same Tekpower supply -- not ideal, but that's just breadboard convenience ). I'd still like 5v (nominal) on all the 1305's +V pins I can get away with. (I assume the logic voltage level limits are for the 1305's V_DDD digital supply voltage only). In other words ... keep V_DDD digital supply voltage (pin 10) at 3.3v; but, supply independent 5.0v to V_DDA analog supply voltage (pin 1) and V_DDO operational amplifier supply voltage (pin 28). Yes?

Bottom line: The ASRC seems to work some mystical stuff. Not sure how or why. Maybe keeping the MCLK out of that tiny SSOP device, where I2S signals are travelling molecules apart, is part of the dreamy magic

===================

2023-04-20 7:06 pm
#93
--------------
rfbrw said:
But is it better than a YSF210B or a CXD1144/CXD1244 nevermind a PMD100 or a custom solution ? I have a dac with one of these things in it and all it seems to do is smear itself over all that passes through it at the same as forcing the different sample rates of my many files into the same bland mess. Fortunately it can be switched out.
--------------------------
========

"Imagination Is More Important Than Knowledge"
Albert Einstein
Yes, but he was talking relativistically -- meaning in a contextual framwork ;)
========
rfbrw: We need more context.
Please provide more details about your setup.
What precedes those DFs you noted (YSF210, CXD1144/CXD1244, PMD100)? E.g., is it in a CDP where you have a decoder chip (which one)?
Is it in a DAC with a receiver and/or USB chip -- which one(s)? Or an Amanero?
And what follows your DF chips ... DAC chip, and its output stage and/or or i/v stage.
But too bad the min. specified for V_DD min = 3.4 v. It does work below that. It's only 0.1v to make it the ultra-common 3.3. I do have some fixed VDO at 3.3v. Including some goodies from Linear.
======================

2023-04-21 6:43 am
#94
Some major improvements to the messy project ...
TDA1305 now using a highly-regulated (see **) 5.0vdc on its V (analog) and V (internal opamp) pwr pins. Pin 10 (V_DDD) now a uses three-pin LP2950-33LPRE to bring 5.0v to 3.3v (ideal for I2S from incoming 3.3v devices). About that 3.3V for V_DDD. It seems that other DAC datasheets --PCM 1792, 1794, etc -- specifically mention V_DDD being fed 3.3v.
And, yes, the sonic improvements are quite significant.
** There is "new" protoboard in these images ... not so new as I built a few Flea's over 13 yrs ago, including the one in the photos... notice the 18.00V. Martin should recognize that

======================

2023-05-08 10:04 am
#95
In this thread, I've successfully implemented the AD1896 and SRC4192 asrc's. After quite bit of time tweaking, some very good-sounding success with the last of the ASIC-based ASRCs in my kit. This one, the CS8421, currently feeding a tda1543 from my DREAM Project test-bed experiment. Not using a downstream digital filter. The 8421 is essentially up-sampling at 2x.

Note the long, variable resistor. The 8421's Output Port Configurations are based on weird resistor values. See DS: Table 3. Serial Audio Output Port Start-Up Options (SAOF).
Note the DIP switch. No. 2 is for toggling between I2S and other modes. Working on an I2S to EIAJ solution for "Sony" format DACs using this/any ASRC alone. Format conversion using glue logic is messy!

======
These I2S to EIAJ converters work but are messy. (These must be a more elegant soln!)

=====================

2023-05-08 11:59 am
#97
= = = =
philipsmarantz said:
the CS8421, currently feeding a tda1543
= = = =

The CS8421 is significantly more sensitive than the other two ASRC's that I noted. For example:
+2.5-V Digital Supply (VD)
+3.3-V or 5.0-V Digital Interface (VL)
Note the to-92 small lp2950-33lpre3 regulator in the photos. This is a 3.3 v reg. part with a 20R in series to drop the 3.3 to 2.5V. I had no 2.5v reg so I had to improvise with a 3.3v part! Works well, but man is that VD/VL voltage difference critical.
Still trying to get another I2S dac -- tda1387 -- to work. And then the 1545 EIAJ. No luck with either those thus far. But 1543 sounds very, very good with no noise and clean separation. Do have some issues where signal not latching, but restarting the CD player fixes this. Again, the CS8421 is very touchy ... but when it works, it does so well.

=================

2023-05-08 12:29 pm
#98
--------------
jean-paul said:
Too much romance to my eyes but if you are looking for them: I have TDA1543 and TDA1545 (soic) in stock in numbers.

Never ever measure a TDA1543 DAC.
-----------------------

Check you inbox [jp].
===
If anyone has seen my DREAM Project thread, you'll note I can swap out DACs and DFs and ASRCs fairly quickly, into and out of the same Maganavox dream machine, with the Rudolf I/V.
About the 1543 ... I never paid much attn to it until recently.
The "Economy" description on the DS cover page did Philips no favors.
This DAC can sound very good and clean if you supply it with decent power. And it does very well in NOS or asrc or DF.
Now listening to Synchronicity, thru asynchronous 1543 ;)

============
2023-05-08 8:18 pm
#105

-----------
bohrok2610 said:
Why don't you just use left-justified output format with CS8421?

---------------
I can do that on-the-fly.

A I said earlier:
Note the long, variable resistor. The 8421's Output Port Configurations are based on weird resistor values. See DS: Table 3. Serial Audio Output Port Start-Up Options (SAOF).
Note the DIP switch. No. 2 is for toggling between I2S and other modes. Working on an I2S to EIAJ solution for "Sony" format DACs using this/any ASRC alone.

But, as I also noted earlier, the CS8421 also does not like another I2S dac like the 1387. I should be able to remove the 1543 and plug in 1387. I can do that with my other test jigs. The CS8421 is a touchy beast. It took me two days to get the 1543 mostly stabilized.

====================

2023-05-08 8:32 pm
#106

--------------
weissi said:
CS8421 would be ready for the task.
TDA1545A:
---------------
I tried that over day ago. Didn't seem to work. Maybe because of 8421 touchiness.
In any case, shouldn't it be L-J which presents MSB first?
https://www.diyaudio.com/community/threads/difference-between-true-i2s-and-converted-eiaj.327227/post-5542397

Another stange thing is that with the 1543 playing music, I can switch from I2S to L-J 16 bit via that DIP switch inserting 16.2kOhm (16.2 k ± 1% to GND Left-Justified 16-bit data). And the 1543, out of I2S now!!!, keeps playing. Change tracks. Keeps playing. I assume I2S and L-J are close enough for signal lock.
===========

This was my last post on DIY on this or any other topic. I (philipsmarantz) was banned 2023-05-08 from diyaudio.com for unknown reasons.

Which is Schiitier , Stereophile or TAS?

The Schiit Modi Multibit 2 is a decent multi-bit dac that is avail today for a reasonable price. It is the only Schiit product that I personally own and have experience with. Some of Schiit's orig design team were involved with Theta Digital. I used to own the Chroma 396 (non HDCD version) -- 1996 model. That was very fine sounding unit!

The Schiit Modi 2 is bit unusual in its power setup. It uses a 16VAC walwart adapter. "Unusual" in that AC is used -- most devices of this type use regulated DC adapters. And that "16v" is not a common AC voltage.
It just so happens that I had a high-quality Triad 16vac transformer in my parts bin. To that I added an Auricap across the MAINS 120Vac leads. Across the 16vac out, I added a snubber: good-quality Wima poly cap + Vishay metal 1/2 W resistor.
Good-quality US-made wire and plugs, obtained locally from tradit. electronics shop, also helps. Plenty of hot glue to secure stuff together. And the glass Mason jar further provides a layer of 'UL"-type security as I have yet to add a fuse and switch!

Sonics improve by a very noticeable amount.

Some DIY projects must be as lowest-effort-reversible as possible. So, I am not going to do any internal mods to either the Modi or orig adapter. I even have the orig box packaging from Amazon.

==========
TAS and its sister publ. The Perfect Vision had some of the ugliest cover art of any mag. No doubt Stereophile is the Schiitier of the two!

Have a closer look at these two AliExpress dual-PCM1794 kit boards. Both seem very similar, but one has prettier pictures; the other has a prettier price.
What do you think?
$82.00 ELNA caps (some electros)
https://www.aliexpress.us/item/2255800850044661.html

$26.00 WIMA poly caps (no electros)

https://www.aliexpress.us/item/3256804662071078.html

Closer inspection reveals more "subtle" diffs, like the $82 PCB having some decoupling at PS connections. And I2S having in-line R's, etc. And higher quality parts like those wafer R's. But still quite a steep $$ multiple of the other.
IAC ... for $26.00, that's two L/R 1792's with Wima caps, all ready to go with I2S input. Certain high-end CDPs and DACs only used a single 179x.
Hmmm ...

btw: Anyone know why the name "fever" keeps popping up in these kits? It is sometimes used for describing resistors, which might be metal fume deposit, or "intense". Not sure.

btw2: Currently own a blank ...

This is an actual photo of what I received:

https://postimg.cc/pmxFzzX0
Photo, complete scroll

... those are some of the manuf. whose various D/A processors and/or disc players I am trying to emulate in these experiments. The Arcam FMJ 33 cdp (reviewed by JA in July 2004) is at the top of my queue.

Squaring up the oscillator (clock) -- i.e., output the MHZ freq as close to digital-friendly square wave geometry as possible -- has been a goal of digital audio designers since the beginning. There are other freq-domain metrics (as well as subjective tests) that are important, too. But there's something pleasing about a nice, clean square wave on an old analog 'scope ...
To that end, I've been experimenting with cheap, small ways to generate common MHZ signals that look very clean and square on the old Tek analog 'scope.
I think I've found a possible "solution". It is in the "Y" output of an obscure TI driver device (SOT-23 pkg!!!) .
Have look ....
https://forum.allaboutcircuits.com/threads/xtal-oscillator-circuit-for-d...
======
That said, the squareness may not be just one goal to aim for ... there may be others as Martin and I discuss starting here:

https://pinkfishmedia.net/forum/threads/got-ticks-and-fleas.280281/#post...

I tested my teeny-tiny SN74LVC1G04-based 11.289 osc design with and w/o a very high quality op-amp based 5.0Vdc voltage reg (martin Clark's Flea)

Below: The teeny-tiny on left with a generic 11.289 xtal. For o'scope images, I used a 1989 Philips 11.289Mhz xtal. On right is a failed experiment with another diy osc. design.

Note that the Flea outputs 5.0 v, so the amplitude of the square wave is taller. But, despite the Flea's very high degree of regulation, the square geometry and visible waveform "fidelity" is very similar to TekPower 4.00 v by iteself.

TekPower (17.0vdc) -- > Flea (5.0 vdc) --> SN74LVC1G04 osc:

https://i.postimg.cc/tTYTSY2q/image.jpg

TekPower (4.0vdc) --> SN74LVC1G04 osc:

https://i.postimg.cc/W3dPbD3F/image.jpg

MC noted here: https://pinkfishmedia.net/forum/threads/got-ticks-and-fleas.280281/page-2#post-5028424

======One should also be able to explain complex ideas in language simple enough that anyone can get it - so with the crayons out, here is a little illustration of the only quality that matters in the clock output, down to 10s-of-pS deviation as I wrote above: this is what 'jitter' is in a clock signal:

Any o'scope internal timebase performance is c. at least 10^4 x worse than the clock in this regard, so it cannot be seen on a scope directly. The crests of the clock signal, high and low, are way above/ below the slicing level where the dac converts the signal (which are usu c 0.6v from each of high and low voltages. Which is what the shape at the extremes, really doesn't matter.
- Now go read the earlier post above, again, with this illustration in mind, until you get it.

======
I replied:

Timebase performance can be somewhat sidestepped using FFT and freq. domain analysis ... i.e, SA. One can squeeze out pS jitter data ... and audio engineers have been do so for decades now.
I can even do it with a cheap TinySA. The speed isn't as great as an old, analog HP or modern APx555. But one can watch the spikes move up and down in real time. If anyone wants, I can shoot a video and let you see with your own eyes.
Meanwhile ...
The tiny osc, using 4.00v TekPower.

Philips 11.289 xtal:

Generic 11.289 xtal:

If you're okay with a soldering iron and basic DIY, this may be an option for some decent low-$ (less than $19) improvements in USB or other I2S interface.
This is not "re-clocking".
https://i.postimg.cc/zGkTLHRB/S932af0fb9bf948d68f9624e0c0a0500fa.jpg

The Analog Devices basis part is here:

https://www.analog.com/media/en/technical-documentation/data-sheets/ADuM4160.pdf

It has been around since 2009 and is a bit $$ by itself (Digikey, Mouser). Ali has them for much less $.
Note its architecture:

https://i.postimg.cc/PqrVNjJg/Screenshot-at-2023-06-28-11-45-35.png

There may be better choices for this part. Unfortunately, Potato Semi is no longer active.
==========
Edit: Better choices ... that AD device seems a bit slow and is NOT the device on the LHY adapter (that uses a TI part).

Some alternative to that digital isolator IC chip, and they are faster than the AD:

TI: ISO7240

NVE: IL715E

Up to 100Mbps
==== Below: from pavouk =======
Circuit description

Circuit use specialized circuit IL715E which have on the one side four TTL inputs and on the opposite side corresponding outputs. It perfectly fits to I2S bus, where we have DATA, BCK, LRCK and SCK signals. Circuit requires dual supplying, one on the input and second on the output side. On the input is supply connected to +5V which I bring out them on the I2S connector for future use. Secondary side is supplied from small DC/DC converter. According to recommendations are around converter inductors L1 and L2 and filter and block capacitors. Their values correspond to changer from Aimtec company but probably parts will be acceptable for changers from other companies. For reliable function of converter is on the secondary side load resistor R1 which guarantee minimal current consumption about 20mA. Datasheet have informations about maximum data rate 100Mbps which is far away from our sample frequency 48kHz, where frequency of BCK is 3.072MHz and SCK is 12.288MHz with 4x oversampling. Circuit should work with 192kHz/24bit samples where frequency of SCK is higher 49.152MHz. DAC without oversampling doesn't use SCK signal that we must care only maximal BCK frequency. It is possible to use isolators from other vendors with same pinout, but we must check their maximum data rate and if they doesn't need ENABLE signal.
==========
See: http://www.pavouk.org/hw/modulardac/en_i2sisolator.html

I purch'd some new-old stock Raltron 11.289 Mhz crystals from eBay at a fairly low $ for a 20-piece pkg.

And those were tested in sn74vlvc1g04 datasheet ckt (first SA image below). The other images are from a new Chinese $24 clock that is quite clean in both TD and FD. And yes, both clocks using same PSU, the $24 clock is cleaner than my diy sn74vlvc1g04 / Raltron 11.289 Mhz crystal clock.

PLEASE NOTE: The Stereophile forum image projector crops off images of certain wide aspect ratios. Use postimg URLs to access entire image.

Example of one of the cropped images that can be shown, smaller/uncropeed, as a thumbnail here -- this is what "uncropped" means:

===================

diy sn74vlvc1g04 / Raltron 11.289 Mhz crystal clock

https://i.postimg.cc/J4WjxCJS/image.jpg

==============

Chinese $24 clock

https://i.postimg.cc/pXYz1cJr/image.jpg

https://i.postimg.cc/W4tZzNLq/image.jpg

https://i.postimg.cc/7ZK7qCcy/image.jpg

https://i.postimg.cc/jSxPV6MX/image.jpg

Note that a mid-priced Siglent SDS 1202x-e digital o-scope shows a strange spike where the old Tek 465 analog o-scope (above) does not show the spike.
https://i.postimg.cc/YSVWm5Mn/image.jpg

USB isolators come in quite a wide variety of packages and price points. Some are industrial units that only filter noise, but do not necessarily improve sonics. That's odd and somewhat confusing !
Perhaps, the fastest way into thee potential of USB isolators is via these YouTube links:

https://www.youtube.com/watch?v=pcDB_wc_llc

https://www.youtube.com/watch?v=jruWfqxApT0

I personally use a $15 Amazon-sold HiLetgo ADUM3160 B0505S 1500V USB to USB Voltage Isolator Module; Supports 12Mbps / 1.5Mbps .
It uses an Analog Devices ADUM3160 $6.00 usb iso chip.
Both Analog Devices and Texas Instr. (among other manufs) produce a series of IC usb iso chips that vary slightly in price and (the latest) are capable of SIGNIFICANTLY faster operations (480 Mbps).
Curiously, the very $$ USB Reinigen $1,650.00 uses the $9.00 AD ADUM4166.

Other refs:

https://www.youtube.com/watch?v=OReJfx74bTE
https://www.youtube.com/watch?v=2zRp0tAW_vs
https://www.youtube.com/watch?v=HCpkb9gALWQ
https://www.youtube.com/watch?v=RulAcLrnPkA
https://www.analog.com/en/parametricsearch/11032#/

It should be clarified that there may be completely different products (and applications) for isolators and filters.
The AudioQuest JItterbug is only a filter; it has no logic IC.

Both JA and Amir (ASR) found no MEASURABLE effect of the AQ device.

https://www.stereophile.com/content/audioquest-jitterbug-usb-noise-filter
https://www.youtube.com/watch?v=RulAcLrnPkA

HOWEVER ... Jonathan Novick, Audio Precision has some thoughts here: "RMAF15: What The Specs Don’t Tell You… And Why"
https://www.youtube.com/watch?v=2V6YN-mshmY

Meanwhile, products like the $50 Hifime High-Speed USB Isolator v2 does isolate via an IC.
It is based on a Silanna ICE USBB chip for data lines and an isolated 5V-5V DC-DC converter for power.
Silanna press release: http://www.prweb.com/releases/2015/10/prweb12995871.htm

Also note that ONLY since 2021 has there been a MASSIVE, COST-EFFECTIVE jump in isolation speed from 12Mbps to 480Mbps (!!). Eg. ADUM4160 to ADUM4166. Look for such chips in the specs of your iso device.

REFS:
https://dranwarali.com/2022/03/10/usb2-isolators/
https://www.eevblog.com/forum/chat/high-speed-usb-isolator/25/

https://hifimediy.com/product/hifime-high-speed-usb-isolator/
https://www.gieseler.com.au/product-page/usb-reinigen
http://www.pavouk.org/hw/modulardac/en_i2sisolator.html

USB isolators, I2S isolators, re-clocking, up-sampling and oversamping

Some of these are low-cost, relatively low-complexity chip- (IC-)-based tweaks that can be used before the DAC chip.

In the posts above, I have mentioned (and or shown my own projects implementing them); all of these "magic tricks" diyers and manufs use to affect sonics of D/A processors, disc players, streamers, etc.

Some quick examples of IC that can had for low cost (see above posts in this long thread for more details):

USB isolators: ISOUSB211, ADUM4166 (480Mbps)

I2S isolators: ISO7240, IL715E, ADUM1400BRW

re-clocking: 74 series logic ICs; see: https://www.diyaudio.com/community/threads/re-clocking-i2s-simple-version.396381/

up-sampling: src from Cirrus/Crystal, and AD: SRC4192, AD1896, CS8420

oversamping: built-in to many chip dacs, but older ICs like SAA7220 and DF1704/6 and PMD100 still avail on eBay and Ali

The 1305 caught my eye when Naim used it in some of their CDPs in the late 1990s to about 2002. They only use a single stereo 1305 and follow it up with a 7-pole opamp output stage, along with their overkill PS and regulation design.
The DutchAudio list ...
https://www.dutchaudioclassics.nl/the_complete_d_a_dac_converter_list/
...shows several "high end" manufs using the 1305, like Conrad-Johnson, Micromega, Mission, Pink Triangle, Roksan, Struder, Sugden, and even a 2010 AMR model:
6moons audio reviews: Abbingdon Music Research CD-777
So all those manuf using the 1305 aroused my curiosity even further.
In mid 2020, I ordered a new Denafrips Ares II (discrete R2R), and also a few other Ali dacs (chips, pcbs or whole units) like that cheap $35 "dual parallel " 1305 kit. Even un-modded, the 1305 had excitement and "PRAT" that the $800 Denafrips simply lacked. Modding the 1305 kit improved sonics much further. And going back to the Denafrips was boring. Even though Ares II and many other DACs excel in some sonics areas, pace-n-rhythm isn't it.
Yes, for the time, the 1305s might just be the best DAC for the price on Ali.
==========
Question from a friend:
You're in possession of the Denafrips so have you examined its power supply feeding the discrete resistor ladder part? That would be my first investigation for what renders it impotent in the PRaT department. Based on TDA1387 designs, getting the lowest noise rails, especially at LF, is required for the rhythm department to work.
===========
My reply:

Yes, it's important to keep that PSU in mind when tweaking for PRAT. I think many UK designers -- esp. Naim -- are known for tight, driving rhythm and dynamic drive. One look at the their PS and reg sections can suggest how they get that timing thing down so well.
Not sure I wan't to invest too much into the Ares II. I am tempted to mod it, but I want to keep it stock because that is a commercial reference; I can easily toss it back into my system and remind myself that those who have promoted the Ares II --- many influencers and audio journalists -- are simply unaware of what its possible with good design, such as what DIYers routinely create for very little cost.
Also: there is something about the discrete R2R topology, Soekris et al, that just doesn't feel right to me ... I think I have commented on this on DIYA. Including overlong traces, Johnson noise in all those R's (that must be at optimum temp), etc. I am a bit more hopeful about DSP (fpga) d/a or signal processing. But, having worked in the manuf industry (electronics and PLC), dedicated/ad hoc ASIC chips devices were always optimal.

A friend noted:
Re: Accuphase with 32 PCM63s inside. He sent me the service manual, I barfed at how it was implemented, what a waste of classic chips.
=====
https://www.stereophile.com/content/accuphase-dp-90-cd-transport-dc-91-d...
=====
My reply:

It's funny you mention Accuphase ... their US hq was less than mile from my house here in Gardena, CA.
In some ways, I can empathize with Accuphase ...
at the time they created their Multiple MultiBit (or MMB,) design, late 1994, the PCM63 was not "classic" and not the at inflated price or status it now commands. Also, Accuphase is a very high-end brand and that D/A processor cost $13,495 (1995) -- so they could use that many 63's at that sticker price. And given the few chips and design topologies they had avail AT THE TIME, I think Accuphase might have picked a viable route.

About urge to mod commercial gear ... yeah, I used to hack models up to the point of profanity. Search for my Magnavox Mess project (a CD650 with an outboard Naim-like PSU). Wow, did this sound good, but that CD650 did not survive the ugly stick beating. And the ugly stick lost!
Now, when working with vintage / stock models, e.g. Philips CDPs, I try to make everything as reversible and modular as possible. Not that I want re-sell or eBay the unit ... just the challenge of plug-n-play reversibility is more fun nowadays. The unit should look and operate as OEM as possible, but with improve sound.
For original veroboard projects, they are a complete mess -- as you may have seen from my DIYA posts. Dead-bug designs, flying wires and verobaords in the most hideous, twisted configs imaginable. Not really concerned about appearance; I simply want to prototype, plug-n-play, quickly, with good-sounding results, and then move on.
If the messy vero proto project sounds and measures "good", then I'm okay with JUST that -- the project can always be upgraded to "excellent" with better caps, tighter reg. rails, etc.

If you haven't heard about a possible magic trick to getting certain headphones (or IEMs) to sound spectacular is to power them with loudspeaker amps.
https://www.youtube.com/watch?v=YtyCRXizSeM
This is not the same as Stax adapters for amps (that sound great!).
Indeed, some major manufs may have known about this magic trick as far back as the mid 1970s; important POWER amps like JVC JM-S7 (V-FET, 1975-77) had a dedicated "PHONES" jack on the front:

http://www.thevintageknob.org/jvc-JM-S7.html

What I want is something smaller and lighter. Maybe one of the magnetic field amps -- aka phase control amps -- that began appearing in the late 1970s. Of course most have heard of the Bob Carver line. But there were other players, including Soundcraftsmen and Yamaha:
http://www.thevintageknob.org/yamaha-B-6.html

... I do own that beyer dt880 ... should be interesting ... to be contiinued ...

My dream project, noted earlier, is not as unique as I had thought. I think I did mention the Counterpoint earlier in this thread. And there as also a CAL dac that came out in late 1991/early 1992. Stereophile reviewed it in Aug. 1992.
See: http://www.wghwoodworking.com/cal/cal.html

Note from the PDF review in that link, there are four plug-in modules, much like computer PCI sound cards. The most $$ is MASH V which adds a whopping $1500 to the cost. The PCM63 "Indus" module is only $650.00.

The controversial, portable Optimus CD-3400 (1993/1994) is well known in audiophile circles. I personally have owned two units, and as well as its successors cd3450 and cd3470. All can be tweaked using external power supplies and Sorbathane feet. Even CD mats.
These units seem to employ a secret weapon: the NPC SM 5871 (digital filter and delta-sigma dac combo).
I found a good deal on eBay recently from Singapore. Six pieces (3 in 28-pin dip; 3 in 28-pin SOIC) for $49/00.
Should be an interesting DIY project, especially with better output stage (tube or SS).

BONUS IMAGE---what is this?

Yes, the CD3400 was an important PLOT DEVICE in a novel called Stereophile.
After Sam Tellig set up the initial controversy (which was surprsigly pub'd by the Novel), something unique happened. Several Stereophile reviewers, including JA, followed up in July 1994 with a formal subjective/objective review. Especially intriguing is JA's Measurement followup in 2010 when he re-measured the 3400's SPDIF jitter performance with a surprising discovery.
========
Radio Shack continued to evolve the 34xx series from 1994-6, with even better-sounding players with better portable performance to boot. But the Zeitgeist that owned the 3400 was long gone.
=========
About that NPC df+dac chip ... if one data-mines the master digital audio-gear chip list here ...
https://www.dutchaudioclassics.nl/the_complete_d_a_dac_converter_list/

... and search for "sm587x", one finds about 36 hits. Early to late 1990s. NPC continued to evolve the sm58xx line with sm5877, sm5878 and SM588x. See:
https://www.alldatasheet.com/datasheet-pdf/pdf/91113/NPC/SM588X.html

Archive.org also has records of NPC product pages from late 1990s-early 2000s with the full lineup and descriptions. By far, NPC had much more success with very popular "SM" series monolithic interpolators (digital filters). And very small success with monolithic (stand alone) d-to-a converters.
Not sure why NPC abandoned audio/video component' their last component was an ASRC.
Maybe they sold their designs to AKM, etc. ????

This obscure chip series used mostly in Kenwood models mid 1990s to early 2000s.
CDP's using an KAN0x chip(s) often had 24 Bit "Fine D.R.I.V.E" on front of device. I've searched for any reference to what D.R.I.V.E might be an acronym to ... found nothing except:

http://marlene-d.blogspot.com/2012/10/review-kenwood-dp-5090.html

Kenwood D.R.I.V.E (Dynamic Resolution Intensive Vector Enhancement)

https://audio-database.com/TRIO-KENWOOD/player/dp-5090.html

You can see some refs at the usual: https://www.dutchaudioclassics.nl/the_complete_d_a_dac_converter_list/

Use the search string: "Kenwood KAN0"

Present in the SM schematics, too, of models like model Kenwood DP -7060, page 2 (block diagram, KAN02). It seems to be quite an intense chip: qfp size with 44- 60 pins. When a manuf discloses so little about a certain part, it often means the manuf. may have ... ahem ... lifted the design from an OEM that has IP/patent claim. Kenwood is no electronic giant like Philips or Panasonic or chip manuf like AD or TI or even NPC. Because of the size of the KAN0x chip (60 pin qfp), it might very well be an FPGA/ DSP with some sort of custom Kenwood "DRIVE" firmware. Possibly a custom interpolating DF.
In any case ... Very intriguing how Kenwood implements this KAN0x device. Sometimes it is labeled in the schematics as an L.P.F.
In the Kenwood dpf-3030 cd player (year 2000) schematics, KAN06 is a 44-pin qpf chip that is placed just before the AD1855 DAC. It is indeed labeled "DRIVE" in the schema.

The 5090 uses KAN03 in a completely novel way ... in conjunction with several 74hc logic chips to construct a custom DAC ?????? --- see page 5 of schematic.
Anyone know more about "Fine DRIVE"?

================

Yes, indeed the world is all about appearances and illusions ....

... .take that stunning-looking CDP from #1 ...

Incredible! Kenwood went the whole hog in the next model up from that: the DP-7090.

A looker!!
It even got Lampi's attention (late 2006ish), but weirdly , Lukasz makes no comment about its unique 3-chip digital filter D.R.I.V.E. "serializer" ; only the 8x PCM1702 dacs that are used after the 3xDF. Ahem: Kenwood's D.R.I.V.E (Dynamic Resolution Intensive Vector Enhancement) -- IOW, a digital laxative or stool softener.

Another weird strategy is that after all that esoteric DFing and DACing, Kenwood does the i/V output with lousy opamps. Lampi ditched the opa's using two different methods:
SOWTER xformer; or: simple R for i/V.
So, no typical Lampi tubes here.
For the SOWTER output, the 'zator pooped this comment:

"After lampization - it is in top league. Absolutely stunning. Jaws dropping, drooling, etc etc. You know the story. The PCM1702 DACs used in R conversion mode are of similar quality as the TDA1541A, AD1865, AD1862, PCM63K and other finest I out DACs."

http://www.lampizator.eu/lampizator/references/Kenwood7090/Kenwood%20CD%207090%20CD%20player%20lampizator.html

Lampizator said: "Highly recommended. The sound is so refined, detailed, analog-like, that the CD playback equals the best vinyl from MC cartridge and tubed phono stage."

Make of those over-the-top statements what you will. BUT: This was well before Lapmi went commercial. Alas, human "temporal effort" has brainwashing effects. So, regardless of profit motive, if the darn thing JUST works, after all that sunk-cost investment, the sleepwalking zombie will defend it to the hilt.

Links to images:
https://i.postimg.cc/MpTF0p9k/dp-5090.jpg
https://i.postimg.cc/mg56mWnP/dp-7090.jpg
https://i.postimg.cc/yYnGSZXX/Screenshot-at-2023-09-13-22-55-21.png
https://i.postimg.cc/Dz4pDYVB/Screenshot-at-2023-09-14-20-55-08.png

Several Sony digital devices in the 1-bit "Pulse" era (roughly, 1990-2000) were favorably reviewed (and measured) here at Stereophile.

Indeed, Sony put out so many models that only lists, like noteworthy DutchAudio master DAC/chip list, can hint at the scope of Sony during the CDP "Golden Age" (an age with huge R&D budgets that were funded by the success of the CD format).

https://www.dutchaudioclassics.nl/the_complete_d_a_dac_converter_list/#S

Technology often goes thru phases ...

-- early (built like a tank, but reliability and performance still in less-than-ideal "immature" stages);

-- middle (Golden Age, very mature, maybe optimal in all respects: sonics, reliability, build quality etc);

-- late (present/now; engineering enthusiasm wanes, economic factors cheapen parts and can negatively affect sonics, reliability, durability. E.g., it can be argued that Delta-Sigma DACs are cheaper to produce than R2R; hence, that was a MAIN reason they were developed. But D-S can go thru their own phase/stage cycle, so MAYBE new AKM and ESS dacs are decent ).

Towards the late middle of the Golden age (mid 1990s), Sony was losing enthusiasm, but still had to keep improving the sonics of their CDP projects.
The result of low-effort (lazy) designs included a strange chip, "CXA8042AS", that was a mini amplifier of sorts [no datasheet as was typical of many mysterious Sony devices; different from Philip's comprehensive SMs , datasheets and schematics. When tech specs or datasheets are "missing" it often means IP theft or obfuscation ].
The inclusion of this chip was not mentioned in reviews, like TNJ's 1996 sony_cdp-xa7es_cd_player review, but this ES series Sony CDP was very highly rated:

https://www.stereophile.com/cdplayers/sony_cdp-xa7es_cd_player/index.html

Below is a segment from an ASR thread on the topic of DSD...it reveals some more on current-pulse (I'll follow thru soon in a subsequent post in this thread).
============

Sony:

On the other hand, Sony, the other leading player in DSD, used a different 1-bit system from Philips, called PULSE DAC, at about the same time. It started in 1990 with the CXD2552 chip (used in the CDP-X777ES). The method is the same as for Philips, first oversampling 44.1kHz to 352.8kHz with an oversampling chip, then pulsing to 1 bit with a CXD2552 chip, but while Philips uses PDM (pulse density), Sony uses PLM (pulse length) to indicate amplitude.

Four years later, in 1994, Sony introduced a bridging "current-pulse DAC" system, in which the voltage pulse output from the CXD2552 is converted into a powerful analogue current pulse by a subsequent chip.
When it first debuted, it looked like a gimmick, but now it seems that what Sony was trying to do was a "direct amplification of 1-bit pulses". In other words, for Sony, 1-bit is not just a digital data format, but a neutral entity that treats digital as analog.

The S-MASTER amplifier, which has since been used in Sony's AV amplifiers and high-resolution Walkmans, is in principle [???? ed.] the same as a current-pulse DAC, with the digital data first converted into PLM voltage pulses, and then driven directly to the speakers or headphones by a powerful current switch. That's how it works. That's where Sony's tradition comes in. [Note that CXD2552 converts to analog before the "CXA8042AS" device].

Unlike Philips, which supplied DAC chips to many audio manufacturers, large and small, Sony's DAC chips were almost exclusively for its own CD players, so their performance is still largely a mystery.

Around the time of the arrival of SACD, Sony, like Philips, ceased to develop dedicated DAC chips. The first Sony SACD player, the SCD-1, used Sony's CXD8594 DAC chip, but from the third generation SCD-XA5400ES, Sony began to use Burr Brown's DSD1796 and other DACs, as did Marantz. The sound quality seems to have changed dramatically during this transition. In 1999, the 2.8MHz 1bit DSD system was a proven high-quality technology in 1bit DACs for CDs, but since then, DAC chip manufacturers such as Burr-Brown and Cirrus Logic have been developing even faster DACs with 128x and 256x oversampling, as well as 4-8bit DACs instead of 1bit. And as a result, 2.8MHz DSD adopted for SACD has been treated as "a format with 'halfway between PCM and multibit' sound quality born in a transitional period". For example, the advanced current segment method used by Burr-Brown since 2003, just after the debut of SACD, combines the best of both PCM and 1bit DAC, with purely resistive conversion for the top 6 bits of 24bit data and high-speed 5 bit delta-sigma modulation for the bottom 18 bits. It's the best of both worlds.

=================

" tired light" is explained in this Wiki entry:
https://en.wikipedia.org/wiki/Tired_light

"glass" (in this context, I am using to make catchy term!!) is: "amorphous structure of glassy silica (SiO2) ", as used in the semiconductor industry.

What I want to explore is whether the actual (physical) gap in semiconductor devices -- between layers of insulation and conduction or between conductive traces and components -- has any effect on sonics? Put simply, does a larger chip (say DIP vs SSOP) sound different, all else held equal? Is this why a discrete output stage (say of a DAC) is preferable to one that uses opamps?
is "breathing room" for electrons and molecules more/less important than speed-of-light factors (e.g., an oamp has shorter trace lengths; hence, should THEORETICALLY be faster and more precise).

Stay tuned ....

... or hand-brain vs computer-software (CAD).
Which might work better ... for sonics, measurements, objective vs subjective metrics, or long-term reliability/durability?

The above photo is from old textbook (mid 60s) on the topical subject. Nota Bene (Jim Austin) EXACTLY what the Figure caption scripts.

Sometimes, us DIYers, and/or repair techs, run into an all-too-common issue. No datasheet, no service manual or schematic; or certain parts (often ICs) used in models (incl. several models from major mainstream manufs), with no datasheet or description, just a part number and, possible, represented in the schema diagram, but just that w/o elucidation . Or there is schematic or SM ... but when it comes to certain components, there are no details or specs.

Smaller high-end manufs (Bryston, Bel Canto, PS Audio, Ayre, Naim, etc) -- more often than not -- produce very hard-to-obtain manuals, or very low quality schematics (sometimes, hand drawn!). They are likely afraid of copy-cats ... or it could be something even big players (like Yamaha, Sony, etc) are up to ... hiding something ... such as possible "theft" on THEIR part of patented topology and stolen designs. It's easy to hide that in a complicated device such as an IC. Sometimes, manufs -- Philips, TI/Burr-Brown, AD -- will go exactly the opposite way, with very detailed datasheets and even Theory-of-Operation sections; this indicates that the manuf probably did play by the rules and are proudly showing off their work.

In my DIY research, I came across particular and specific example where Yamaha* may have incorporated some ICs that are not their entirely own. [* I noted Kenwood as another obfuscator, in another recent AudioKarma thread.
https://audiokarma.org/forums/index.php?threads/kenwood-cdps-with-mysterious-d-r-i-v-e-kan02-to-06-chips.1039480/
Interestingly, in that AK member cpt_paranoia himself attempted obfuscation.
https://audiokarma.org/forums/index.php?threads/kenwood-cdps-with-mysterious-d-r-i-v-e-kan02-to-06-chips.1039480/#post-16626080]

In particular, DAC chips such as:

YAC501
YAC506
YAC507
YAC514

All from (roughly) the early 1990s; all w/o datasheets TTBOMK. And Yamaha on used them in their OWN cdp's; unlike, say, Philips, NPC or Burr-Brown.

For example, the CDX-860 (1993).

This cdp uses the same elusive chip-set as the ultra-high-end GT-CD1 and GT-CD2. YSF210 (digital filter) + YAC507 (dac)

"I-PDM"

What is very odd about the YAC506/507 is its weird "I-PDM" topology: AFTER the YAC chip are two (per channel; four total) TC74HC00AP's, Quad 2-Input NAND Gate, and these two logic chips feed a typical I/V (output) stage. This is super bazaar -- never seen anything like it nor could guess how the DAC chip and 74 chips work together.

https://audio-database.com/YAMAHA/player/gt-cd2.html

========= GT-CD1/2========
The D/A converter uses Yamaha's original I-PDM 1-bit DAC and secondary noise shaping.
Accurate D/A conversion is achieved by combining the I-PDM method, which achieves accurate analog conversion even if the digital waveform is not ideal, and the secondary noise shaping method, which ensures a high dynamic range while reducing out-of-band noise to a low level.
In addition, by using DACs with a processing capacity of 2 ch in one chip independently on the left and right sides, mutual interference can be prevented, positive and negative phase signals can be output, dither components are completely cancelled, and random noise is reduced by 3 dB.

The audio section is composed of discrete elements composed entirely of transistors. It has differential-input, complimentary service output-class A operation.
Audio parts used in GT-CD1 are used for parts.
=================

The yamaha-cdx-660-cdx-860-cd-players were the ONLY other models (TTBOMK) to use that chips-set. Especially elusive is YAC507 dac chip. I think it was only used in 3-4 Yamaha models:
See:
The complete d/a DAC converter list - DutchAudioClassics.nl
www.dutchaudioclassics.nl www.dutchaudioclassics.nl
The DF , YSF210, is a SOP24 device from late 90s. It’s readily available on Ali and eBay.

What is very odd about the YAC506/507 is its weird topology: AFTER the YAC chip are two (per channel; four total) TC74HC00AP,
Quad 2-Input NAND Gate, and these two logic chips feed a typical I/V (output) stage. This is super bazaar -- never seen anything like it nor could guess how the DAC chip and 74 chips work together.

above: YAC507 _ 74hc001

Links to raw, full-frame images above:
https://postimg.cc/VdqkQscj
https://postimg.cc/gnF2TX1z
https://postimg.cc/ts9CX0X8
https://postimg.cc/rDgyshss

Below are schematic (and SM) image captures from a Yamaha CDX-1050 (1991). It uses a very similar topology to the one described above (use URL links to see the full-frame images)

https://postimg.cc/rd3kR7kV
https://postimg.cc/Vrtcq4fj
https://postimg.cc/7JM8c6Th
https://postimg.cc/0Kq1cmjh

-------------------

Another weird chip from Yamaha is the YAC514, which has no datasheet nor can it be found on Ali or eBay. It is one of those mysterious 64-pin chips that was used in several Yamaha cdp’s in the 90s. See that dutch audio list.

Not on that list, but an ssop24 dac from Yamaha, is the YAC516.

https://datasheetspdf.com/datasheet/YAC516.html

Cheap and available on Ali and eBay -- I have several in my parts bin. But I have yet to find a player or DA processor to employ the yac516. Weird!

above: YAC514

Some- where/when/how/why in the Universe, all this makes sense.

I attempted to post this on pinkfishmedia.net (UK audio forum) but forum admin (Tony L) deleted it twice for reasons he claims violated PFM's AUP. I have been a member or PFM as "13mh13" since Oct. 2007. And have FREQUENTLY run into the same micro-moderation issue (forum mods have too much time on their hand; or they need to be active to justify their normally-boring lifestyles). Also, certain forum communities are existentially sensitives and easily (personally) offended when their lack of knowledge is put to the test. In any case, ...
==================

JVC's K2 digital and mystery chip

Probably Martin Clark has some clues about what JVC might have been up to in the early '90s. Their K2 technology cd players got some decent praise in HFN and Stereophile.
Looking that that reference link of digital dutch insights ...
The complete d/a DAC converter list - DutchAudioClassics.nl​
... we can see some of what's goin' on ....
JVC XL-Z1050 (1992) decoder dsp --> JCE4300 [mystery chip] ->> YM3414 [digital filter] ->> JCE4501 [dac]
JVC XL-Z674 (1994) decoder dsp --> JCE4300 [mystery chip] ->> MN35502 [df+dac] -
The mystery chip, JCE4300, is placed right after the decoder chip.
The service manuals for various JVC cdp's only give very sketchy details about this 20-pin "Gate Array for K2 Interface". Obfuscation? Methinks it's a 74-style logic chip being used to re-clock the I2S lines. Typical practice. But, then again, it is a 20-pin device, maybe like: SN74AS804BN. So Uh, Martin, when you're done with the Dewars, you might want to un-embarrass yourself with a home run ;)

[edit: use URL for full fame images]

https://i.postimg.cc/vT9L0C4n/Screenshot-at-2023-10-07-22-01-02.png
https://i.postimg.cc/0QzCwzxc/Screenshot-at-2023-10-07-21-58-27.png
https://i.postimg.cc/qMVGNBFx/Screenshot-at-2023-10-07-21-57-23.png
https://i.postimg.cc/Gp7QtF04/Screenshot-at-2023-10-07-21-56-59.png
https://i.postimg.cc/VsbRs0KN/Screenshot-at-2023-10-07-21-56-02.png
https://i.postimg.cc/yd2yX78B/Screenshot-at-2023-10-07-21-55-21.png

I OCR'd the service manual and, maybe for the first time on the internet, and exclusively for the Stereophile group, some insights into K2.

[For figures, use the URLs to view full frame image (below). The messy OCR will be cleaned up soon, and this post will be added to with more pages from the service manual....a work in progress!! To BE CONT. ....]

Description of New Technology

New K2 Interface

1) K2 Interface

The K2 interface has been designed to eliminate digital distortion occurring in digital audio processing.

2 ) K2 Interface in the Digital Audio Application

The digital audio application consists of digital and analog signal processing sections. In the digital signal processing sect ion, the signal load fluctuation continuously changes depending on disk data input, but also on other conditions. Whereas, the analog signal processing section always performs a constant operation on the input digital signal waveform. However, both sections are connected for digital signal transmission, even if the power source radiation is separated so that a non-correlation component (sound quality factor) generated by the load fluctuation of the digital signal processing section is transmitted to the analog signal processing section. Consequently, the analog signal section is always adversely influenced by the quality of the digital signal processing section (Figure 1).

Intrinsically, the purpose of the digital signal processing section is to read binary data from a disc and transmit the data without error . The waveform distortion, which is the non-correlation component and the jitter component are added to the digital signal transmission waveform. Although the processing circuit should not be affected by any load fluctuation, it always adversely affects the sound quality and necessitates redesign of the sound-generating procedures. The K2 interface is the best design method for solving these problems (Figure 2 ).
=================
Figure 1.

Conventional digital signal transmission mode

Figure 2. Binary transmission mode using K2 interface

==========

3 ) What is the K2 interface?:

It is a mode in which the binary 1 and 0 from the digital signal, including the non-correlation component, are read using a separate reference clock signal and a new digital signal is generated in a block (analog signal processing section) completely separated from the first block (digital signal processing section)

4 ) K2 interface configuration: The K.2 interface operates according to the following design:

(1) Separation between block processing: The transmission block (first step) and receiving block (second step), including the power source and grounding, are totally electrically separated;

(2) Sign transmission: The receiving block has a unidirectional transfer element, an optical coupler, for receiving binary from the transmission block, and has an extremely fast timing control circuit for completing the reception (Figure 3 ).

(3 ) Timing data transmission: A uni-directional transfer element transmits the timing data between the timing control circuit and the transmission block.

Figure 3. K2 interface diagram

===

5) K2 Interface Operation: Figures 4 and 5 show the circuit configuration and binary transmission timing chart, respectively. The crystal oscillator used as the synchronous reference is placed in the analog signal processing section, in which all synchronizing signals and timing control signals are generated. The synchronizing signal is transferred to the digital signal processing section via transfer element B. The binary data from a disc is transmitted as a digital transmission signal ® to the analog signal processing section via transfer element A.

It is assumed that a current flows to the LED of the transfer element A when the logic of digital transmission signal @ is 0, and that binary det ect ion switch G is closed only when the logic of binary det ect ion timing switch © is 1. Therefore, only if the logic of digital transmission signal © is 0, does current flow to the transfer element A phototransistor when binary detection switch G is closed.

As a result, the binary data of the digital transmission signal can be transferred with influence of the waveform distortion component for only an extremely short time in which the binary pattern is detected. The binary data transfer is not affected. Binary output ® is generated as digital receiving signal © in the D flip-flop.

6) K2 Interface Effect

By performing the D/A conversion of the ideal digital signal generated without a "non-correlation component " in the analog signal processing section, the sound quality alteration fact or can be eliminated prior to the transmission element. This elimination has never before been accomplished. Consequently, the source sound can be reproduced faithfully. In addition, the crystal synchronization reference in the analog signal processing section enables a timing signal without jitter to be supplied for D/ A conversion. This causes the sound quality to be improved without any FM modulation of the analog-converted signal in the jitter component.

Figure 4. Conceptual diagram of K2 interface operation

===

====

====
https://i.postimg.cc/13zDRFjg/Screenshot-at-2023-10-10-03-02-47.png
https://i.postimg.cc/m2jQGj8f/Screenshot-at-2023-10-10-03-03-10.png
https://i.postimg.cc/zDnCx8hf/Screenshot-at-2023-10-10-03-03-27.png
https://i.postimg.cc/0NbGnSGg/Screenshot-at-2023-10-10-03-03-34.png
https://i.postimg.cc/nhdYq2DY/Screenshot-at-2023-10-10-03-04-45.png
https://i.postimg.cc/Wz60zwVZ/Screenshot-at-2023-10-10-03-04-53.png
https://i.postimg.cc/QxhpPwgJ/Screenshot-at-2023-10-10-03-05-58.png
https://i.postimg.cc/9fKdnf1D/Screenshot-at-2023-10-10-03-06-14.png
https://i.postimg.cc/0QSYRqHc/Screenshot-at-2023-10-10-03-06-29.png
https://i.postimg.cc/C5Vjw63Y/Screenshot-at-2023-10-10-03-06-45.png

==========

Early CD players get a bad rap. Not sure why (see my comments at the end).

"it was only fair to assume that the CDP 502 ES also had the potential of blowing most modern CD players out of the water, simply because of the impressive components that were used inside."

That's from a 2023 review: Sony CDP-502ES (Gear & Review) - Explorations in Audio:

https://eiaudio.de/gear-and-review/cd-players/sony-cdp-502es/

There are quite a few posts, on forums, in which various CD players of that 2nd gen era (~ mid 1984 - early 1986) which are generally praised for very good sound and solid construction. In many ways, this is not surprising because of success of 1st gen. And all the resources the major manufs like Sony were putting into evolving the CD technology.

The mag journalists also wrote glowing praise for most of these players.

First things first: I think that from the introduction of the CD player format (1982) until roughly the end of the 1980s, the press and mag. reviewers categorized CD players based on technology and year. Correct (or add to) the following list:

First (1st): 1982 1983: e.g. Sony CDP-101 and Philips CD101 (14-bit 4x os)

Second (2nd): 1984: 2x oversampling. e.g. Yamaha CD-X1; Philips CD104 (14-bit 4x os)

Third (3rd): 1985/86: 2x and 4x oversampling.

Fourth (4th): Late '86-1987: e.g. Philips cd-650 (16-bit, 4x)

Fifth (5th): Late '87-1988. 8x oversampling, 18-bit. e.g.Yamaha CDX-1100

Sixth (6th): 1989-??. 1-bit introduced (PLM, PDM)

=====
Refs:

Philips CD player Timeline
https://www.stereophile.com/content/four-3rd-generation-cd-players-adcom-magnavox-onkyo-yamaha
(note: Stereophile defines 3rd gen in a different time/technology frame)
=====

Another reason for bringing up this issues to get better figure out whether "early digital" playback gear really was (is) as hard and nasty-sounding as many audiophile and certain Golden Ear reviewers claim it was. Reading non-audiophile magazine reviews from that time -- in the US: Audio, High Fidelity, even gullible Stereo Review -- reviewers were all quite on board with CD and its sonics. And it's not as if they didn't have the best analog-playback gear at their comparative disposal--as well as (of course), the ultimate reference: live, un-amplified sound.
My own immediate impressions of my first CDP (1986 Philips 104, 2nd gen, completely stock) -- comparing to, say, a Dual CS-505/Ortofon Om-30/Van Alsitine super-PAS phono pre-amp -- were all very positive. I even (henceforth) opted for CD rather than LP as my default music format in purchases.
That Philips (TDA1540 based) has a bold, detailed, punchy sound that is not easy to come by in the mid-grade vinyl or digital gear of today.
It might be possible to disguise one of those 1st-3rd gen CDPs, put it in modern chassis or "black box", take it to local high end show or shop or fanyboy meetups, with their fancy Chords and Mola Mola's, and pull a fast on on 'em. [I know what my 1986 super DIY modded CD-650 is capable of -- and why my VPI-19/Rega RB300 / Sumiko BP special are in "like new condition!!] And yes, we can include the early hot-rod mods by Meridian, Mission and cottage-industry Magnavox mod jobs.

I think the audiophile crowd likes a certain sound quality that is simply different to the mainstream. And by all means, neither is better or more nuanced than the other. I've been in both camps, over the decades. I actually prefer the mainstream sound which is punchier and rawer (and much of my collection is classical music!).
If you have vintage gear ... Replace the orig. electro caps, and use clean, filtered power for a lifetime of enjoyment. And don't sell any of that vintage gear. That's a problem audiophiles and journalists have ... they sell/trade-in/swap stuff so often that they lose important reference points, so no way compare whether their latest new toy is actually better than prev.

I'm still a bit mystified about why certain audiophiles and audiophile journalists, like M. Fremer, et al, vilified early digital. And continue to do so of early and modern digital.
Here's what Mr Holt said of a certain Kycocera CD player in 1985. [A time when vinyl playback was very mature and advanced; and Holt had access to some excellent other analog gear, including the 15ips open-reel he was particularly fond of]:
https://www.stereophile.com/content/kyocera-da-910-cd-player

=======

In short, the DA-910 is now the CD player to beat. It raises CD sound quality to the point where even CD's most resolute opponents are going to have to take a long, hard listen and reconsider their feelings about this fledgling medium. This is the first time I have heard any player whose high-end quality is comparable to the best I have ever heard from an analog disc. (My basis for comparison is analog reproduction whose spectral balance is as nearly as possibly identical to that of a disc's CD equivalent.

I do not feel that any other comparison is valid.)

To Sum Up:

This is an incredible player. I can't afford it either, but if Kyocera wants it back, they may have to sue me!
============

Before I bring Mr Fremer into this discussion, I will reference similar sentiments from one of his fellow Stereophile journalists:
Onkyo DX-7555 CD player
Robert Deutsch | Jan 23, 2008
https://www.stereophile.com/cdplayers/108onk/index.html

=====
"I first heard a CD player in my own system in 1984 or 1985, several years before I began writing for Stereophile.

[...]

I have a feeling that if the first CD player I heard in my system had been an Onkyo DX-7555, my impression of the format would have been much more positive."

======================

Mr Fremer has made even more critical comments AGAINST digital, notably early digital recordings, CDs and players.

As noted prev. in this thread, that was not my personal auditory opinion from the get-go. And recently revisiting 2nd - 4th gen models (1984-88) by manufs like NEC, I am more convinced than ever before that even several completely un-modded (stock) CD MID-PRICED players from 1984-88 may DEFINE the golden age of 16-bit/44.1-khz playback.
And, yes, the best of the bunch may EVEN be 16-bit 2x oversampling chipsets, incl. those DAC chips with parallel inputs (PCM54) and S/H outputs.

To that end, I have just acquired an NEC cd-650 -- made in 1986. Chipset: 2 x PCM56 (dacs) + NEC µPD6352ACA (2x df); CD mech: Sony KSS-123A
https://www.gammaelectronics.xyz/audio_12-1986_nec.html

==========

(Source: Audio magazine, Dec. 1986)

Use and Listening Tests

I have found that the sound of a piano is one of the most difficult to reproduce faithfully. Furthermore, even if mike placement and recording technique are good, some CD players will impart to piano sounds an unnatural and some what strident character which is immediately apparent (even to less-experienced listeners). It was for this reason, in addition to the fact that I like the music, that one of the CDs I used to check out this player was a new Telarc release offering Prokofiev's Piano Concerto No. 3 and Tchaikovsky's Piano Concerto No. 1, played by Jon Kimura Parker with Andre Previn conducting the Royal Philharmonic Orchestra. I'm told that Schoeps MK-2 and MK-4 mikes were used and that the digital recorder was a modified version of Sony's PCM-1610. Monitor speakers were B & W 801Fs (no wonder the recording engineers got the balance right). To put it succinctly, the sound reproduced by the NEC CD-650E hooked up to my reference system was absolutely superb-as good as I've heard from just about any CD player.

The supplied remote control worked flawlessly, and when listening to non-classical fare, I found the ability to do random-access programming from my easy chair rather a nice convenience. I'll even confess to having pushed the open/close button a couple of times from across the room, even though that feature serves no practical purpose. Oh, well, NEC can be forgiven for this slight over-embellishment, in light of the excellent player they have created at a very reasonable price.

-Leonard Feldman

(Source: Audio magazine, Dec. 1986)

==========

I paid about $110 usd incl tax via eBay offer sale. No remote or xport screws, but unit is (arrived) in excellent cond.
I have only played unit for about 12 hours. No issues, but for certain functions , I did have to dig out a Philips (!!) universal remote and re-code it NEC (0670).
I assume this unit still has all orig electros, but none have leaked or buldged. Plan to replace all electro's at some point -- will adhere as close to orig uF as possible. Increasing uF, a common modding mistake, can actually make the "tuning" worse. NEC knew what they were doing.
As is, un-modded, it sounds just fine. I'm not going to over-hype with subjective evals. too much as I want the used-market value to remain ... ahem ... accessible. NEC knew what they were doing.
This model has the same chip-set and manuf year as that 903 model. Ahem ... NEC knew what they were doing.

Some photos ...

https://postimg.cc/bdz66PXB
https://postimg.cc/TLj4K2zr
https://postimg.cc/56sD506g
https://postimg.cc/cKnjNBTk
https://postimg.cc/NL73MbtW
https://postimg.cc/ts7wBj6j
https://postimg.cc/LYFW4gb9
https://postimg.cc/QH8RW5xV
https://postimg.cc/DSDDDMQb