OOMSAP

I might be one of the most stupid audiophiles in this planet.
I have used a pair of elegant Quad ESL-57, designed more than a half of a century ago in United Kingdom for domestic use, supposed to distribute sound from the air... Its sound is so transparent and clear, sounds from air.

But the music sources have been changed a lot, now the bottleneck is obviously located at the UK made speakers. I want to have more solid sound.

First mistake happened then. I had had used second-handed Bose 501X system, probably for automotive audio. I just connected its base unit to Quad in an accident, but sounds okay, really good.

THE AMERICAN BOOM BOX, THE REAL THING has been connected. One of the most tempestuous box was left alone in the very middle of the British gentlemen. This stupid combination make some progres ssomehow. They worked togather to make the sound fully bodied with elegant suits.

Second mistake happened a few minutes ago. Another American pipe organ thing ; BOSE AWCS-1, which was listed in an auction site, is now located at the back of the old gentlemen. That pipe had been situated in a club house in a city, looked down considerable number of auditory damaged dancers, and contributed to reduce their acoustic ability, for years. That killer pipe came into my place, kicked out the boom box completely away, now rests peacefully at the back of the British gentlemen.

Unlike the boom box, it needs some adjustment to make these fire and water colleagues to help each other. The longest Bose came with wind, having potential to physically blow its British colleagues away, currently under the strict control of the pairs. It generates wind rather than sound. Some accident may physically damage QUAD, or even the window glasses, that is real. The pipe organ thing rated to handle 3,000 Watt power, now required to work with 51 years old twins that are designed for domestic use with a few fraction of a watt, or in micro-ampere order current.

Eventually, the sound lost its base that is rigidly connected to the ground took off into the wind. It may be just a giant blower and 51 year old plastic films, forced to work togather to make air move. That is up to an One Of the Most Stupid AudioPhile in this planet.

Fixing broken amplifier

eBay and other auction sites have a lot of Junk and broken power amplifiers. It is relatively hard to fix it to the original condition, because most of these broken amplifiers are fairly old, and some critical parts are obsolete and unavailable.
In most case, the housing and heat radiator are usable. In addition, if it claims that the power can be turned on, at least transformer and AC circuit is also usable.
I bought SONY's TA-N330ES, explained as Junk, but its power can be turned on. The situation is relatively simple, it makes buzzer like sound. I tried to locate the source of the noise, and the driver IC (thick film hybrid IC) makes the sound. It were possible to replace the IC if I were able to find that obsolete IC, but I decided to use modern IC LM4702 from National Semiconductor.
The modification is relatively simple, because LM4702 requires only a few registers and capacitors. TA-N330ES's output stage, protection circuit are also alive and all I need is to remake very simple driver circuit.
I am planning to include DA converter (CS8416, SRC4192 and PCM1794A/ch plus Kaneda style I/V) and power amplifier part as differential-to-single-ended converter.

I hope that would sound better than current system (which has same D/A stage with discrete I/V, and discrete differential-to-single-ended converter, preamplifier and power amplifier.

Kaneda Style outside Japan

I have heard that Kaneda-Shiki DC Amplifier (Kaneda-Style DC Amplifier) is well known in some areas, including France.
Despite the information or articles in non-Japanese language are very limited, there are enthusiasts. I believe that the sound quality of Kaneda-Style amplifier’s quality attracts so many people. I hope more information translated in English or French for those enthusiasts outside Japan, presumably by the publisher or Dr. Kaneda himself, so that to clear his copyright.

DAC modification

DAC modification
I have implemented Asynchronous Sample Rate Converter (ASRC) clocked by a Xtal oscillator to my D/A Converter designed by h_fujiwara.
There is an accessory connector (in place of dumping registers) between Digital Audio Interface (DAI) and PCM1794A D/A converters on h_fujiwara's DAC1794-3 main board. The connector is designed to plug a cable from external ASRC and memory buffer boards. I thought it is possible to reduce jitter of ASRC by replacing PLL with Xtal.
There are only 2 devices (TI’s SRC4192 & Xtal Oscillator), so I made a tiny circuit board that directly plugged into the accessory connector.

It is relatively hard to make the board because SRC4192 is SSOIC, and the oscillator is also a surface mount type, so it took almost one day to complete it.
It worked just fine. I have not expected that the new one would be as good as external ASRC & buffer
But, the difference from ASRC/memory buffer is insignificant. So, I will spend some time to find out the sound difference, or any change.

0.000,000,000,075 second vs 0.000,000,000,001 second

I wrote to h_fujiwara to make sure what the advantage of the memory buffer right after PLL locked ASRC. Although, the effect of the memory buffer is obvious, and the clarity is much better than without it, I thought that is the placebo effect.

I wondered that the jitter occurs before the DA converter, mostly in SPDIF connection (signal reflection by mismatched cable, connectors, and termination, or the optical reflection), and I misunderstood that the order of this jitter is far beyond the jitter of the PLL.
He suggested me that ASRC is locked by PLL which may have up to 0.000,000,000,075 (75p) Second jitter, while the memory buffer’s jitter is about 1pS. His answer is clear and rational in theory. I can understand that may distort the audio signal in some way. But, it is still a mystery that quite a number of his subscribers can distinguish that effect.

I recently found that I have ability to hear ultrasound in very specific condition, i.e. some resonant frequencies around 40KHz, right front of my own ear from certain orientation. The wide-band microphone shows that there is no significant sub-band sound from that disassembled earphone. 40KHz is just 0.000,025 S (25,000,000pS), it is 333,333 times of the PLL jitter.

Minor modification made a big change

DIY Digital to Analog converter, DAC1794-III, which is designed by h_fujiwara is just great. As it burns in, the sound comes to be more natural, but, the only thing is it is just being too smart and performs exactly what it should do, but nothing more than that.

H_fujiwara provides several accessories for his DACs, including Asynchronous Sampling Rate Converter (ASRC), Memory buffer for zero jitter, and digital isolator. I put ASRC to up-sample the input source into 96KHz followed by a memory buffer. I am wondering about the memory buffer, because ASRC’s output is synchronized to its crystal and there should be no jitter practically.
But, what happens is just amazing. Its sound has solid body along with extraordinary expression of what is in the CD, and I am fun to listen to the music. I find that I spend much more time to listen to the music after I installed ASRC and memory buffer.

DA converter

I made a DA converter that is produced by h_fujiwara. It consists of 2 8x8 inch boards, one of which is the power supply and the other is the digital, DAC & analog circuit. It is one of h_fujiwara’s high-end DAC with four of TI’s PCM1794 DA converters and six discrete operational amplifiers. I made the power supply board with ordinary parts. But, I made a big modification to the DAC board, I used ceramic capacitors in place of all electrolytic capacitors, i.e. there is no electrolytic capacitor on the board. I used metal-can transistors in place of plastic molded transistors except a pair of FETs for each opams. All registers except I/V converting registers are metal film registers from DALE, and I/V converting registers are tantalum nitride from Western Electric.

I thought it should be better than previous version of the DAC (also designed by h_fujiwara, consisting of 32 TI’s PCM63P DACs and simplest I/V converter (2 registers) and one discrete OPamp per channel. I implemented arranged Kaneda’s #168 opams.

It seems that I have to run the new DAC for a while to burn-in to the stable state. I will report when the new one is ready to compare.