FANUC Teach Pendant Black Screen, Deadman Issues, and iPendant Lockups: Diagnosing the iPendant Without Throwing It Out

A teach pendant that boots to a black screen, a deadman switch that drops the servos at random, or an iPendant that freezes mid-edit are three of the most common service calls we get at Probot Systems. The instinct is to suspect the controller and start swapping boards. In practice, almost every TP problem is mechanical: a cable, a connector, a worn switch, or a deteriorating membrane. Skip a $40 cable inspection and you can lose a day to phantom troubleshooting.

This post is written for maintenance technicians and integrators working with FANUC iPendants on R-30iB and R-30iB Plus controllers, including the legacy iPendant on R-30iA Mate and the CRX tablet pendant on CRX-10iA and CRX-30iA cobots. The same diagnostic mindset applies across the family.

How the teach pendant talks to the controller

The iPendant is a thin client connected to the controller by a single multi-conductor cable carrying both power (24V), Ethernet for the GUI, and the discrete safety signals from the deadman switch and the e-stop button. Two of those safety signals come back as dual-channel monitored circuits, and they have to agree at every poll cycle or the controller throws a safety alarm.

The official FANUC alarm code manual entry for SRVO-206 reads: “The teach pendant deadman switch was released while the teach pendant was enabled, but the EMERGENCY STOP line was not disconnected. Remedy: Refer to the Controller Maintenance Manual for more information. Note: Need power cycle to release this alarm.” In plain English, the controller saw the deadman go open while the TP was active, and the e-stop did not trip in sync, which is what you would expect if the deadman switch were healthy.

For the related SYST-045: “The mode selector is in AUTO and the TP ON/OFF switch is in the ON position. Remedy: Turn the TP ON/OFF switch to OFF. Press RESET.” Two different alarms, both involving the same TP enable switch logic, and both pointing at the same mechanical part suite (cable, switch, membrane).

Most common causes, in order of probability

1. TP cable damaged or partially unseated. Number one. A factory floor cable that has been flexed, kicked, or pinched between cabinet doors will fail intermittently before it fails completely. The Teach Pendant Black Screen following a cable failure thread is the canonical example: complete black screen, controller still running, fix was the cable.
2. Deadman switch worn. The three-position deadman has a mechanical detent that wears out around 100k presses. When it cannot reliably hold the middle position, the controller throws SRVO-206 at random. The SYST-302 PLEASE POWER OFF and SRVO-206 Deadman Switch SVEMG ABNORMAL thread starts with the same symptoms.
3. iPendant comm card failure. Less common than the cable but real. The iPendant has its own internal comm card that fails after thermal cycling for years.
4. Software lockup during edit. The iPendant runs IE-based Java for legacy menus. Heavy edits on a big TP program (BG Logic, PalletTool3, etc.) can hang the rendering. The Roboguide Crashing thread shows the same blank-screen and “out of memory” pattern on iPendants whether the cell is virtual or physical.
5. Membrane keypad keys failing. After years of service, individual keys (often the F-keys or the most-used number keys) stop registering. Membrane is replaceable; the TP does not need to be scrapped.
6. TP locked by a stuck edit context. Pressing FCTN > END EDIT closes a held edit. If that does not clear, a cold start releases the in-memory lock. The PalletTool3 with the iPendant discussions touch on this lockup.

How to diagnose in under 15 minutes

Step 1. With the cabinet on and the TP showing the failure (black screen, frozen, or whatever), do not power down yet. Look at the controller’s POWER and FAULT LEDs. If the controller itself shows normal status and only the TP is dead, the TP or its cable is the prime suspect, not the controller.

Step 2. Power down the controller. Unscrew the TP cable on both sides (cabinet and pendant). Look for visible wear, bent pins, or strain damage right at the strain relief on the pendant connector. That is where 80% of cable failures show up.

Step 3. Reseat the cable, power up. If the TP boots normally now, replace or repair the cable as a precaution before it fails again on a less convenient day.

Step 4. If the TP still does not boot, swap with a known-good pendant at the same firmware revision. The thread error XML perse CRX10-ia/L explicitly suggests trying a standard teach pendant when the CRX tablet acts up. If the spare boots, your TP has an internal fault.

Step 5. For a deadman complaint, go to MENU > 4 STATUS > Safety Signals and watch DSWLDM and SVOFF (or equivalent on your controller) while pressing the deadman through all three positions. They should change cleanly between Released, Half-press (enabled), and Full-press. Bouncing or stuck signals point at the deadman switch.

Step 6. For a TP enabled in AUTO complaint (SYST-045), the fix is in the manual: flip TP ON/OFF to OFF and press RESET. If the alarm reappears immediately, the TP enable switch itself is bad, not the operator action.

How to fix it

Cable damaged. Order a replacement TP cable for your controller generation. R-30iB and R-30iB Plus share a cable family, R-30iA uses a different one. The OEM part is worth the price; aftermarket cables in our experience fail again within months. 

Deadman switch worn. The deadman is replaceable as a sub-assembly on most iPendant models. The thread SYST-302 PLEASE POWER OFF and SRVO-206 walks through the cable check first, then the switch replacement.

Membrane keys failing. Several third-party rebuilders will replace the membrane and return the TP within a week. FANUC also accepts pendants for refurb. The CRX Tablet Style Pendant thread covers ordering options, and worn iPendants often come back in like-new condition for a fraction of a new unit.

iPendant lockup during edit. FCTN > END EDIT first. If that does not unlock, cold start the controller. The Teach screen freezes thread describes the same symptom appearing during USB backups; the fix is to use a different USB stick (some no-name sticks cause the TP rendering to time out) or to do the backup at a time when the cell is idle.

iPendant comm card failure. This is a board replacement, not a field repair. Send the unit in.

SYST-045 won’t clear after RESET. TP enable switch is mechanically stuck or wired wrong. On older controllers, that may also be a sign the E-Stop board itself is failing.

When to call a specialist

You should bring in a specialist if any of these are true:

• You have swapped the cable and tried a spare TP, and the controller still does not show the pendant.
• The deadman behaves intermittently after switch replacement.
• You see TP issues correlated with other alarms (SRVO-206 plus SYST-302 plus SVEMG abnormal in the same log), pointing at an e-stop chain or operator panel fault rather than the TP alone.

The fix for any of those is usually a controlled start and a board swap, which is not 11pm work. contact us for an on-site intervention, or set up a maintenance preventive contract that includes annual TP and cable inspection.

Related errors to check

• SRVO-206 Deadman switch (SVEMG abnormal): canonical deadman fault, see the alarm body for the dual-channel logic.
• SYST-045 TP enabled in AUTO mode: TP ON/OFF switch is in ON while mode selector is in AUTO. Cycle the switch, RESET.
• SYST-302 Please power off: a SYSTEM severity alarm posted; cycle power to recover.
• SRVO-046 OVC alarm: shows up in logs alongside TP failures when the cell is partway through a motion that gets killed by a deadman release.
• SYST-322 Auto status check time out: appears on some cells when the TP loses sync with the controller, can be relieved through the legacy TP view.

Probot Systems is a FANUC integrator based in Lévis, Quebec. We service iPendants on R-30iA, R-30iB, R-30iB Plus, and CRX tablet pendants on CRX-10iA and CRX-30iA, across Canada and the US. If your pendant has been intermittent for weeks and you want it diagnosed before it fails completely, that is a contact us conversation.