Gleneagles Hospital: Power-Quality Monitoring to Protect Medical Equipment

In a hospital, power quality is a patient-safety issue. Imaging systems, ventilators and laboratory analysers are built to tight voltage tolerances, and a voltage sag or swell that lasts only a few cycles can corrupt a scan in progress, trip a protective relay, or shorten the life of a power supply that nobody budgeted to replace. By the time a fault shows up as a failed machine, the damage and the disruption to care have already happened.

About the client

Gleneagles Hospital operates the kind of critical-care environment where electrical reliability is non-negotiable. A single building floor can carry diagnostic imaging, operating theatres, intensive care, and a clinical laboratory, each with equipment that is expensive to repair and impossible to take offline at will. Facilities and biomedical engineering teams share responsibility for keeping that equipment available, but they work in different parts of the building and rarely sit in front of a switchboard watching waveforms.

The challenge

Gleneagles needed early warning of power-quality events, voltage sags and swells, before they reached sensitive medical equipment, and a way to alert the right people instantly, wherever they were.

Power disturbances are usually invisible until something breaks. A sag from a large motor starting on the same feeder, a swell after a load is shed, or a disturbance riding in from the utility supply can pass in milliseconds. Conventional energy meters log averages and totals; they are not built to catch a sub-cycle event or to tell anyone about it in time to act. The result is a familiar pattern in hospitals: equipment behaves strangely, a service engineer is called, and the root cause is reconstructed from fragments days later, if at all.

The solution

CobiNeural continuously monitors for voltage sags and swells and sends real-time Telegram notifications the moment an event is detected. The biomedical and facilities teams get an immediate, actionable alert rather than discovering a problem after equipment has been affected.

A CobiNeural voltage-waveform capture used to confirm a swell event: the three phase voltages over the event window, with per-channel min, max, average and RMS statistics underneath.

The platform deploys as an intelligent overlay on the hospital's existing electrical monitoring, so there was no need to rip out installed metering. Two CobiNeural modules carry the work:

- Insights → Energy captures power-quality parameters, including voltage and power factor, at location and equipment level, so a disturbance can be tied to a specific feeder or piece of switchgear rather than a vague "the building had a glitch."
- Alerts turns those readings into instant notifications. Thresholds for sag and swell events are configured per location, and when one is breached the right people receive a Telegram or email message with the context they need to respond.

Because the same data hierarchy maps sites, equipment and sensors, an alert points to where the event happened and what it affected, which is what lets a biomedical engineer act instead of investigate. Where a hospital later wants to close the loop automatically, for example shedding non-critical load or signalling backup systems, CobiNeural's Actions and automation layer can extend the same monitoring into control.

The result

Real-time visibility of power disturbances, faster response, and an added layer of protection for the hospital's most sensitive and critical equipment.

The shift is operational, not just technical. Power quality moves from something discovered after a failure to something the teams see as it happens, with the alert reaching a phone rather than sitting in a log nobody reads. For a facility where uptime and patient safety are the same conversation, that early warning is the point.

Gleneagles is one of several deployments where CobiNeural turns monitoring into action. See more in our case studies, and if you run a hospital or other critical facility, ask us about power-quality monitoring.