DWORD

**DWORD (Double Word) is a 32-bit unsigned data type** holding values from 0 to 4,294,967,295 (hex 16#00000000 to 16#FFFFFFFF). It is the unsigned counterpart to DINT and is heavily used for bit-packed status registers, IP addresses, file timestamps, and any application requiring a large positive-only integer.

**Range:** 0 to 4,294,967,295 unsigned. No negative values.

**Memory:** exactly 4 bytes (32 bits).

**Common uses:** - Bit-packed flag registers (32 BOOLs in one variable) - IPv4 addresses (one DWORD = the four octets) - Status words from intelligent drives, robotics, vision systems - Encoder counts and large positive counters (where DINT's −2 billion to +2 billion isn't enough) - Hash values and CRCs - Time-of-day in milliseconds since midnight (max 86,400,000)

**Syntax across major IDEs:** - Siemens SCL: `myFlags : DWORD := 16#00000000;` - Studio 5000: Allen-Bradley uses DINT (signed); no native unsigned 32-bit. Use UDINT in newer firmware. - CODESYS / TwinCAT: `myFlags : DWORD := 16#00000000;` - Mitsubishi GX Works3: DWORD or KW (signed) per declaration

**Bit access:** address individual bits as `myDWord.0` through `myDWord.31`.

**Common pitfalls:** confusing DWORD with DINT — DINT is signed (-2,147,483,648 to +2,147,483,647), DWORD is unsigned. Adding a negative offset to a DWORD requires casting; doing the math in DWORD silently wraps if the result would be negative. Mixing DWORD and DINT in expressions also requires explicit conversion on Siemens; CODESYS will sometimes auto-promote.

**Conversion:** `DINT_TO_DWORD()`, `DWORD_TO_DINT()`, `DWORD_TO_REAL()` (loses precision above 16 million).

**See also:** DINT (signed 32-bit), UDINT (explicitly unsigned 32-bit), LWORD (unsigned 64-bit), WORD (unsigned 16-bit).