EBCDIC (Extended Binary Coded Decimal Interchange Code) is an 8-bit character encoding created by IBM for its mainframe and midrange systems. It is still used on z/OS and IBM i platforms, which is why converting between EBCDIC and ASCII remains a common task in data migration.

Which EBCDIC code page does this use?

It uses Code Page 037, the US and Canada EBCDIC variant. EBCDIC has dozens of national code pages that differ mainly in the placement of currency symbols and a few punctuation marks, so results for those characters can vary by code page.

Why are EBCDIC letters not in a simple sequence?

In ASCII the letters A to Z occupy consecutive codes, but in EBCDIC they are split into blocks with gaps. The letters A to I, J to R, and S to Z each sit in separate ranges. This is a legacy of the punched-card column layout EBCDIC was designed around and is a classic source of sorting bugs.

What happens to characters not in the code page?

When encoding, any character with no Code Page 037 equivalent is skipped and listed below the output. When decoding, a byte that has no printable CP037 character is shown as a replacement marker so you can see exactly where it occurred.

Does the conversion happen on a server?

No. The full Code Page 037 table is built into the tool and all conversion runs locally in your browser. No text or byte data is uploaded.

EBCDIC ↔ ASCII Converter

EBCDIC — the Extended Binary Coded Decimal Interchange Code — is IBM’s 8-bit character encoding for mainframe and midrange systems. It is unrelated to ASCII and assigns completely different byte values to the same letters. This converter translates between ASCII text and EBCDIC byte values using IBM Code Page 037, with output in hex, decimal, or binary.

How it works

The tool carries the full printable Code Page 037 table, which maps each EBCDIC byte to the ASCII character it represents. Encoding looks up each character of your text in the reverse table to find its EBCDIC byte, then formats that byte in your chosen base. Decoding parses the byte values you enter, validates that each is a value from 0 to 255, and maps it back to its character.

Characters that have no place in Code Page 037 are handled explicitly: when encoding they are skipped and listed, and when decoding any non-printable byte is shown as a replacement marker, so a conversion never quietly loses or invents data.

The non-contiguous letter layout

The most surprising feature of EBCDIC is that its letters are not in one continuous run. The capital letters fall into three separate blocks — A to I, J to R, and S to Z — with unused byte values between them. This reflects EBCDIC’s origin in the column layout of punched cards. The practical consequence is that a naive byte sort of EBCDIC text does not match alphabetical order across those boundaries, a well-known trap when porting code from mainframes.

Example

The text Gera 037 encodes in Code Page 037 to the hex bytes:

C7 85 99 81 40 F0 F3 F7

Here G is C7, the lower-case era are 85 99 81, the space is 40, and the digits 037 are F0 F3 F7. Decoding those bytes reproduces the original text. Everything is computed locally in your browser.