Sort the causes before you book a lift. Log every shutdown for 14 days with its clock time and the standby light's state: repeats at the same minute point to a timer or schedule, scattered drops point to heat or power. Probe the air behind the wall on a warm afternoon; above about 40 °C heat is the working suspect, and the fix is airflow that brings the cavity to 30 °C or less.
Twelve screens that switch off a few at a time, on no pattern you can see, are sorted by two tests before anyone books a lift. First, log every shutdown with its clock time and the state of the screen's standby light. A shutdown that repeats at the same minute points at a timer or a schedule; scattered shutdowns point at heat or power. Second, tape a temperature probe to the back of a top-row screen and read it on a warm afternoon, against the rating the panels are built to. Work in that order, because the extenders and the players are the last suspects on this fault, and a wall of this size is expensive to reach.
Symptoms and what each one points to
The screen's state after a drop tells you which path failed. Sort every event into one of these before you test anything:
- Standby light lit, panel dark. Something switched the screen to standby: a timer, a schedule, a command from the management software, or an overheat trip. Power reached the panel.
- No light at all. AC stopped upstream: a breaker, the distribution feeding the wall, or the panel's own supply. This is the electrician's path, and repeat faults belong to one.
- Picture black, panel lit. The signal path dropped out, the extenders or the players. With signal-loss behaviours switched off in the screen menus, a signal dropout leaves the panel lit, so this is a picture fault, not a shutdown.
- Same minute, different days. A schedule firing by clock. Daily and weekly timers both show this way.
- Different screens, different times, no clock pattern. Heat first, per-screen power second. On a wall set into a shallow niche, heat leads.
- Every screen dark at once, after a weekend or a storm. A mains interruption. Screens that wake to standby wait for a command, which is why they come back one by one through the management software.
The log: every shutdown with its clock time
Record date, clock time to the minute, screen position by row and column, and the standby light's state afterwards. Run the log for 14 days; one full week is the floor, because a weekly timer needs its weekday to come round again.
Read the log against the clock. Two shutdowns at 03:00 on two different Tuesdays are a weekly schedule. Two at 07:30 on consecutive days are a daily one. Mark every repeat and clear it before you look at hardware.
A schedule that fires at a strange hour points at the display's clock. A clock without battery backup resets when power blinks, and the on/off timer then fires against a wrong time, differently on each screen. That is how one panel drops at 14:10 and its neighbour at 15:40, and how a schedule looks random while it drifts. Set every screen's clock right, clear the on/off timer and any eco or power-saving schedule in each screen's menu, and clear the schedules held in the management software. Check the players too, because a player can switch a screen off over HDMI-CEC on a schedule, and that command path works even after every screen menu is clean. A daily on/off timer behaves the way a retail display scheduled on and off each day behaves, so the timer you are hunting may be one somebody set on purpose and forgot.
Keep logging for another week after the clearing. A pattern that stops is the schedule found. Shutdowns that continue with no clock pattern send you to the probe.
The probe: air temperature behind the wall
Standard LCD modules are typically rated to operate from 0 to +50 °C, and that is the number to read the wall's air against. The display's own heat then adds to the air around it: count on the inside of a working panel running 10 to 20 °C above the air behind it, 10 with free air and a moderate backlight, 20 in a tight cavity. That arithmetic sets the working threshold at about 40 °C of air behind the wall. Past it, the gap to the 50 °C end of the rating is gone before the panel's own rise is counted, and what remains is overheat trips and shortened panel life.
Tape a contact probe to the back panel of a centre screen in the top row, because heat stratifies and the top of a wall runs hottest. Read it on a warm afternoon, after the wall has been switched on since the morning, and write the room temperature beside it. Above about 40 °C, heat is the working suspect. Between 30 and 40 °C, heat can still build on the warmest days, so keep the probe readings beside the shutdown log. Under 30 °C, heat is off the table and power moves up the list.
Read the log's geography too. Shutdowns clustered on the top row or the centre columns are heat's fingerprint, because those panels sit in the hottest air.
A shallow niche is the usual reason the air behind a wall runs hot. Twelve large panels together put kilowatts of heat into a cavity with one open side, and a closed top lets it pool. The same tight-gap physics shows up behind a flush wall mount, where the clearance behind the panel decides what fits and what breathes.
Kill the wall's circuits at the breaker before you reach behind any panel, and let the access crew do any reaching their platform is built for.
Example: twelve screens in a shallow niche
A 2 by 6 wall of twelve commercial displays sits in a cavity about 150 mm deep, in a room held at 22 °C. The players sit in a rack and reach the screens over HDMI over Cat6 extenders, and the screens are managed over the network.
The 14-day log holds 9 shutdowns, each on a different screen, at times from 13:20 to 16:45. No minute repeats on any two days and no weekday repeats, so no schedule fits. Every dropped screen's standby light was lit afterwards, so power reached each panel and something switched it to standby or tripped it. On a test afternoon the probe taped behind a top-row screen reads 43 °C at 14:00, with the room at 22 °C.
The arithmetic against the rating: 43 °C of cavity air plus the panel's own 10 to 20 °C rise puts the inside of the panels at 53 to 63 °C (43 + 10 = 53, 43 + 20 = 63), past the 50 °C end of the standard rating. Scattered afternoon times fit heat building through the day, and lit standby lights fit an overheat trip.
The decision: cool the cavity before touching anything else. Target 30 °C or less behind the wall, because 30 + 20 = 50 lands the panel insides at the rating's edge even with a full 20 °C rise. If the log had shown shutdowns at 03:00 on three consecutive Tuesdays with the probe reading 28 °C, the same two tests would have sent the work to the timers, and the extenders and players would have stayed in the rack.
The fix for each cause
Schedule. Set every screen's clock from the management software or by hand. Clear the on/off timer and any eco schedule in each screen's menu, clear the schedules held in the management software, and check the players for a CEC power-off schedule. Re-log for a week; a pattern that stops is the cause found.
Heat. Give the cavity a path for air: an opening low down and an opening high up, so warmed air can rise out, or a quiet extraction fan at the top pulling air through. Re-probe on the warmest afternoon of the next fortnight. The target is 30 °C or less behind the wall.
Power. Screens found dark with no standby light point upstream: the circuit feeding the wall, the connections behind the rack, and any switched outlet or timer socket on that circuit. Hand repeat faults to an electrician with the log attached, because the logged times are the evidence an intermittent supply fault needs.
Recommendation
Recommendation: Run the log and the probe before you book a lift. Log every shutdown for 14 days with clock time, screen position and the standby light's state, and probe the air behind the wall on a warm afternoon. Same-minute repeats mean timers and schedules: set the clocks, clear every timer, and re-log for a week. A probe above about 40 °C with scattered shutdowns means heat: ventilate the cavity to 30 °C or less behind the wall. Screens dark with no standby light mean the AC path, and an electrician. The extenders and players stay in the rack until the log and the probe have had their say.
Questions
Why do the screens come back only through the management software?
A commercial display that loses AC, or is switched off by a command, wakes into standby and waits. Most display menus carry an option for what the screen does when power returns. On the keep-last-state setting, a dark screen stays dark until a network command or the power button wakes it, which is exactly the behaviour you are seeing. Set that option to power on, and a mains blink stops leaving the wall dark until someone commands it back.
The room is air conditioned, so how can the wall overheat?
The thermostat reads room air, and the panels live in the cavity behind them. Twelve 55-inch commercial displays draw on the order of 100 to 200 W each, which is 1.2 to 2.4 kW of heat going into that cavity. Air conditioning that never reaches the cavity leaves it well above the room. The probe settles it: 22 °C in the room with 43 °C behind the wall is a wall the air conditioning has not touched.
How long does the log need to run before the pattern means anything?
One full week at minimum, because a weekly timer fires by weekday and needs its day to come round again. Two weeks is better: it catches daily and weekly schedules and gives the warmest afternoons time to produce shutdowns. Log the quiet days too. Shutdowns that stop during a cool spell and return when the room warms up are heat showing its pattern in your own data.
Planning information, not legal or engineering advice. Rules differ by place: check codes and permits with your local authority and your licensed trades.