IMDTouchKnowledge BaseShop main store

Troubleshooting

Why Vijeo Designer cannot connect to an M580 PLC

HMI and touchscreens Published 8 min read

Short answer

The HMI fails to reach an M580 because one leap hides which layer broke. Match the project target to the HMIGTO, prove the computer-to-panel transfer path and the panel-to-PLC runtime path as two separate tests, add the Modbus TCP equipment the controller actually serves, then prove communication with one forced read and one approved write before any faceplate. Each layer gives a pass or a named fault.

On this page
  1. Match the target and the software installation first
  2. Prove the two network paths as two separate tests
  3. Add the PLC equipment the controller actually serves
  4. Worked example: reading %MW100 over Modbus TCP
  5. Map faceplates by parameter, not by fixed address
  6. What each symptom tells you
  7. Questions asked

An HMIGTO with finished screens that will not exchange data with an M580 is almost never one bad setting — it is six layers being commissioned as a single leap. Work them in order: project target and software, engineering transfer path, runtime network path, PLC equipment and driver, one forced read and one approved write, then faceplate parameters. Stop at the first layer that fails, because that layer names the fault and holds the fix. This order takes a panel with finished screens and no live data to live values without guessing, and it is the shortest route when you inherited the job without the original vendor's files.

Match the target and the software installation first #

The project target must be the exact HMIGTO model printed on the panel label. Driver availability and the generated runtime both follow from the target, so a project can hold finished graphics and still have no valid communication layer underneath. If the family you need is missing from the target list entirely, the article on activating the XBTOT target family in Vijeo Designer 6.2 shows where a target family gets switched on.

Schneider's FAQ on Modbus TCP support records the driver on HMIs in Vijeo Designer as validated with the M580, and notes it requires Unity V8.0 or higher on the controller side. If no PLC driver can be added at all, the installation is incomplete — repair or complete it before editing a single address.

Run the project build now and read what it reports: a target or driver-generation error is a software problem, while a runtime timeout is a network problem, and mixing the two up costs hours.

  • Read the full model designation from the HMI label and set the target to match it.
  • Build the project and clear every target or driver error before touching addresses.
  • Treat a greyed-out .vdz export or import as a project-management issue — select the project root, close open editors, confirm the project location is writable — not as evidence about the PLC.

Prove the two network paths as two separate tests #

There are two independent connections, and confusing them is the classic inherited-project trap. The engineering computer transfers the application to the panel; the panel runtime then talks to the PLC. Reaching the PLC from the laptop proves only the laptop-to-PLC path. It says nothing about the panel's address, route or equipment definition.

Path What it depends on Proof it works
Engineering PC to HMIGTO PC interface, panel address and subnet, transfer settings A transfer starts and completes
HMIGTO to M580 Panel interface, PLC address and subnet, route, enabled Modbus TCP service A runtime variable returns live PLC data

Record the configured addresses from the running M580, from the panel and from the computer, then check for duplicates and confirm each device sits on the interface you think it does. A laptop whose Wi-Fi owns the address you meant for the panel is the standard cause of "it pings from here but not from there"; the article on why a PC fails to reach an HMI when Wi-Fi owns its address walks that failure directly. If panel and PLC sit on different subnets, configure the real gateway and route instead of widening masks to fake local connectivity. If the panel is detected but the transfer still dies, the article on why an HMI is detected but project download fails covers that path end to end.

Only a completed transfer proves the engineering path — and the runtime path is still untested.

Add the PLC equipment the controller actually serves #

The runtime needs a PLC equipment entry under the target, and its driver must match the communication service the M580 application actually enables, not merely the controller family. An M580 can serve Modbus TCP from the CPU's Ethernet port or from a BMENOC0321 communication module, and each port has its own address; Schneider publishes the driver settings for an HMI talking to a BMENOC0321 over Modbus TCP/IP, and your equipment address must be the address of the port that serves your data. Modbus TCP itself runs on TCP port 502, so there is nothing to tune there — the fields that matter are the address, the register addressing and the poll setup.

  • Add the PLC equipment beneath the target in the communication configuration.
  • Pick the Modbus TCP driver when the PLC project enables Modbus TCP on the port you are addressing, and do not choose by family name when several drivers are listed.
  • Enter the PLC address from the commissioned PLC configuration, never from an old drawing.
  • Fill only the route, unit or slot fields the selected driver actually requires, taking those values from the PLC configuration or the driver documentation.

Build again and read the diagnostic: a driver that is unavailable during editing points back to installation or target support, while a driver that builds but times out at runtime points to addressing, routing or PLC-side access. Keep the map checkable by naming HMI variables after the names the PLC project exports — the guide on how M580 device DDT aliases import bulk I/O names shows how those names come across. And write the register map once, then reuse it so every client asks for the same registers; the guide on how Modbus register maps stay reusable across SCADA covers that discipline.

Worked example: reading %MW100 over Modbus TCP #

This is the smallest test that proves the runtime path on the hardware in front of you.

Inputs. Reserve %MW100 as the test word — your first unused %MW, a 16-bit register you can force from the PLC programming software. The addressing inputs come from the Modicon convention Schneider documents for its controllers: %MWn is served as the holding register numbered 40001 + n in the 1-based five-digit notation, or as offset n in a 0-based field. Read the test word on a plain diagnostic screen that shows the value and its communication quality.

Rule. To read %MWn, add n to 40001 to get the five-digit register number; a 0-based driver field takes the bare offset.

Arithmetic. For the test word: 40001 + 100 = 40101.

Result. %MW100 is holding register 40101 in 1-based notation and offset 100 in 0-based notation. Enter 40101 in a 1-based register field or 100 in a 0-based field; both read the same word, and which applies depends on the driver field, not on the controller.

Decision. Force %MW100 to a value you choose — 12345 here. If the panel shows 12345, read addressing is proven. If it shows the value of a neighboring word, you are one register off: move the address by one and retest, and leave the poll rate and timeout alone, because timing was never the problem. If quality stays bad while ping passes, the controller is not serving that register — enable the Modbus TCP service in the PLC application and retest.

Then prove the write once, on one approved, non-critical word the application logic does not own: change it from the panel, then confirm at the PLC that the word moved and that the machine's separate feedback followed, because a button animation on the screen is not proof. Keep the PLC permissives and interlocks live through the whole test — the point is to prove communication, not to work around machine protection.

Map faceplates by parameter, not by fixed address #

A reusable faceplate must receive a motor-specific reference each time an instance opens. If its internal objects stay bound to one fixed PLC variable, every instance drives the same motor; if no reference is passed at all, the faceplate shows stale or invalid data even though your direct variables communicate fine. Replace the fixed bindings inside the definition with the parameter or alias mechanism the faceplate supports, and give the definition a parameter for every reference it uses: command, running feedback, fault and each displayed value. It is the same rule that gives each instruction instance its own tag — one instance, one set of data — and the guide on why each instance needs its own tag covers that principle.

Test the first motor completely: map every parameter explicitly, open the faceplate, and confirm command, feedback and fault all follow that one motor. Then duplicate the instance for a second motor with different mappings and switch between the two; values and commands must follow the selection while the faceplate definition itself never changes. Similar variable names hide crossed bindings, so compare the command, status and fault maps separately. Match each HMI variable's data type to the PLC object while you map — a correct path still shows wrong values with the wrong type, and a REAL brings its own word-order trap covered in displaying a PLC real value over Modbus on an HMI.

What each symptom tells you #

When the panel is still dead, match the symptom to the layer and fix only that layer.

Symptom Failing layer Next move
No Modbus TCP driver in the list Target or installation Match the target to the panel label, repair the installation
.vdz export or import greyed out Project management, not the PLC Select the project root, close editors, check write access
Laptop reaches the PLC, transfer still fails Engineering path Correct the panel's address, subnet and transfer settings
Transfer completes, runtime timeouts Driver, addressing or PLC service Check the served port address and enable the service
Ping passes, quality bad everywhere Register addressing Force a test word, prove one offset, then scale up
Reads live, writes dead Write mapping Re-map the write to a word the logic does not own
Direct variables live, faceplates dead Parameter binding Replace fixed bindings with per-instance parameters

Work the table top to bottom: the first row that matches is the layer to fix, and every layer above it is already proven.

Questions asked

Does the M580 need to be stopped or reprogrammed so the HMI can read its registers?

No. Reading %MW words over Modbus TCP does not change the controller application, so the only download in this job is the HMI project to the panel. You need a PLC intervention only if you must create data the application does not already expose — reserving a new test word, or enabling the Modbus TCP service on the port the panel will address.

Can two HMIGTO panels share one M580 over Modbus TCP?

Yes. Modbus TCP is a client-server service, so each panel opens its own session to the controller and no serial multidrop wiring or address sharing is involved. Give each panel its own equipment entry and its own poll list, and keep the combined poll rate reasonable, because every added panel adds read traffic to the same link and the same controller.

The original vendor kept the HMI files — can the panel still be made to work?

Yes. The panel runs whatever runtime you transfer to it, so with the vendor's files lost you rebuild against the same target, the same Modbus TCP driver and the same register map the running PLC program already serves. The controller side needs nothing new, which makes the addresses in the live program the authoritative source for every HMI variable you recreate.

Where does the REAL data type change the addressing?

A REAL is 32 bits, so it occupies two consecutive 16-bit registers rather than one: a real value beginning at %MW200 needs %MW200 and %MW201, and the HMI variable must read both in the order the controller serves them. If the displayed number has a nonsense magnitude but plausible-looking digits, the two words are swapped rather than the address being wrong.

Keep reading

  • How-to

    Connecting an HMISTO735 to an M241 PLC with Vijeo XD

    Commission the HMISTO735 to M241 link in three layers: a driver and endpoint both Vijeo XD and the SoMachine project support, a proven Ethernet path with addresses and a ping, then one exposed test variable passed read and write end to end. Build screens only after that loop closes, and fix each fault at the first layer that fails.

    HMI and touchscreens

  • How-to

    Adding a Modbus RTU Driver and External Variables in Vijeo Designer

    Add one Modbus RTU equipment driver on the HMI serial port, set the baud, parity and stop bits every slave shares, and give each device a unique slave address. Then create one external variable per value, pointing it at that device and its documented register with the matching data type. Mismatched serial settings or duplicate addresses show up as read timeouts, so commission one device at a time.

    HMI and touchscreens

  • How-to

    Connecting an HMI to a GE Fanuc PLC Over Ethernet SRTP

    Yes. GE controllers speak SRTP, their native Ethernet protocol, on TCP port 18245, and the HMI editor's built-in GE Fanuc SRTP driver needs no Modbus TCP and no purchased add-on. VersaMax CPUs with integrated Ethernet serve SRTP only, and a Series 90-30 needs CPU firmware 8.xx before its IC693CMM321 Ethernet module will cooperate.

    HMI and touchscreens