Straight explanations of the standards and checks behind our work: the same things we'd tell you on site, written down.
"Safety audit" gets used loosely. A proper one is a defined, repeatable set of tests, not a walk-through with a torch.
At minimum, it should cover: loop impedance measurements, earth resistance measurements, insulation resistance measurements, RCD trip time and current tests, feeder load measurement against cable capacity, and testing of circuit breakers, RCDs, and ELCBs for trip time and tripping current, alongside a full physical inspection of the wiring installation for integrity and standard.
Where thermal imaging is included, it adds a second layer: hot spots at loose connections or overloaded cables show up before they fail, not after. That's the difference between an audit that documents a problem and one that prevents an incident.
The output should be a report with clearly stated remedial measures, not just a pass/fail. If a report doesn't tell you what to fix and in what order, it hasn't done its job.
Portable Appliance Testing (PAT) is often confused with a general safety audit. It's narrower and more specific: it tests the portable equipment itself: Class 1 (earthed) and Class 2 (double-insulated) appliances, not the fixed wiring of the building.
A proper PAT test includes insulation resistance measurement, earth continuity checks on Class 1 equipment, and lead polarity checks, using dedicated test equipment rather than a visual once-over.
What it doesn't cover: your distribution boards, fixed wiring, earthing system, or protective device coordination. That's the job of a full electrical safety audit. Sites with a lot of portable equipment, such as kitchens, workshops, or offices with heavy IT/appliance turnover, typically need both, on a schedule, not one or the other.
BS 7671 (the IET Wiring Regulations, currently in its 17th/18th edition lineage) is the technical baseline most electrical design and installation work in Kenya's commercial and industrial sector is measured against, alongside local wiring regulations.
In practice, "BS 7671 compliant" means the design and installation account for things like correct cable sizing for the load and run length, proper protective device selection and coordination, adequate earthing and bonding, and safe separation of systems (power, low-voltage, fire, security) where required.
It's not a document you read once. Every design we produce is checked against it before drawings go out, and every installation is tested against it before handover. Compliance is the floor, not a selling point.
Arc flash risk comes from a specific, calculable combination of factors: high fault currents, protection devices that aren't coordinated correctly, and system design that doesn't account for how much energy could actually be released at a given point.
The reason it's worth studying formally, rather than assuming standard PPE is enough, is that incident energy levels vary hugely by location in a system. A panel that's perfectly safe to work on in one configuration can carry a materially different risk after a system change (bigger transformer, longer cable run, altered protection settings) that nobody re-assessed.
An arc flash study calculates incident energy levels and arc flash boundaries per IEEE 1584, specifies the actual PPE category required at each point, and gets that information onto physical labels on the panels themselves, so the answer to "is this safe to work on" doesn't depend on memory or assumption.
The right solar system type depends on one question more than any other: how reliable is your grid connection, really?
Grid-tied systems maximise savings where grid power is reliable: you draw solar during the day and grid power otherwise, with no batteries to buy or maintain. Stand-alone systems make sense where grid access is unreliable or absent, operating independently but requiring battery storage sized to your actual usage pattern. Hybrid systems sit between the two: grid-connected for cost savings, with battery backup for resilience during outages.
None of this should be decided before an energy audit of actual consumption. Sizing a system to a guess, rather than measured daily energy requirements, is the most common way solar projects underperform their promised payback period.
