SEPA File Checker

Invalid characters in a SEPA file

Last updated: 28 October 2026

A character outside the SEPA character set in a reference, or a bare ampersand in a name, is enough for a bank to reject the file. The first is a SEPA rule, the second makes the file unreadable as XML. Upload the file below to see which fields contain characters that are not allowed.

Check your file for invalid characters

Choose file

Checking is free. Repair if needed: €14.95 excl. VAT per file.

Our server checks your file in memory only, never saves it to disk and deletes it after 30 minutes at the latest.

See how it works

Error
Character error
Where
References such as EndToEndId; & in names
Repair
Automatic in EndToEndId, PmtInfId and MsgId

What the error message looks like

Validators and banks report these errors in English: in your bank portal when uploading, in the status file pain.002, or in your software. A bare & or < stops the XML parser before any schema check; libxml2 writes “xmlParseEntityRef: no name”. A character outside the SEPA set in a reference is usually only noticed by the bank, or by a check that knows the SEPA rules.

XML schema validator (libxml2, for example xmllint and many online validators)

PARSE xmlParseEntityRef: no name, line 25, column 27 (<string>, line 25)

Java validators (Xerces)

The entity name must immediately follow the '&' in the entity reference.

.NET validators

An error occurred while parsing EntityName. Line 25, position 27.

Your bank (status report pain.002)

  • CH16 ElementContentFormallyIncorrect
  • FF01 InvalidFileFormat

In the status report, banks put the reason in the Rsn/Cd field. Online banking often shows a translation of it. CH16 means the content of an element is formally incorrect, FF01 that the file format is invalid, which is typical when the XML cannot be read.

Our check

  • Transaction reference (EndToEndId) 'E2E_2030/0001' contains characters that are not allowed, starts or ends with '/' or contains '//'.
  • Mandate reference (MndtId) 'M_0001' contains characters that are not allowed, starts or ends with '/' or contains '//'.
  • The file is not a valid XML document (error around line 25).

Cause

The European Payments Council defines a SEPA character set: a–z, A–Z, 0–9, the characters / - ? : ( ) . , ' + and the space. The references MsgId, PmtInfId, EndToEndId and MndtId must stay within that set. They may also not start or end with / and may not contain //. An underscore, as in E2E_2030/0001, is not allowed.

For names and descriptions, banks must at least process the SEPA set; umlauts, accents or the euro sign may be handled differently per bank.

The ampersand is a special case. In XML, & starts a reference such as &amp;, so a bare & in Müller & Söhne makes the whole file invalid XML. So does a bare <. Usually the software that writes the file must fix this.

How to fix it by hand

Fix the XML first, then the references.

  1. Search the file for & characters that are not followed by amp;, lt; or a similar name, and replace each bare one with &amp;; replace a bare < in text with &lt;.
  2. Check MsgId, PmtInfId and EndToEndId for characters outside the SEPA set and replace them, for example _ with -.
  3. Make sure no reference starts or ends with / or contains //.

Before

<PmtId>
  <EndToEndId>E2E_2030/0001</EndToEndId>
</PmtId>
...
<Dbtr>
  <Nm>Müller & Söhne</Nm>
</Dbtr>

After

<PmtId>
  <EndToEndId>E2E-2030/0001</EndToEndId>
</PmtId>
...
<Dbtr>
  <Nm>Müller &amp; Söhne</Nm>
</Dbtr>

In the example the underscore in the EndToEndId becomes a hyphen, and the ampersand in the debtor's name is written as &amp;. The umlauts stay.

How the tool repairs it

The check names each reference that breaks the rules, for example “Transaction reference (EndToEndId) 'E2E_2030/0001' contains characters that are not allowed, starts or ends with '/' or contains '//'.” The repair replaces characters that are not allowed in EndToEndId, PmtInfId and MsgId, such as _ with -.

The mandate reference (MndtId) is reported but deliberately not changed, because it must stay identical to the signed mandate. Change it in your software and submit it as a mandate amendment; see SEPA mandate reference. Names are not checked for characters: the tool does not reject a name for umlauts or accents.

What the repair writes in the change report:

  • Transaction reference (EndToEndId) E2E_2030/0001 → E2E-2030/0001.

A file with a bare & or < is reported as “The file is not a valid XML document (error around line 25).”, but it cannot be repaired, because it is not readable XML. Fix that line by hand or in your software.

Related pages

Frequently asked questions

Which characters are allowed in SEPA references?

a–z, A–Z, 0–9, / - ? : ( ) . , ' + and the space. A reference may not start or end with / or contain //.

What does xmlParseEntityRef: no name mean?

There is a bare & in the file. Write it as &amp;. Until then the file is not valid XML and cannot be checked or repaired any further.

Why does the tool not correct my mandate reference?

Because it must stay the same as on the signed mandate. Change it in your software and pass it on as a mandate amendment.

Sources

Check file