What is the difference in CD transports?

You have to look at construction, both mechanical and electrical. Also, digital jitter plays a role. Some transports deal with this better than others. Power supply is a big deal. The better the supply, the better the sound usually providing all else is up to snuff. So, it's not just about reading 1s and 0s.

A transport is no different than amps or preamps with respect to the internal circuit components - the better the quality, the better the sound. This is because preserving an audio signal is is tricky (and expensive). A transport that sounds better does a better job of preserving the signal that it generates and transmits, meaning that the circuitry and power supply are designed to a higher level.

A transport will read the data off the disc and convert it to an electrical signal similar to that found in an amplifier, with some differences. The digital data stream read off a CD contains the "ones and zeros" which represent the music signal. But it also contains the critical timing data that tells the DAC exactly when to start the conversion process. If this clock signal is off, the DAC converts the signal too late or too early, resulting in distortion (jitter). Both the timing and music data are stored as one signal superimposed on one another. The transport, ideally, must recreate the waveform without any "overlaps". To do this, it has to have a very precise clocking circuitry.

There are many things that cause the timing signal not to coincide with the audio signal. One is the power supply. If it is not fast enough to switch at the right time or does not maintain a dead-on DC, the error results. The power supply is affected by the transport motor, the drawer motor, the panel LED circuits and other electrical devices (caps, inductors). Also, if the clock circuitry is not well designed, it will introduce errors in and of itself.

To minimize the errors, the circuiry requires a lot of thought and execution which means cost. Even if timing and music data are perfect, the signal still has to travel through the transport making it subject to distortion from circuit elements much in the same way as in other electronics. To preserve the signal path, again, expensive circuitry, highly regulated and discrete power supplies, and solid construction works best. Bottom line - design and parts quality affects all audio components.

A learned engineer @ one of the well-known & respected digital player companies wrote this once. I have picked this snippet up from his post:-

There are really only 3 three main parts to any transport.
1) the CD mechanism to read the data off the disc
2) the clocking and digital output methodology.
3) the powersupply

CD mechanism overall rigidity, smoothness of operation of the lens as it reads the disc, clamping the disc to reduce/eliminate wobble in (1) all have an effect on sound. Maybe subtle but there.
Clock stability in (2) above reduces jitter & hence read errors.
Poor power supply design in (3) can pollute quiet supplies with clock infested supplies. This could create spurious energy in the output data stream & reduce SNR &/or dynamic range thru an elevated noise floor.

Better designs pay more or much more attention to these details & the price goes up accordingly.

Logic would dictate that using a common reference clock for the transport and DAC would improve performance. Has anybody ever thought about finding a quality transport & DAC and using one lab grade ( like HP, etc... ) external clock for both of them? Sean
>

Sean,

I think that your logical thinking is on the right track!
I think that Wadia uses 1 clk for both transport & DAC when one buys their 270 transport & 27 DAC. The clock is in the DAC unit & is fed back to the transport using a glass fibre (ST) cable. They call this their "Clock Link" technology, I think. I believe that they do the same thing inside their CD players but it is transparent to the users as the routing is all internal. I think that the fed back clock is put in a FIFO & this FIFO can be compensated for time (which is phase) delay. The time delay, of course, occuring in the fed back clk signal.
FWIW.