MAY25 2008

This is one of  most intriguing products I ever had.

It was recommended to me by very nice man from Romania - Mr. Florin Baciu. I am infinitely grateful for it.

He told me that this player has a magical  sound thanks to a VERY RARE DAC chip.
Thanks to the power of internet - one day later I had it at home.
It was late at night and just to check this player - I plugged it in.

My expectations of un-modified, un-lampized player were very low. BUT ...
This must be the best sounding stock player I ever had. It totally beats almost all players, and it is about equal with the best stock machines, if memory serves me, like for example Copland 289, JVC1050K2, Naim CDS-1 and maybe Cambridge CD-4. It would even beat the stock Meridian 506.20

That is already a HUGE gap in price and expectations. 2 leagues apart in perceived value.
But the SOUND   !  Suono ! Klang ! le Son ! Dźwiek ! Zwuk !
Wonderful. Even in STOCK FORM IT WAS ALMOST EQUAL to totally tweaked to the maximum Marantz CD with lampized TDA1541A.

I have no way of proving this very strong statement, but it was a damn shocking experience.
Next evening I lampized it after some heavy additional tuning and right now I am sitting and listening to my new reference king of CD.
Ah and I forgot - I paid 20 Euro for it with remote and box.


Manufacturing date of the DAC chip: August 1994
CD player is from Singapore and part from Malaysia (mechanism).


I used finally the cathode follower (one tube 6H6P) system with 330 Ohms in cathodes, 250 K in grids, and 150 V at anodes.

Transformer and power supply of anodes. The heaters in cathode followers can be AC which I used. 6,3 Vac from transformer.

The tube with two huge Obbligato caps.

This is the new 7805 regulator supplying clean 5 VDC to the DAC chip. I added the regulator because the player's regulator 7805 supplied apart from the DAC also all the player - tens of other elements. So now tha DAC has exclusivity for this clean and strong 5VDC.

This dac  chip is the rarest Philips DAC on Earth. Only used during one year - 1994 - and only by 4 companies: Philips, (5 players)  Micromega, Alchemist and Sugden.
Google has almost no information about it.
EVEN GOOGLE which knows every f***g thing.

ALCHEMIST NEXUS APD32A       2 x TDA1549                     CDM 12.1

MARANTZ CD-48                             TDA1549                            CDM 12.1

MARANTZ CD5000                          TDA1549 - SAA7378       CDM 12.1 (VAM 1201)

MICROMEGA MINIUM                      TDA1549                           CDM 12.2

PHILIPS CD750                                 TDA1541A / TDA1549    CDM 12.1

PHILIPS CD751                                 TDA1549 – SAA7378     CDM 12.1/05

PHILIPS CD753                                 TDA1549 – SAA7378     CDM 12.1 / VAM 1201

SUGDEN MASTER                           TDA1549                          CDM 12.1

From the Philips datasheet.


· Finite-duration impulse-response (FIR) filtering and noise shaping incorporated
· 2nd-order noise shaper
· Wide dynamic range (true 18-bit resolution)
· Low total harmonic distortion
· No zero-crossing distortion
· Superior signal-to-noise ratio
· Bitstream continuous calibration conversion concept
· Inherently monotonic
· Voltage output (1.5 V RMS) at line drive level
· Single supply rail (3.8 to 5.5 V)
· Optimum output voltage level over the entire supply range
· Small outline packaging (SO16)
· Wide operating temperature range (-30 to +85 °C)
· Standard Japanese input format
· No analog post-filtering required
· Low power consumption
· Integrated operational amplifiers.

GENERAL DESCRIPTION from Philips datasheet:

The TDA1549T (BCC-DAC1) is the first of a new generation of digital-to-analog converters featuring a
unique combination of bitstream and continuous calibration concepts.
A system of digital filtering, high oversampling, 2nd order noise shaping and continuous calibration digital-to-analog
conversion ensures that only simple 1st order analog  filtering is required. The circuit accepts 18-bit four times
oversampled input data (4fs) in standard Japanese format. Internal FIR filters remove the main spectral components
and increase the sampling rate to 96 times (96fs). A 2nd order noise shaper converts this oversampled data to a
5-bit data stream. For low signal levels the converter operates in the 1-bit bitstream mode with attendant high
differential linearity. Higher level signals are reproduced using the dynamic continuous calibration technique,
thereby guaranteeing high linearity independent of process variations, temperature effects and product
ageing. High-precision, low-noise amplifiers convert the digital-to-analog current to an output voltage capable of
driving a line output. Externally connected capacitors perform the required 1st order filtering so that no further
post-filtering is required. Internal reference circuitry ensures that the output voltage is proportional to the supply voltage, thereby making
optimum use of the supply voltage over a wide range (3.8 to 5.5 V). This unique configuration of bitstream and
continuous calibration techniques, together with a high degree of analog and digital integration, results in a
digital-to-analog conversion system with true 18-bit dynamic range, high linearity and simple low-cost application.

Fig.1 Block diagram.

(after some time I discovered that the DAC chip has IDENTICAL datasheet and block diagram and pinout as their own TDA1305. Whatever that means - go figure.

My proposal was initially (I was wrong, anyway):

1. To  increase the main capacitor for power from 4700 to 10 000 uF. (good idea)
2. To add os-con or black gate to legs 11 and 15 power input (good idea)
3. To kill the internal op-amp by removing its power supply from leg 10. (not so good idea)
4. To remove the op-amp filter feedback loop - external capacitors 1 nF - legs 4-5 and 6-7 respectively.  (not so good idea)
5. This way we gain a pure output not only from DAC straight - more - FROM INSIDE GUTS OF THE DAC. This is really upstream !!!!

This output I hoped was going to be current, but no - it is still Vout. With plus 1 V DC component (necessity of capacitor.)

I went further and cut off the supply line from the 7805 regulator to DAC. This one regulator feeds the whole goddamned player - 20 power consumers or so.
I took a wire from the hot leg of main 10 000 uF capacitor and wired it's 12 V to the DAC and just near the DAC I put another 7805 (no heatsink needed) and I got a dedicated +5V supply just for the DAC (legs 11 and 15 and respective oscons).

I did not touch the clock because it is 33 meg - I dont have such clock on hand.

Of course red is for CUT and blue is for ADD output.

Then I lampized it using SRPP lampizator with 250 Ohms resistors in both cathodes and 250 K in the input grids. Tube 6H6P plays here best for me.
Input caps are MKP 1 uF and outputs - Obbligato oil caps 2 uF.

After intense testing i concluded that the signal is distorted. So I just reversed to the power to leg 10, and signal from legs 4 and 7.
It plays VERY VERY WELL now.

So the list of things to do is this:
locate the chip and the legs
cut off the circuit after the legs 4 and 7.
remove feedback capacitors from these legs
wire the signal to Lampizator tubes or just via capacitor - to the outputs.

I am very excited by the feedback this whole post can generate, I wonder if myself and Baciu are alone on this planet who admire this dac and player.

The good thing is that it becomes MOST RECOMMENDED player for people who dont want to do anything DIY.

If you want high end without Audionote price - GET ONE OF THESE PLAYERS.
They have good display
good remote
easy to find laser
one half opamp per channel only
airy open and pure highs
mids to die for
superb bass
and LOTS of space inside shall you ever want a lampizator inside.

In fact, I can imagine that  with just 2 oscons and a snipit bypass from DAC legs 4 and 7 via capacitors to RCA - that would be a giant killer without any tubes.

lampizator and TDA1549

Is there lotsa sapace for lampizator or what ?! It is just begging for lampization.

lampizator and TDA1549

The mistery magic chip - TDA1549 which looks like nothing at all.

lampizator and TDA1549

The TDA with traces cut (don't quote me on that - it was early experiments)

lampizator and TDA1549

The red circles show caps to be removed.
Yellow ones -show  the upgraded ones (oscon and tantalum)

lampizator and TDA1549

this upgrade of the demodulator chip capacitors was a failure because of tight space - they had to be removed.

lampizator and TDA1549

The better attempt - I added new caps across the blue ones as on the photo one above this one.

lampizator and TDA1549

The general look on the main PCB. You must agree - nothing to impress audiophiles. Nothing to display at the HighEnd show in Germany .

lampizator and TDA1549

Three audiophiles who listened to this player lampized with 6H6P (dont dream of any other tube  for this one because it is Voltage U out type  !!!) agreed that it bettered the TOTALLY tweaked and perfect Marantz CD50 with TDA1541A. I agree. It BEATS THE 1541 believe it or not. At least to my ears.

And today I just bought the 753 hahahahahahahaha !