The Level 2 method for monitors: logs, technical alarms, parameter isolation and swap tests

Read a monitor's technical alarms, error history and boot behaviour, place the fault in a host, a channel, an accessory or the installation, isolate one parameter at a time with a simulator and swap tests, and know which settings and service functions you may and may not touch.

Where Level 1 ends

Patient Monitor Service — Level 1 gave you the measuring principle of every channel, the accessories and their compatibility traps, the planned maintenance sequence, simulator and manometer checks, and the IEC 62353 electrical safety test. It also gave you the first rule of monitor service: when one channel misbehaves, swap the accessory for a known-good one before you suspect the monitor. Most calls end there. Level 2 starts with the calls that do not: the error that survives a new cable, the SpO2 that reads low on one monitor only, the NIBP that times out with every cuff, the monitor that reboots twice a day, the bed that drops off the central station every night.

The method is the one taught in the Fault-Finding Method course: facts first, then the subsystem, then cheap and likely causes before expensive ones, one change at a time, and a proof at the end. What is specific to monitors is the structure of the device. A monitor is a host (processor, display, alarm sound and lights, power supply, battery, network interface) plus several independent measuring channels, each with its own front end, its own accessory and its own way of failing, plus an installation it depends on: mains, earth, the central station, other equipment around the bed. The expensive mistake at this level is to believe the block that raised the message. A noisy ECG is reported by the ECG module, but the cause may be an unearthed power supply. A 'module error' may be electrostatic noise on the link to an input unit, not a dead module.

Read what the monitor already knows

Before you open anything, collect three kinds of evidence. The first is the technical alarm on the screen at the moment of the fault. The alarm standard IEC 60601-1-8 separates physiological alarms (the patient) from technical alarms (the equipment); manufacturers call the latter technical alarms, INOPs, equipment messages or system messages. Write the exact wording down. 'Check electrodes', 'ECG noise' and 'ECG module error' sound alike to a nurse and point to three different places.

The second is the error history. Most current monitors keep one, with a time stamp for every self-test failure and every error detected in operation. Nihon Kohden's BSM-6000 series, for example, shows it on a history page of its diagnostic check screen and keeps it until a factory initialisation; Philips IntelliVue monitors keep a status log that the manufacturer's support tool can download. Read the whole history, not only today: a module reset every few days for a month, a battery event before every unexpected shutdown, a communication error with the same input unit, are the past of today's fault. Photograph or export the history before any reboot, reset or software action, because some of those actions clear it.

The third is the boot sequence. A monitor that starts normally passes through visible stages: power indicator, alarm lamps, the logo, a lamp test with a start-up tone, the main screen, then the first measured values. Service manuals describe that sequence for each model; Philips gives it with approximate times for the MX400-MX550 family. The stage at which a start-up stops tells you which block failed to report: no power lamp points to the supply side, no start-up tone to the speaker or its driver, a hang on the logo to the main board, its memory or its software. Check the date and time while you are there. A wrong clock makes the history unreliable and often means a flat backup cell on the main board.

Then ask the people. What exactly did they see, on which bed, with which patient category and accessories, after what change: a new cable supplier, a moved bed, a power cut, a software update, a monitor that was dropped or wet, an electrosurgical unit used in that room?

Place the fault before you test it

Sort the complaint into one of four places. It costs nothing and it decides which tools you take.

Many complaints are not equipment faults at all. A wrong electrode count, the respiration measurement switched off, a demo mode left on, pacing detection off for a paced patient, an IBP channel labelled for the wrong pressure: the monitor is doing exactly what it was set to do. Check the active settings against what the ward expects before you suspect hardware.

• The whole monitor: no power, no boot, restarts, no sound, blank or frozen screen, no network. Look at the host: supply, battery, main board, display, speaker, network interface.
• One channel on one monitor: one parameter fails, the others work. Look at that channel's accessory, socket, front end or module.
• One function on many monitors: every bed in a ward loses the central station at night, or every monitor in one theatre shows noise. Look at what they share: network, central station, mains, earth, the electrosurgical unit.
• One place, any monitor: the fault follows the bed space, not the monitor. Look at the installation: socket, bonding, nearby equipment, Wi-Fi coverage.

Isolate one parameter at a time with a simulator

A simulator is your known-good patient. At Level 1 you used it to verify that every channel passes. At Level 2 you use it to cut the signal path into pieces. The chain of each channel is patient, sensor or electrodes, extension or trunk cable, input socket, front end, module link, host. Inject a known signal at different points and see where it stops being right.

Start with the simulator at the input socket, through the shortest cable the simulator allows. If the channel is correct there, the module and host are working and the fault is in the patient side: accessory, sensor, electrodes, cuff, or the way they are applied. Then add the ward's own cable between the simulator and the socket. If the fault appears, it is in that cable. Nihon Kohden's troubleshooting tables follow this logic for SpO2: replace the sensor and cord first, then put the vital-sign simulator on the socket; if the message disappears, the monitor and its accessories are not the problem, and if it stays, the SpO2 board is.

Some channels allow an even cheaper split. For ECG, the same manual suggests joining all the lead-wire ends together: with the electrodes and the patient removed, a 'check electrodes' or noise message that remains points to the input unit. For a channel with several identical sockets, move the cable and sensor to another socket: if the fault moves with the cable, it is the cable; if it stays with the socket, it is the channel.

Two rules keep this honest. The simulator must match the channel: the right SpO2 technology setting, the right IBP interface cable, the right thermistor family. And a simulator that passes does not clear a complaint about a real patient; it clears the equipment and moves the question to the accessory, the application or the clinical situation, which you then report as such.

Swap tests: modules, sockets and accessories

Swapping identical parts is the fastest test a monitor offers, because so much of it is modular. A plug-in parameter module can be moved to another rack. A multi-measurement unit or input unit can be moved to another host. An accessory can be replaced with a labelled known-good one from the workshop. The rule is the one manufacturers write into their own service guides: replace one suspect with a known-good part, test, and if the symptom stays, put the original back before you try the next one. A pile of swapped parts with no record of what changed is how a workshop loses a week and a good module.

Check compatibility before every swap. A module that fits the rack is not automatically supported by that host's software revision; manufacturers publish compatibility tables for hardware and software combinations, and some replacements need a software or configuration step before they work. Never move a module into or out of a monitor that is connected to a patient, and never borrow a module from a monitor in use. The ward swaps the monitor first; you work on the free one.

Swaps give answers in both directions. If the fault follows the module to a good host, the module is faulty. If a good module shows the same fault in this host, look at the host, its rack connector and its link. If neither changes anything, you are probably looking at the installation or the settings.

What you may and may not change

At Level 2 you will see service menus, configuration screens and calibration functions. The line between what you may touch and what you may not is not a matter of skill, it is a matter of authority and procedure.

Clinical settings belong to the hospital: alarm limits and defaults, alarm volumes, profiles, units and bed labels. You record them before service and restore them after, as in Level 1. Configuration settings that make the monitor fit the installation, such as mains frequency, enabled parameters, the number of electrodes, measurement modes and network addresses, are set by the procedure for your authorised level and by agreement with the owner; if a wrong one is the cause, correct it and write down what you changed. Service functions such as factory initialisation, calibration of a channel, firmware loading or option settings are used only by the manufacturer's procedure, by people trained and authorised for them.

A factory reset is the clearest example. It returns the monitor to defaults that do not match the hospital's configuration and, on some models, clears the stored history and service settings too; one field case below started with a reset meant to clear an error and ended with a monitor that could not show its parameters until the configuration was rebuilt. Do not use passwords, service keys or option activation you were not given, and do not copy configuration files or firmware images between monitors outside the manufacturer's procedure: they can carry unit-specific data and option settings that do not belong to the target. Do not turn potentiometers that the manufacturer marks as factory-set; Criticare, for example, says this about the ECG baseline trimmers on the 8100E analog board.

When you reach your line, stop and hand over a package the next level can use: the technical alarms and history, the conditions of the fault, your simulator results at each point, what you swapped and what changed, and what you suspect.

On-site checklist

• Work only on a monitor that is off the patient.
• Write down the exact technical alarm text and save the error history before rebooting.
• Watch the full boot sequence and check date and time.
• Place the complaint: whole monitor, one channel, one function on many monitors, or one place.
• Check the active settings before suspecting hardware.
• Use the simulator at the socket first, then through the ward's cable and sensor.
• Swap one part at a time, check compatibility, and put the original back if nothing changed.
• Record configuration before any reset; stay within your authorised service functions.

Check yourself