Archiving String Tags in WinCC 6.0 SP4 with Time Stamping

Problem Overview

WinCC 6.0 SP4 (SIMATIC WinCC V6.0 + SP4) provides a powerful Tag Logging archive system that natively handles numeric process values, binary states, and raw/process tags. However, archiving string-typed tags such as batch names, operator IDs, recipe identifiers, lot numbers, or shift descriptions with a time stamp is not directly supported by the standard Tag Logging archive. When the user attempted to configure a string tag as a process value archive, the archive simply refused the data type, and the batch name vanished from the report.

This limitation applies to the classic Tag Logging archive and to the corresponding print/Excel report layouts in the WinCC Report Designer. The archived numeric columns accept BYTE, WORD, INT, REAL, BOOL, and similar data types, but no native STRING column is exposed for time-stamped rows.

This article documents three field-proven methods to capture, persist, and report string-typed batch data with time stamping in WinCC 6.0 SP4:

1. WinCC User Archives (a relational table that natively holds string columns).
2. An integer code mapping approach using a C-script lookup table.
3. An ASCII decomposition approach that stores each character as a numeric byte tag and reconstructs the string at report time.

Why Native String Archiving Is Limited

Internally, Tag Logging stores values in compressed SQL Server tables where each row contains a fixed-size numeric payload plus a millisecond timestamp. The archive configuration dialog in WinCC Explorer (Tag Logging → Archives → Properties) only offers the data types supported by the underlying logging driver. A STRING tag is a WinCC Management Level construct — it exists in the tag database and can be read by scripts and faceplates, but it has no equivalent archive column.

Three practical constraints follow:

• The Tag Logging wizard rejects STRING tags in the selection list.
• Report Designer table columns and OLE-DB queries against the archive database return numeric payloads only.
• C-script functions such as GetTagChar() return the live value at execution time but do not write to a time-stamped archive.

The C runtime helpers commonly referenced for numeric conversion are:

None of these primitives writes a string to a time-stamped archive row. The string must first be reduced to a numeric form, then stored through a numeric archive, then reconstructed at report time.

Prerequisites

Before implementing any method, verify the following:

• WinCC Explorer V6.0 SP4 with a running WinCC project (RT or RT-PM).
• Microsoft SQL Server (installed automatically with WinCC) reachable from the WinCC service account.
• The string tag is already declared under WinCC Explorer → Tag Management → [your connection] → String Tags.
• For Method 1, the WinCC User Archives option must be licensed and present (it ships with WinCC V6.0 but check WinCC Information System → Options → User Archives).
• Global Script C editor access (Global Scripts → C-Editor) and rights to add functions under Project Functions.
• The Report Designer is installed (default with WinCC V6.0).

Method 1 — WinCC User Archives

User Archives is a WinCC option that provides a relational table view on top of SQL Server. Each row carries an automatic time stamp and any column can be a string. This is the cleanest method when the batch name is free-form text.

Step 1.1 — Create the User Archive Table

1. Open WinCC Explorer and select User Archives in the navigation tree.
2. Right-click and choose New User Archive. Name it UA_BatchReport.
3. Open the archive editor and define columns:

Activate the Time Stamp property on the archive; WinCC then writes the system time automatically when a new row is appended.

Step 1.2 — Write a C-Script to Append a Row

Create a project function ArchiveBatchName():

// Project Function: ArchiveBatchName
// Appends the current BatchName string tag to the User Archive UA_BatchReport
{
    // Open the archive
    UAConnect("UA_BatchReport");

    // Read the live string tag
    char* szBatch = GetTagChar("BatchName_Tag");
    char* szUser  = GetTagChar("@CurrentUser");
    DWORD  dwPV   = GetTagWord("ProcessValue1_Tag");

    // Move to a new row
    UAInsert("UA_BatchReport");

    // Set fields
    UASetFieldValue("UA_BatchReport", "BatchName",    szBatch);
    UASetFieldValue("UA_BatchReport", "OperatorID",   szUser);
    UASetFieldValue("UA_BatchReport", "ProcessValue1", &dwPV);

    // Commit
    UAWrite("UA_BatchReport");

    // Cleanup
    SysFree(szBatch);
    SysFree(szUser);

    UADisconnect("UA_BatchReport");
    return 0;
}

Step 1.3 — Trigger on Batch Start

Call ArchiveBatchName() from a button click on the batch faceplate or from a tag-triggered action:

• Graphics Designer → Button → Event → Mouse Click → C-Action: ArchiveBatchName();
• Or schedule on a value change of BatchStart_Tag via Global Script → Actions.

Method 2 — Integer Code Mapping With Translation Script

If your batch names come from a fixed lookup list (recipe IDs, product codes, or shift codes), storing an integer code in Tag Logging and translating it at report time is the most compact solution.

Step 2.1 — Define the Code List

Maintain the lookup table as a header file under Project Functions:

// Project Function: GetBatchNameByCode
// Returns the string for a given integer batch code.
{
    switch (dwCode)
    {
        case 1: return "BATCH_RED_A";
        case 2: return "BATCH_RED_B";
        case 3: return "BATCH_GREEN_A";
        case 4: return "BATCH_BLUE_001";
        case 5: return "BATCH_BLUE_002";
        case 9: return "MAINTENANCE_CLEANING";
        default: return "UNKNOWN_BATCH";
    }
}

Step 2.2 — Archive the Integer Code

Create a process tag BatchCode_Tag of type WORD (unsigned 16-bit) and configure it for archiving in Tag Logging at a 1-second or event-triggered cycle. The archive captures the time stamp automatically.

Step 2.3 — Translate in the Report Designer

Open Report Designer → Batch Report layout. In the table column that should display the batch name, add a Dynamic field bound to the BatchCode_Tag archive column and use a C-action on the field to substitute the text:

// C-Action on report field
{
    DWORD dwCode = GetTagWord("BatchCode_Tag");
    char* szName = SysMalloc(64);
    sprintf(szName, "%s", GetBatchNameByCode(dwCode));
    return szName;
}

This returns the human-readable name directly into the printed report or Excel export.

Method 3 — ASCII Decomposition Into Per-Byte Tags

For free-form strings with no preset vocabulary, decompose each character into its ASCII code (0 – 255) and store it in a numeric archive. The Report Designer then concatenates the bytes back into text.

Step 3.1 — Declare Byte Tags Per Character Slot

For a 16-character batch name, declare:

• BatchChar_01 through BatchChar_16 as BYTE tags

Configure each one as a process value archive in Tag Logging with a 500 ms acquisition cycle. Total archive growth: 16 bytes × 1 728 samples/hour ≈ 28 KB/hour — manageable for multi-month retention.

Step 3.2 — Decompose on Batch Start

// Project Function: DecomposeBatchName
{
    char* szBatch = GetTagChar("BatchName_Tag");
    int   nLen    = strlen(szBatch);
    int   i;
    BYTE  bZero   = 0;

    for (i = 0; i < 16; i++)
    {
        if (i < nLen)
        {
            BYTE bChar = (BYTE)szBatch[i];
            char szTag[32];
            sprintf(szTag, "BatchChar_%02d", i + 1);
            SetTagByte(szTag, bChar);
        }
        else
        {
            // Pad remainder with 0 (null terminator equivalent)
            char szTag[32];
            sprintf(szTag, "BatchChar_%02d", i + 1);
            SetTagByte(szTag, bZero);
        }
    }

    SysFree(szBatch);
    return 0;
}

Step 3.3 — Reconstruct in the Report

// C-Action in report designer field
{
    char  szOut[32];
    int   i;
    memset(szOut, 0, sizeof(szOut));

    for (i = 0; i < 16; i++)
    {
        char szTag[32];
        sprintf(szTag, "BatchChar_%02d", i + 1);
        BYTE b = GetTagByte(szTag);
        if (b == 0) break;
        szOut[i] = (char)b;
    }

    char* szRet = SysMalloc(strlen(szOut) + 1);
    strcpy(szRet, szOut);
    return szRet;
}

Configuring the Batch Report With Time Stamping

Regardless of the storage method, the report layout requires these elements:

1. Open Report Designer → File → New → Batch Report.
2. Insert a Time Stamp Column from Dynamic Dialog → Archive Tags → [your archive]. Bind to TimeStamp with the format yyyy-MM-dd hh:mm:ss.000.
3. Insert a column for each numeric value archive (Method 2: BatchCode_Tag; Method 3: each BatchChar_NN; Method 1: nothing — the string is read directly from the User Archive OLE-DB query).
4. For Method 1 in Report Designer, choose Insert → Database Connection → User Archive and select UA_BatchReport. Drop the BatchName, StartTime, and EndTime columns onto the layout.
5. Save the layout as @Batch_Report.rpl and assign it as the scheduled print job in WinCC Explorer → Print Jobs.

Verification Steps

After commissioning, perform the following checks:

1. Tag write check: In WinCC Explorer → Tag Management, set BatchName_Tag to "TEST_RUN_001". Confirm the value displays in the Graphics Designer faceplate.
2. Script execution check: Click the Archive Batch button. In the global script diagnostic window, watch for the function call log.
3. Database check: Open SQL Server Management Studio, connect to the WinCC instance, and query:
   SELECT * FROM UA_BatchReport ORDER BY BatchID DESC; — confirm the row exists with the expected string and time stamp.
4. Report print check: Trigger the print job manually and inspect the generated PDF. Verify the BatchName column is populated, the time stamp matches the trigger moment to within ±1 s, and no field is blank.
5. Unicode / non-ASCII check: If batch names contain non-Latin characters, confirm the SQL Server collation is set to SQL_Latin1_General_CP1_CI_AS or Latin1_General_100_CI_AS_SC; otherwise non-Latin characters appear as ? in the report.

Troubleshooting Matrix

Method Selection Checklist