Solutions: Your Engineering Bible

Every credible power solution starts with one thing the cheapest suppliers skip: an honest understanding of the load. That means a demand profile — not a nameplate tally, but how the load actually behaves through the day, the largest motor that can start while everything else runs, the power factor, the harmonic content, and the growth expected over the next few years. A connected-load total can overstate the real demand by a wide margin because not everything runs at once; a diversity (demand) factor turns the nameplate sum into the figure the system must really serve.

Skipping this step is how sites end up with equipment that is simultaneously too big (idling, wet-stacking, wasting standing losses) and too small (collapsing on the worst motor start). We measure with data loggers where the load already exists, or build the profile carefully from the schedule where it does not — because every downstream decision, from genset frame size to cable to tariff, rests on this number being right.

A generator, a UPS, a solar array, an ATS and a distribution board bought separately are not a power system — they are a pile of boxes that may or may not cooperate when the grid drops. The single-line diagram (SLD) is where they become one designed system: it shows every source, switch, protective device and load, how the changeover sequences, how the UPS bridges the genset start, how the solar ties in without back-feeding, and where the system can be isolated safely for maintenance.

This integration is where most field failures are quietly designed out — or designed in. A UPS that fights the generator, a transfer switch that is itself a single point of failure behind a redundant UPS, a solar inverter with no anti-islanding interlock: these are integration faults, invisible on any individual datasheet and obvious on a well-drawn SLD. We design the whole chain on one diagram, and that diagram becomes the document the installation, the commissioning and every future engineer works from.

A compliant installation is designed to recognised standards (IEC 60364 / the Kenyan equivalents) and proven by verification, not assertion. Cables are sized for current capacity and volt-drop and fault withstand; protective devices are selected and graded so a downstream fault trips the nearest breaker and leaves the rest of the site live; earthing is designed and tested so faults clear and people are safe. These are not optional refinements that separate a professional install from a dangerous one — they are the install.

The output is a documented, certifiable system: test certificates, protection settings, an as-built SLD and an O&M manual. For regulated clients this is what satisfies the auditor; for everyone it is what lets the next engineer work on the system safely years later. We hand it over as a matter of course, because an undocumented installation is a liability waiting for an incident.

Equipment that has never been tested under real conditions is a hope, not a system. Proper commissioning runs a defined sequence: factory acceptance where it applies, then on-site checks of every connection and setting, a load-bank test that proves the generator carries full rated load (and burns off any wet-stacking), a simulated mains failure to verify the changeover and UPS bridge happen cleanly and within the specified transfer time, and a check that protection grades as designed. Only then is the system signed over.

This is the step that separates a working installation from one that fails on its first real outage — the day nobody can afford a surprise. NFPA 110 formalises it for standby systems precisely because untested systems fail when tested by reality. We commission to a written protocol and leave the client the records, so the first time the system proves itself is on the test bench, not in a crisis.

The day the system is commissioned is the start of its working life, not the end of the engagement. Standby systems fail quietly between outages — a flat battery, stale fuel, a drifted setting — so reliability is sustained by a maintenance SLA: scheduled service by running hours or calendar, periodic load testing, remote monitoring where it earns its keep, and locally-held fast-moving spares so a fault is measured in hours, not weeks. Availability, as the maths shows elsewhere on this site, depends as much on speed of repair as on the quality of the iron.

This is why we sell partnerships rather than boxes. One contract, one SLA, documented service history, and the same engineers who designed the system maintaining it — across one site or fifty. It is the least glamorous part of power engineering and, over a ten-year life, the part that decides whether the investment actually delivered the resilience it promised.