Diagnosing Slow Screen Transitions on WinCC OS Clients: Terminal Bus Latency and Network Optimization

When slow page-to-page transitions occur exclusively on WinCC OS Clients while the OS Server remains responsive, the bottleneck is caused by Terminal Bus network latency, mismatched network adapter duplex settings, misconfigured Simatic Shell bindings, or DNS/WINS name resolution delays. Because OS Clients resolve tags, alarms, and server data packages over the Terminal Bus, dropped network packets or RPC communication timeouts force multi-second retransmission pauses. Reviewing the client diagnostic log (C:\Program Files (x86)\Siemens\WinCC\diagnose\WinCC_Sys_01.log), hardcoding 1 Gbps Full Duplex on switches, and adding local hosts entries resolves the client latency.

Terminal Bus Communication Architecture in Distributed PCS 7 Systems

In a distributed SIMATIC PCS 7 architecture, client-server communications are decoupled from the industrial controller network:

Plant Bus (System Bus): Carries high-speed S7 Named Connections between S7-400 CPUs and redundant OS Servers.
Terminal Bus: Carries operator HMI traffic between OS Servers, OS Clients, Engineering Stations, and Process Historian.

When an operator on an OS Client clicks an area overview button to switch pictures, the client must query the OS Server's Data Manager across Microsoft RPC (Remote Procedure Call) and TCP/IP, request updated process variables, evaluate faceplate scripts, and load graphic resources. If the Terminal Bus suffers packet loss or latency, the client interface freezes during every picture transition.

Why Client Latency Occurs Independently of Server Performance

A common diagnostic puzzle is observing that the OS Server operates at lightning speed (or has zero CPU load in headless server mode), while all OS Clients exhibit 5 to 12 second delays between pictures:

The server accesses its SQL Server database and Tag Management locally in RAM across high-speed internal inter-process pipes.
The client, however, relies entirely on network round-trip time. If an RPC call experiences a network timeout (default 5000 ms), the client graphics thread (pdlrt.exe) pauses synchronously until the retry completes.
A single misconfigured network switch port can cripple client responsiveness without affecting AS-OS controller polling.

Diagnostic Inspection: WinCC_Sys_01.log and Channel Diagnostics

The primary source of truth for client delays is the diagnostic log:

Location: C:\Program Files (x86)\Siemens\WinCC\diagnose\WinCC_Sys_01.log on the client PC.
Look for error signatures:
Connection to server broken / reconnecting: Indicates physical cable faults or switch port flapping.
RPC timeout / Call failed: Indicates DNS name resolution latency or firewall packet throttling.
DMClient: Tag request timeout: Indicates the OS Server is saturated or the network link is saturated.

Network Root Causes: Simatic Shell, DNS Latency, and Duplex Mismatches

Three dominant technical flaws cause client-only slowness:

DNS / NetBIOS Name Resolution Lag: If the OS Client queries the OS Server by hostname and the DNS server is slow or non-responsive, Windows attempts NetBIOS broadcast resolution, delaying every remote transaction by 3 to 7 seconds.
Duplex Mismatch on Managed Switches: If the client PC is set to Auto-Negotiate while the managed Cisco/Scalance switch port is hardcoded (or vice versa), the port falls back to Half Duplex, causing massive collision rates under graphic burst traffic.
Simatic Shell Binding: If Simatic Shell is accidentally bound to a secondary or virtual network adapter (such as a VMware bridge or wireless adapter), multicast discovery broadcasts fail or route out of order.

Comparative Performance Matrix: Network Bottlenecks vs Diagnostic Indicators

The following table correlates client transition latency symptoms with network causes:

Symptom on OS ClientDiagnostic Log SignatureUnderlying Network DefectCorrective ActionFixed 5-second delay on every screen switchRPC runtime error 1722 / timeoutDNS resolution delay resolving server nameAdd static IP and hostname to hosts fileProgressive screen lag during shiftNetwork buffer overflow in CCDataMgrDuplex mismatch (collisions on switch port)Hardcode 1 Gbps Full Duplex on PC and SwitchClient disconnects periodicallySimatic Shell multicast timeoutSimatic Shell bound to incorrect network cardRe-bind Simatic Shell to Terminal Bus adapterSpecific screen takes 15s to loadPicture not cached locallyLarge bitmap graphics loaded across network shareEnable local picture caching in Client propertiesAlarms update with 10s delayServer data synchronization lagTerminal Bus traffic congested by office backupsSegregate Terminal Bus onto dedicated VLAN

Step-by-Step Network and Client Optimization Workflow

Follow this sequence to eliminate OS Client latency:

Audit Diagnostic Log: Open C:\Program Files (x86)\Siemens\WinCC\diagnose\WinCC_Sys_01.log on the client; identify timeout warnings.
Configure Windows Hosts File: On the OS Client, open C:\Windows\System32\drivers\etc\hosts as Administrator. Add static mapping for all OS Servers: <Server_IP> <Server_HostName>.
Hardcode Network Port Speed: In Windows Device Manager, open the Terminal Bus NIC properties -> Advanced -> Speed & Duplex. Set to 1.0 Gbps Full Duplex (matching the Scalance/Cisco switch configuration).
Re-bind Simatic Shell: Open Windows Explorer -> Simatic Shell -> Settings. Ensure the selected network adapter strictly corresponds to the physical Terminal Bus IP address. Verify encrypted communication settings match the servers.
Enable Picture Caching: On the OS Client project, open WinCC Explorer -> Computer Properties -> Graphics Runtime. Increase Picture Cache to 20–30 screens.
Verify Terminal Bus VLAN: Ensure the Terminal Bus resides on a dedicated, isolated VLAN free from corporate broadcast storms.

Verification

Validate client performance through these benchmarks:

Ping Latency and Jitter: Ping the OS Server from the client (ping -t <Server_IP>). Confirm response time is consistently < 1 ms without packet loss.
Screen Transition Timing: Switch between 10 different complex plant pictures. Confirm every screen renders in under 1.0 second.
Log File Cleanliness: Inspect WinCC_Sys_01.log; confirm zero RPC timeouts or communication broken warnings.
Full Plant Alarm Burst: Trigger a test alarm burst in CFC; verify all OS Clients update their alarm banners synchronously without UI freeze.

Limitations and Operational Precautions

Observe the following networking precautions:

No Wireless Operator Consoles: Mission-critical OS Clients must never communicate over Wi-Fi on the Terminal Bus; RF interference and roaming handoffs cause unrecoverable WinCC session drops.
Managed Switch Port Security: Do not configure aggressive port security policies (such as MAC flapping shutdowns) on redundant server links.
Commercial Accuracy Disclaimer: Software order numbers (MLFBs), package structures, and hardware capacity prerequisites vary across PCS 7 versions and hardware platforms. System integrators must confirm exact software entitlements and licensing requirements against the official Siemens Catalog CA 01 and obtain an authorized Siemens distributor quotation prior to purchasing.

Engineer FAQ

Should network adapter speed and duplex be left on Auto-Negotiate?

In mission-critical industrial networks, it is recommended to hardcode both the PC network adapter and the managed switch port to 1 Gbps Full Duplex to avoid negotiation mismatches.

What does RPC timeout 1722 mean in WinCC diagnostics?

RPC 1722 (The RPC server is unavailable) indicates that the client attempted to communicate with a remote server process that was unresolvable, blocked by a firewall, or stopped.

Can too many loaded server data packages slow down an OS Client?

Yes. A client loading dozens of foreign server packages must evaluate larger tag dictionaries; prune unused server packages to improve responsiveness.