IMDHMIControls LibraryShop main store

Troubleshooting

TIA Portal v14 cannot go online with an S7-1500

PLC programming Published 7 min read

Short answer

Only a complete power cycle restores engineering access: a STOP-to-RUN transition or a port change leaves the failed download service in place. Power down under your site's procedure, confirm recovery with a read-only online session before any download, then capture CPU firmware, TIA Portal version and download details so the next occurrence can be classified as project-triggered or a software defect.

On this page
  1. What the error text names, and what it does not
  2. Check the workstation and network path first
  3. Confirm the boundary before touching the controller
  4. Recover with a controlled power cycle
  5. Working the evidence: four lockups in five months on two controllers
  6. Capture this before the next restart
  7. Where to put the effort while support works the case
  8. Questions asked

Only a complete power cycle clears this fault. When a download to the S7-1500 ends in a "download to device" error with "service is not started", every TIA Portal v14 station loses engineering access while the program keeps running and the HMI keeps updating; a STOP-to-RUN transition or a port change does nothing. Power the controller down under your site's procedure, restart it, and prove recovery with a read-only online session before anyone downloads again. Then work it as an open defect: the same project has produced this lockup on two controllers, on two different switches, in two different countries, so capture the firmware, project and download evidence that separates a project-triggered condition from a firmware or TIA Portal defect.

What the error text names, and what it does not #

"Service is not started" names the service the download requested, not the component that stopped. Read it as "the engineering endpoint refused the session", and locate the fault boundary from the behavior around it:

Observation What it proves Next check
Program keeps executing CPU and user program are healthy Watch a live process value; no mode change needed
HMI values keep updating Ethernet path and at least one communication service work Confirm values are new, not a cached screen
Every programmer fails to go online The fault is not one workstation Read-only test from a second station
Full shutdown restores access Power-on initialization clears the failed state After restart, go online read-only before downloading

Engineering access and HMI traffic use separate protocol endpoints and resources on an S7-1500, so normal machine operation proves nothing about the download service. That separation is why this lockup looks like a network problem and is not one.

Check the workstation and network path first #

Complete these before disturbing the controller:

  • Stop all download attempts and nominate one workstation for testing. One failed download leaves several clients watching the same controller-side state, and parallel retries add traffic that pollutes the record.
  • Confirm the engineering software uses the intended network adapter and the configured controller address.
  • Check link indications at the workstation and switch, and review switch counters for lost link, errors, or a port transition at the failure time.
  • Test IP reachability where network policy permits. A reply proves the IP path, not that the engineering service accepts sessions.
  • Run device discovery through the intended adapter; discovery without an online session narrows the fault above the physical network.
  • Repeat the read-only online test from a second workstation. Never repeat a download.

If one station alone shows link but never connects, that is a different fault: work the single-station steps in S7-1200 link lights but TIA Portal cannot connect. In your case the physical layer is already half-eliminated: two controllers, two switches, two countries, one project. Normal results from two independent stations make a workstation or cable cause unlikely; proceed to the controller.

Confirm the boundary before touching the controller #

  • Verify the HMI displays newly changing values. A static screen can hold its last received values after communication stopped.
  • Confirm execution from a process indication already visible to operations, such as a sequence step or a changing production value, not from a new command invented for diagnosis.
  • Record the exact action underway when access failed: connection, compile, download preparation, or transfer.
  • Record the complete error text, screenshots and the workstation's local time.

Do not cycle STOP-to-RUN. The S7-1500 documentation describes a warm restart as a menu-triggered restart with the CPU in STOP; it reinitializes execution, and in this lockup it left engineering access dead. Do not swap CPU ports either: the failure survived the swap in the reported case, which is what you expect when both ports sit behind the same communication subsystem.

If the HMI values freeze along with engineering access, you are on a broader communication fault: start with why HMI values show question marks: PLC comms timeouts.

Recover with a controlled power cycle #

A power cycle stops control execution and de-energizes outputs exactly as the installed design behaves on power loss, so run it under your site's operating and electrical procedures with the machine in its defined safe state.

  • Obtain operating authorization and place the machine or process in its defined safe shutdown condition.
  • Save the project, and preserve the screenshots, timestamps and network observations before removing power.
  • Isolate controller power by the approved method and confirm full de-energization from the indicators. No specific wait time is established for this case; what matters is a true power removal, not a timed blink.
  • Restore power, let startup finish, and confirm process readiness.
  • Open a read-only online session from the nominated station before any download.
  • Where your workflow permits, compare the online state with the controlled project copy.

Expect startup to complete, HMI updates to resume, and the online session to open without the error. The restart clears the lockup, not its cause.

Working the evidence: four lockups in five months on two controllers #

Inputs, from the case record itself: 4 lockups on one controller in roughly 5 months; a second controller in another country, on a different switch, showing the same lockup with the same project; every engineering station locked out each time; HMI traffic and program execution unaffected; only a power cycle recovers access.

Rule in words: divide the observation window by the number of events to get the exposure between events, and count the independent network paths to place the common cause.

Arithmetic: 5 months x 30 days = 150 days. 150 days / 4 lockups = 37.5 days per lockup. 37.5 days / 7 = 5.4 weeks. Expect the next lockup within five to six weeks of the last one.

Decision: with the threshold at two or more controllers on independent switch paths reproducing the same lockup on the same project, hardware, cabling and a single switch drop below the line. Classify the case as project content or toolchain, meaning TIA Portal v14 or CPU firmware, instrument the downloads, and treat any recurrence with a different project as the signal to move firmware to the top of the list.

Capture this before the next restart #

  • CPU order number and installed firmware version, read from the installed equipment.
  • Exact TIA Portal edition and service pack: v14 and v14 SP1 are different evidence.
  • The failing operation step, the full error text, screenshots, workstation time, and the network path used.
  • How many engineering clients were connected or downloading at that moment.
  • Free communication resources on the CPU after recovery. A Siemens support discussion of a related PLC-to-HMI communication loss recommends exactly this check after the error occurs.

The same signature is reported publicly for a CPU 6ES7 516-3AN01-0AB0 at firmware v2.0 under TIA Portal V14 SP1, so record your own identifiers rather than substituting another machine's. One documented trigger to rule out first: Siemens documentation states a download to an S7-1200 or S7-1500 can require a module reset when the versions of the modules in the project differ from the target, so verify the project's module versions against the installed firmware before the next download.

Where to put the effort while support works the case #

  • Remove concurrent engineering sessions. One nominated station performs every download while the others stay offline during the transfer. The one condition present at every event so far is a download on a CPU shared by several programmers; that is a hypothesis to test, not a proven cause.
  • Align project module versions with the installed firmware, because a version mismatch is a documented reason a download ends in a reset demand.
  • Upgrade the engineering toolchain once, to a version that both programs your controllers and can service their firmware; firmware servicing is gated by the TIA Portal version that supports it. If you are weighing which edition to move to, what STEP 7 is in a PLC project, and which one you need sorts the variants apart.
  • Keep the capture list ready at the cabinet from the day of recovery, because the five-to-six-week interval opens the next evidence window soon.

Questions asked

Why does the HMI keep working when nobody can go online?

Engineering sessions and HMI traffic use different communication services on the CPU, so one can fail while the other runs. The pattern fits already-established connections continuing while new session setup is refused, which is why every fresh online attempt fails at the same time. A live HMI proves the Ethernet path and program execution only; it never certifies the engineering service.

Will a warm restart or a port swap clear the lockup?

No. The S7-1500 documentation defines a warm restart as a menu command from STOP mode, and it reinitializes execution, not the failed engineering service; both CPU ports also sit behind the same communication subsystem, so swapping ports changes nothing. Only full power removal has cleared it, and the lockup can return on a later download, so plan the safe state before each recovery.

Is the defect in the CPU firmware or in TIA Portal?

No one can assign it from the message, because the text names the service the download requested, not the component that failed. The two levers you control are the project and the firmware: a lockup with a different project points at firmware, and one that appears after a firmware update points at the toolchain. File your captured case with vendor support and let that analysis assign it.

Can the CPU firmware be updated from TIA Portal v14?

Only to the firmware that version supports. In the reported case, updating that CPU's firmware required TIA Portal V15, so the engineering toolchain gates the fix rather than the controller alone. One engineering-software upgrade chosen to cover your controllers and the firmware you intend to run unlocks both the update path and a newer service stack.

How soon should we expect the next lockup?

Four lockups in about five months on one controller works out to one event roughly every five to six weeks. Keep the capture list and screenshot procedure ready at the cabinet from the day of recovery. The next occurrence is your best evidence, so capture it completely before you start the recovery restart.

Keep reading

  • Troubleshooting

    S7-1200 shows link lights but TIA Portal cannot connect

    Green Link and blinking Rx/Tx on the S7-1200's PROFINET port prove cable and traffic only, never a working TIA Portal connection. The fault is almost always the wrong PC adapter or a subnet mismatch: select the cabled interface as the S7ONLINE (STEP 7) access point, scan Accessible devices, give the PC a unique address in the CPU's subnet, then go online. If the scan shows an unconfigured CPU, assign its IP address and PROFINET name first, then connect.

    PLC programming

  • Explainer

    What STEP 7 is in a PLC project, and which one you need

    STEP 7 is the engineering software you use to program, configure, diagnose and download to SIMATIC PLCs — a workstation tool, not a part of the PLC itself. SIMATIC Manager, also called STEP 7 Classic, configures S7-300 and S7-400 CPUs; STEP 7 inside TIA Portal configures S7-1200 and S7-1500. Read the CPU family from the module label or the existing project, open the matching environment, and confirm the project's hardware configuration matches the installed CPU before you edit or download.

    PLC programming

  • How-to

    Write an S7-1200 heartbeat in ladder logic in TIA Portal

    Tie the light to a detector, not to the heartbeat level: time the on half with one TON and the off half with a second TON, both at three times the measured state time, and OR the two done bits into one communication-fault coil. With a PC heartbeat that changes once per second, PT is 3 s and the fault trips 3 s after the last change; add TCON and TDIAG supervision when the PLC owns the TCP connection.

    PLC programming