get i2s from CEC superlink?

[Marlowe]

Hi all,

I have got a CEC TL-51X transport.
There is a 9-pin D-sub connector at the rear panel of the transport, and it is marked as 'superlink'.

I have searched for the term 'superlink'.
It seems an digital interface between the transport and DAC of CEC.
This is the diagram of 'superlink':
http://www.clofis.nl/nl/cec/cec_superlink.htm

Since the superlink is consist of data, bitclock, master clock, could/how I got the i2s data via this interface?

[Marlowe]

After a series of experiments, something new regarding to the CEC superlink has been found.

1) The pinout of the 9-pin D-sub connector is:
Pin 1 - Ground.
Pin 2 - Bit clock (BCK)
Pin 4 - Word clock (WCK)
Pin 6 - Audio data (DOUT)
Pin 8 - De-emphasis (DEEM) (inverted)
Pin 9 - Master clock (MCLK) input

2) To activate the superlink output, it is required that a master clock sent to the CD-transport. The frequency is 16.9344MHz.

3) The digital output from superlink is encoded in Sony/BB format. It is difficult to convert to I2S. Fortunately, the TDA1543A works well with superlink, if you are a NOS fan .

4) Whether or not to get the digital output from superlink, it is always a good idea to reclock the CEC transport through the superlink interface.
In my current configuration, a 16.9344MHz clock generated by a TXCO is sent to CEC TL-51X, and the SPDIF output from the transport is sent to the DAC (A NOS TDA1541A).
Under the new clock, the CEC is actually improved on the whole. The sound is much solid (especially in the bass) and refined.
Other clocks such as KC-7 are also good for this application.


More information regarding to the CEC is posted on my blog
(in Chinese at all).

[tubenut]

Originally Posted by Marlowe

4) Whether or not to get the digital output from superlink, it is always a good idea to reclock the CEC transport through the superlink interface.
In my current configuration, a 16.9344MHz clock generated by a TXCO is sent to CEC TL-51X, and the SPDIF output from the transport is sent to the DAC (A NOS TDA1541A).
Under the new clock, the CEC is actually improved on the whole. The sound is much solid (especially in the bass) and refined.

More information regarding to the CEC is posted on my blog
(in Chinese at all).

I have recently fitted an ultra clock from audiocom to a TL51 transport, replacing the original crystal obviously.

Are you saying it may be worthwhile to link this same clock to the pin 9 of the D Sub? would this enable/activate some synchronising of the digital output with the main clock like Theta does for eg?

[Marlowe]

In my case, the NOS DAC does not need the master clock, thus the external master clock is sent to the transport independently.

In contrasting to NOS DAC, OS DAC usually need the master clock to be synchronized with the transport. In general, the master clock is generated in the transport, and the DAC obtains the master clock by decoding the SPDIF in the digital receiver chip (PLL). However, it is better to place the master clock in the DAC and sent it back to the transport in a separate connection.

As you said, it is a good idea to synchronize the DAC with the transport via the superlink interface. To do that, a little modification of the DAC should be performed. Perhaps the following way works though I have not tried:

The DAC still receives the digital data from SPDIF, but now the master clock (16.9344Mhz) is generated in the DAC, and sent it back to the transport by a 9-pin D-Sub connector. At this time, only 2 rails are used in the connector, pin 1 for the ground and pin 9 for the master clock. That is all. No modification on the transport is needed.

It is actually worthwhile to sync the both in this way rather than the usually PLL decoding.

[tubenut]

OK

Question then is, if the original crystal/cap clock has been removed of the digital board and now a better clock (audiocom ultraclock) feeds it, is there any point in feeding a clock in to the d sub (assuming this clock was not better then the new internal clock)?

[Marlowe]

If I understand your post correctly, your question is that a good clock in transport vs. a not-so-good clock in DAC, which is the better?

It depends on how good the clock in the transport is and how bad the clock in the DAC is. Anyway, the best solution is that you put the good clock in the DAC, and feed it back to transport with the superlink interface.

[tubenut]

No, not at all, rather lets say you put your tent clock inside the TL, replacing the original basic clock cct of the player, would there be any benefit running the set up like you are doing with an identical clock circuit feeding in to the d sub as well?

[Marlowe]

Okay. Your intention is to feed the transport an external clock through the superlink rather than synchronize the transport with the master clock generated in the DAC.

In the TL-51X, there is a simple logic circuit deciding the source of the master clock between the internal crystal and the superlink input.
If there is a clock input from superlink, then the internal clock is disabled.
Therefore, it is non sensible to feed the superlink a clock worse then the internal one.

[tubenut]

So, assuming internal clock and external stand alone clock are of identical quality, there is no benefit to the external clock if not part of the DAC? ie no buffers are enabled as a result of ext clock for eg.

[Marlowe]

However, it is complicated for such cases.

Given two identical clocks, one is in the transport, and the other is in an external box. The internal one shares the power of the transport, and has a clock rail shorter than the external one. The external clock has a stand alone box, a dedicated power supply, and a longer path to the transport. Considering these factors, it is hard to say which one is better.

The external clock is preferred to me because I do not want to mod the TL-51X at all.