FactoryTalk View SE: Configuring Distributed Applications

Mark Townsend7 min read
Allen-BradleyFactoryTalk ViewTechnical Reference
Licensed PE Working through this on a live machine? A Maine-licensed engineer can take it from here — included with IMD hardware, by the hour for everything else. Book an engineer

Overview: Where FactoryTalk View SE Fits

FactoryTalk View SE (Site Edition) is the supervisory-level HMI product in the Rockwell Automation visualization family. Selecting the correct product and the correct SE application type up front determines how many servers you deploy, how clients resolve tags, and whether the system can be scaled or made redundant later. Retrofitting a distributed architecture onto an application that was created as a stand-alone one is a rebuild, not a setting change.

Product Scope Deployment model
RSView32 Legacy machine/process supervision and control Component-based integrated package: single server and single client on one PC, expandable to remote clients (maximum 20)
FactoryTalk View ME (Machine Edition) Machine-level operator interface Single-application runtime, typically downloaded to a PanelView Plus terminal or run on a PC
FactoryTalk View SE (Site Edition) Supervisory, line, area or site-wide visualization Client/server, single computer or distributed across multiple computers on a network
Design rule: ME projects are self-contained and address a single machine. SE projects are multi-user and multi-server. Do not use ME as a "cheap SE" for a system that will later need more than one concurrent operator station, historical alarming across areas, or remote clients — the migration path is a rewrite of the graphics and tag references.

SE Application Types

FactoryTalk View SE creates the application in one of two structural forms, chosen in the New Application dialog of FactoryTalk View Studio. The choice binds the application to a FactoryTalk Directory scope.

Attribute Local (stand-alone) application Network Distributed application
Directory used Local FactoryTalk Directory (that PC only) Network FactoryTalk Directory (one directory server for all participating PCs)
HMI servers One HMI server, on the local machine Multiple HMI servers, each able to run on its own computer
Clients Clients on the same computer SE Clients on any networked computer that points at the same Network Directory
Redundancy Not available Primary/secondary HMI and data servers supported
Typical use Single-PC operator station, small skid Plant/site system, multiple operator stations, area segregation

Components of a Distributed SE System

A distributed application is an assembly of services. Understand each one as a separate installable role, because in a real deployment they frequently live on different computers.

  1. FactoryTalk Directory (Network) — the name-resolution and security backbone. It stores the application tree (system, areas, servers), user accounts and policies. It does not store graphics or tag values. Every SE Studio, server and client on the system must be pointed at the same Network Directory host using the FactoryTalk Directory Server Location Utility.
  2. HMI server — supplies graphic displays, HMI (memory) tags, the HMI tag database, alarms of the HMI tag-alarm type, macros, and data logging. One HMI server is typically created per area. It runs as a Windows service and can be started automatically on computer boot.
  3. Data server — supplies live controller data. This is either RSLinx Enterprise (native Logix/EtherNet/IP browsing with direct references such as ::[SHORTCUT]TagName) or an OPC data server for third-party devices.
  4. Tag Alarm and Event server / Logix alarms — centralizes alarm detection and subscription so that clients receive the same alarm state regardless of which PC they run on.
  5. FactoryTalk View SE Client — the runtime operator station. It is configured by a .cli client configuration file that specifies directory location, application, area, startup display, and window behavior.
  6. FactoryTalk View Studio — the single engineering environment for both ME and SE. In a distributed application, Studio edits the application over the network; multiple engineers can edit different components concurrently.

Deployment Procedure for a Distributed Application

  1. Choose and build the Network Directory server first. Install FactoryTalk Services Platform on that computer and leave it running continuously; if the directory host is unavailable, clients cannot resolve the application.
  2. On every other participating computer, run the Directory Server Location Utility and set the Network Directory to the host computer name. Use the same computer name resolution method (DNS or hosts file) everywhere; mixing name and IP causes intermittent "cannot connect to directory" faults.
  3. Verify time synchronization across all nodes. Alarm ordering, data logging and security token validity all depend on consistent clocks.
  4. In FactoryTalk View Studio, create a Network Distributed application. Add areas that mirror the physical or process segmentation of the plant.
  5. Add one HMI server per area, specifying the target computer for each. Set the startup type so the server loads automatically and continues running when no one is logged on.
  6. Add a data server. For Logix controllers use RSLinx Enterprise and define shortcuts; keep the shortcut name identical between development and production so display tag references never need editing.
  7. Build graphic displays referencing tags through the area path, not through absolute machine paths, so displays remain portable between areas.
  8. Create client .cli files per operator station. Test each one on the actual target hardware, not only on the engineering PC.
Licensing: SE server and SE client activations are separate. Count concurrent clients, not installed clients, and confirm activation availability before commissioning — an unlicensed client will fail at runtime start, not at configuration time.

Verification and Commissioning Checks

Check How to verify Expected result
Directory participation Directory Server Location Utility on each node All nodes list the same Network Directory host
HMI server running FactoryTalk Administration Console / Windows Services on the host Server state Running; no repeated restart entries in the Diagnostics log
Data server connectivity Browse tags from Studio through the RSLinx Enterprise shortcut Controller tag tree expands with live values
Client display data Launch the .cli file on the operator PC No #### or asterisk placeholders on numeric fields
Alarm delivery Force an alarm condition Alarm appears on all clients in the area within the configured scan/subscription time
Security Log in with a restricted operator account Secured commands and displays are denied per FactoryTalk Security policy

Common Distributed-Application Failures

Symptom Probable cause Corrective action
Client cannot find the application Client pointed at Local Directory instead of Network Directory Reset directory location on the client and restart the FactoryTalk services
Tags show as invalid on one client only Firewall blocking RPC/DCOM or FactoryTalk service ports on that PC Apply the required firewall exceptions; verify with the services stopped-and-restarted
Displays open but no live data RSLinx Enterprise shortcut name differs from the one used in tag references Rename the shortcut to match the reference string rather than editing every display
Intermittent loss of all clients Directory host offline, renamed, or on a different subnet without name resolution Stabilize the directory host; use consistent DNS entries for all nodes
Duplicate alarms on multiple stations Alarming implemented per-client instead of at a server-level alarm service Move alarm detection to a Tag Alarm and Event server or Logix-based alarms
Studio edits not visible at runtime Client cached an older display, or the HMI server was not restarted after structural changes Refresh/restart the affected client and, for component additions, restart the HMI server

Selection Guidance

  • Single machine, single operator terminal, no networked clients → FactoryTalk View ME.
  • One PC supervising a process, no plans for remote operator stations or redundancy → SE Local (stand-alone).
  • Multiple operator stations, multiple areas, server redundancy, centralized security → SE Network Distributed.
  • Existing RSView32 system near its practical limits (single server/client base package, remote client ceiling of 20) → plan an SE migration; the client/server model removes the single-server constraint and adds FactoryTalk Directory-based security.

When the eventual scope is uncertain, build as Network Distributed from day one on a single computer. The architecture supports collapsing all roles onto one PC, and adding a second HMI server later is then a configuration task rather than an application conversion.

What is the difference between FactoryTalk View SE and ME?

ME is a machine-level operator interface producing a single self-contained runtime, typically for a PanelView Plus terminal. SE is a supervisory client/server product that can distribute HMI servers, data servers and clients across multiple networked computers under a common FactoryTalk Directory.

Can I convert a local SE application into a distributed one?

Not by a setting change. Local applications are bound to the Local FactoryTalk Directory and support one HMI server on one PC. Moving to distributed requires creating a Network Distributed application and importing or rebuilding the components under the Network Directory.

How many remote clients did RSView32 support?

RSView32 shipped as a single-server, single-client package on one PC with expansion to remote clients up to a maximum of 20. Systems approaching that limit are candidates for FactoryTalk View SE.

Why do my SE clients lose connection intermittently?

The most common causes are an unreliable or renamed FactoryTalk Directory host, inconsistent name resolution between nodes (mixing computer names and IP addresses), and firewall rules blocking FactoryTalk service traffic. Verify all nodes point at the same directory host and that time is synchronized.

Do I need a separate license for each SE client?

Yes. SE server and SE client activations are licensed separately and are consumed by concurrent runtime clients. Confirm activation availability before commissioning, since a client without activation fails when it starts, not when it is configured.

Back to blog