Philips CDD522 Operating Instructions Manual page 27

Chapter 4. Layer Four: The Sector Structure
One of the advantages of the Compact Disk is fast access. In
audio applications, only accurate positioning to the beginning of a
song is required. This is done at Track level (see Chapter 5 p 28:
The Track Structure).
For data applications, the data has to be accessible in much
smaller blocks.
This is where the data-sectors come in. Data-sectors are data-
blocks that carry extra information, e.g. sector identification
number and sector type. In that extra info, we may also find
EDC, Error Detection Code, and ECC, Error Correction Code,
depending on the sector-type. These EDC and ECC are used in
the Third Layer Correction.
The Sector Structure is placed on top of the Subcode Structure
(in fact it is placed on top of the Red Book Standard). When it
comes to encoding and writing to the disk, the information in a
data-sector is handled just like that of an audio-sector, i.e.
including EFM-coding, C1-coding, C2-coding, re-mapping and
interleaving.
One sector contains 2352 bytes, which is precisely the size of
one Audio Sector (remember, 98 Audio frames containing 24
bytes). We did encounter the number 98 somewhere else: one
Subcode frame contains 98 bits per channel, being 12 bytes plus
two Sync bits. This means that a Subcode frame and a Sector
correspond one to one.
The basic data-sector always contains the following information:
Sector Sync: 12 bytes (00 FF FF FF FF FF FF FF FF FF FF 00)
Header : 4 bytes
The remaining 2336 bytes are sector-type dependent. At this
moment only five different sector-types exist (Audio sector
included):
Type 1. Red Book Sector or Audio Sector: