Solar Technical Training — East Africa

What EPRA assessors actually look for in method

Assessors are watching the order of your decisions far more than the speed of your hands. This is the sequence being marked.

What an assessor is actually watching while you work.

1. 1. State the licence class scope of the work before startingSafety-fail

   Working outside the class you hold is a compliance failure regardless of technical quality.

2. 2. Verbalise the risk assessment: sources, access, weather, bystanders

   Assessors mark stated awareness. Silent competence reads as luck.

3. 3. Isolate, lock, tag and prove dead — instrument proved either sideSafety-fail

   The most common automatic fail in Kenyan practical assessment.

4. 4. Follow one diagnostic sequence and narrate the elimination

   Two candidates can reach the same answer; only one shows a method that transfers to an unfamiliar system.

5. 5. Select the right instrument and range, and say why

   Using a clamp meter where an insulation tester is required shows the concept, not just the tool, is missing.

6. 6. Complete the record as you go, in ink, with units

   Reconstructed paperwork is visible and is treated as unreliable evidence.

7. 7. Give a plain-language handover of findings and residual risk

   T2 and T3 candidates are assessed on the ability to explain to a non-technical customer.

What the assessor is watching, minute by minute

The observation sheet behind an EPRA-style practical: what earns marks while you work, and the gate that ends the assessment.

What an assessor is actually watching while you work.

Observed practical assessment — EPRA T1 to T3.

1. 1. You brief the customer and restate the fault in your own words before touching anything

   The assessor is checking that you diagnose the reported system, not the one you assumed on the drive over.

2. 2. You name every energy source out loud — PV, battery, grid, generator — before the first isolationSafety-fail

   Naming sources is the observable evidence that you have understood a hybrid energy path.

3. 3. You prove the instrument, prove dead, then prove the instrument againSafety-fail

   An assessor cannot mark an isolation they did not see verified on a known source.

4. 4. Your measurements follow the sequence you stated, and you say why each one comes next

   Marks are given for the decision order. Random sampling reads as guessing even when it finds the fault.

5. 5. You write values down as you take them, with units and conditions

   Values reconstructed at the end of the visit are treated as estimates, and estimates are not evidence.

6. 6. You state where your licence class stops and what you would refer onSafety-fail

   Working beyond class scope is a competence failure even when the work is technically correct.

7. 7. You close by summarising fault, action, result and next steps to the customer

   The handover conversation is part of the observed assessment, not an optional courtesy.

T2 hybrid-ready system — no output and a fault code

A full practical brief marked the way an EPRA-style practical is marked: sequence, measurement choice, fault priority, safety and documentation.

A full scenario brief with the scoring dimensions and the errors that end the assessment.

EPRA Technician Class T2 — practical competence under observation.

T3 hybrid / grid-tied — intermittent fault and incomplete commissioning evidence

A T3-level brief where the fault and the paperwork are both on trial: control the sources, separate a real intermittent fault from a process failure, and rebuild the evidence pack with measured values.

A full scenario brief with the scoring dimensions and the errors that end the assessment.

EPRA Technician Class T3 — commissioning and handover under observation.

T1 service call — standalone solar home system reported dead

An entry-level observed scenario: arrival, safe method, structured checks and a report, on a system a T1 technician may legitimately work on.

A full scenario brief with the scoring dimensions and the errors that end the assessment.

EPRA Technician Class T1 — standalone systems under observation.

EPRA method cards — T1, T2, T3

The four method answers assessors ask for in one form or another, with the reasoning and the errors that cost marks.

Correct answer, why it is correct, and the mistakes people make.

T1 arrival method, T2 diagnostic order, T3 evidence pack, and the safety gate that applies at every level.

• T1 — You arrive at a small standalone solar home system and the customer says it is not working. What are the first four actions, in order, before any tool touches the live system?

  1) Introduce yourself and take the history from the customer. 2) Visually identify every energy source — PV, battery, and any generator or grid supply. 3) Isolate and prove dead at the point of work. 4) Prove your instrument on a known source before and after the test.

  Each step removes an assumption. The history narrows the fault, the source survey stops you isolating only half the system, isolation makes the work safe, and proving the instrument makes the dead reading trustworthy.

  • × Starting with the multimeter before asking a single question
  • × Isolating the PV and forgetting the battery is still a source
  • × Reading zero volts on an instrument that was never proved

• T2 — An inverter shows a generic fault code and produces no power. Battery voltage looks normal. What is the correct diagnostic order?

  Isolate and prove dead where the work requires it, verify DC input voltage and polarity at the inverter, record the exact fault code and look up its precise meaning, check string voltages and connection integrity, and only then move to battery charge settings or communications.

  The order runs from the cheapest, most likely and most dangerous checks outward. Replacing an inverter before the DC side is understood is how a wiring fault gets paid for twice.

  • × Reading the fault code as a diagnosis instead of a symptom
  • × Trusting a resting battery voltage as proof the bank is healthy
  • × Jumping to a firmware or communications theory before measuring

• T3 — You are commissioning a hybrid system and the assessor asks you to demonstrate it is safe and correctly commissioned. What evidence should you be able to present?

  Isolation and lock-out record; measured Voc, Vmp and Isc or operating current per string; polarity confirmation; earth continuity and insulation resistance results; inverter configuration and protection settings; functional anti-islanding / grid-interaction test; and a signed commissioning sheet carrying measured — not estimated — values.

  Commissioning is a claim about the system. Without measurements taken under stated conditions there is nothing to defend the claim with, and nothing for the next technician to compare against.

  • × Writing typical or nameplate values instead of what was measured
  • × Omitting the functional protection test because the inverter 'has it built in'
  • × An unsigned or undated sheet, which carries no evidential weight

• All levels — A safety-critical error occurs during a lab or assessment, for example working on a conductor without proving dead. What is the grading outcome?

  Automatic fail of that lab or assessment, regardless of how well everything else was performed.

  Safety method is not a scoring category that good technical work can compensate for. One unsafe act is the whole result.

  • × Assuming a strong technical answer buys back an unsafe step
  • × Treating the inverter display being off as proof of isolation
  • × Skipping the battery-side isolation because 'it is only 48 V'

East African handover pack

What a Kenyan customer, a utility and an assessor each need to see after the job.

The paperwork, filled in the way it will be audited.

• Commissioning record (EPRA-ready)

  Evidence the system was tested to spec by a technician working within their class.

  • · Technician name, licence class and number
  • · System configuration as built, including storage
  • · Isolation and prove-dead record
  • · Measured values with irradiance, temperature and time
  • · Protection settings and any grid-interaction configuration
  • · Customer handover and residual-risk briefing
  Open the printable document →

• Service visit report

  The narrative record that turns a repeat call into a solved problem.

  • · Symptom as reported and conditions on arrival
  • · Checks performed, in order, with results
  • · Diagnosis and supporting evidence
  • · Work done, parts used and verification
  • · Recommendations and next service date
  Open the printable document →

Evidence-pack exercise — rebuild what the assessor will ask for

A working exercise, not a blank form: take a completed job and assemble the pack that proves it was done properly.

The paperwork, filled in the way it will be audited.

EPRA T2/T3 evidence expectations and NITA practical assessment records.

• Isolation and safe-working record

  Exercise: reconstruct the isolation you performed on your last hybrid job. If you cannot name every source you controlled, the record is incomplete.

  • · Sources identified — PV, battery, grid, generator
  • · Isolation points used and the order applied
  • · Instrument used and its prove-test result before and after
  • · Locks, tags or physical control applied
  • · Who else was on site and how they were kept clear

• Measured values sheet

  Exercise: write down only values you actually took, with units and conditions. Leave gaps visible instead of filling them in from memory.

  • · String Voc and Isc per string, with time, irradiance and temperature conditions
  • · Insulation resistance results and test voltage
  • · Earthing and bonding continuity results
  • · AC voltage and frequency at the point of connection
  • · Battery voltage, state of charge and configured charge parameters

• Functional test and settings record

  Exercise: prove the system behaves as designed, including the protection you cannot see working.

  • · Protection and anti-islanding behaviour observed
  • · Inverter and charger settings as configured, not as shipped
  • · Changeover and backup behaviour under a simulated loss of supply
  • · Monitoring baseline captured with date and time

• Customer handover and referral note

  Exercise: write the page the customer keeps. An assessor reads it to see whether you communicated, not just repaired.

  • · Fault reported and fault found, in plain language
  • · Work carried out and parts changed
  • · Outstanding items and anything referred beyond your licence class
  • · Date, technician name, licence class and signature

EPRA T1–T3 practice questions

Licence-scope, safety and hybrid-system questions in EPRA phrasing.

Exam-style questions with the reasoning behind every answer.

Supports EPRA Technician Class T1 → T2 → T3 practical and written assessment.

Practise with Maya

East Africa calculation drills

Off-grid and hybrid sizing numbers you should be able to produce on paper.

Numbers you must be able to produce without a spreadsheet.

1. Daily energy demand and battery autonomy

   A household in Nakuru wants one day of autonomy on lithium storage.

   • · Daily load: 4.2 kWh
   • · Usable depth of discharge: 80%
   • · Round-trip efficiency: 92%
   • · Autonomy required: 1 day

   Find: What nominal battery capacity is required?

   Answer: ≈ 5.7 kWh nominal

   1. Energy to be delivered from storage = 4.2 kWh
   2. Account for round-trip losses: 4.2 / 0.92 = 4.57 kWh
   3. Account for usable depth of discharge: 4.57 / 0.80 = 5.71 kWh
   4. Select the next standard size up — never round down on autonomy.

2. Array sizing from insolation

   Size the array to recharge the same load in a single day.

   • · Daily load: 4.2 kWh
   • · Peak sun hours: 5.4 h
   • · System derate (soiling, temperature, wiring, conversion): 0.75

   Find: What array size in kWp is required?

   Answer: ≈ 1.04 kWp

   1. Required array output = 4.2 / (5.4 × 0.75)
   2. = 4.2 / 4.05 = 1.037 kWp
   3. Add margin for degradation and load growth; 1.2 kWp is the practical selection.

3. Voltage drop on a DC run

   A 12 V standalone system with a long run to the load.

   • · Current: 15 A
   • · One-way length: 8 m
   • · Conductor: 4 mm² copper (resistivity 0.0175 Ω·mm²/m)

   Find: What is the voltage drop, and is it acceptable at 12 V?

   Answer: ≈ 1.05 V, about 8.8% — not acceptable

   1. R = ρ × (2 × L) / A = 0.0175 × 16 / 4 = 0.07 Ω
   2. V_drop = I × R = 15 × 0.07 = 1.05 V
   3. 1.05 / 12 = 8.75%, far above the 3% target for low-voltage DC
   4. Increase to 10 mm² (drop ≈ 0.42 V, 3.5%) or shorten the run.

EPRA T2 timed mock

Two hours, method-weighted, marked the way a practical assessment is marked.

Sit it against the clock and read the score report honestly.

Proving isolation on a hybrid system

The isolation method every region assesses, in the order it is assessed. Switching something off is not isolation; isolation is a state you prove and keep proving.

What an assessor is actually watching while you work.

1. 1. Identify every energy source before touching anythingSafety-fail

   PV, battery, grid and any generator are separate sources. Missing one is how people get hurt on a system that 'was off'.

2. 2. Shut down in the manufacturer's stated orderSafety-fail

   Most hybrids specify inverter to standby, then AC, then PV, then battery. The order decides where energy is left trapped.

3. 3. Prove the instrument on a known live source before and after testingSafety-fail

   Testers fail silently. A dead reading from an unproven instrument is not evidence.

4. 4. Test every combination at the point of workSafety-fail

   Positive to earth, negative to earth and positive to negative on DC; line to neutral and line to earth on AC. One reading is not a proof.

5. 5. Lock and tag every isolator you operated, and keep the key

   A note on the panel does not stop a homeowner or a colleague restoring power while you are inside the equipment.

6. 6. Respect the stated DC-link discharge time, then verify it

   Capacitors hold charge after isolation. The waiting time is a minimum, not a guarantee.

7. 7. Re-prove dead after any interruption or absence from the work area

   Isolation you did not personally maintain is isolation you no longer own.

8. 8. Photograph the isolation state and the applied lock

   It is a safety record and diagnostic evidence at the same time, and it is what a defensible job file rests on.

The systematic diagnostic sequence

One order of work, used on every call, in every region. It is the order that keeps a hard fault from costing a whole day.

What an assessor is actually watching while you work.

1. 1. Write the symptom as reported, then separate it from any assumed cause

   'The inverter is broken' is a conclusion, not a symptom. Assessors mark whether you can tell them apart.

2. 2. Prove isolation before any intrusive workSafety-fail

   The single most common automatic fail across EPRA, NABCEP and QCTO practice.

3. 3. Observe a full start-up cycle before intervening

   Where the sequence stops names the gating condition that failed — and eliminates everything downstream of it.

4. 4. Take measurements that answer a specific question

   Random voltage checks score poorly everywhere. Each measurement should eliminate a branch.

5. 5. Compare output against a condition-corrected expectation, never nameplate

   Without irradiance and cell temperature you cannot yet say whether there is a fault at all.

6. 6. Eliminate the simple high-probability faults before the expensive part

   Connections, settings, soiling and shading account for most calls. Methodical elimination beats confident guessing.

7. 7. Verify the repair under the conditions that produced the fault

   A fault that only appears after rain or at midday is not fixed because it is quiet at 9am.

8. 8. Record what you actually measured, including inconvenient numbers

   Inflated commissioning sheets are recognised instantly and destroy credibility.

Faults that waste a whole day

The handful of misdiagnoses that account for most wasted visits — the correct call, and the reasoning that usually wins instead.

Correct answer, why it is correct, and the mistakes people make.

• Inverter showing a generic fault code

  Verify DC voltage, polarity, earthing and battery voltage against the inverter's operating window before condemning the unit.

  A fault code names the protection that operated, not the cause. Most are downstream of a DC-side or storage condition.

  • × Replacing the inverter first and inheriting the same code
  • × Reading the display instead of measuring at the terminals
  • × Ignoring the fault-history timestamps that would have shown the correlation

• Battery not charging, or charging very slowly

  Measure voltage at the battery terminals under charge and compare it with the voltage at the charge source. The difference tells the story.

  Charge settings, high-resistance connections, temperature sensing and one weak cell in a bank all present the same way.

  • × Replacing the bank without checking the configured charge profile
  • × Missing a single weak battery dragging a string down
  • × Treating a BMS temperature limit as a hardware fault

• Worked yesterday, dead today

  Visual inspection plus systematic voltage checks from source to load, looking for an intermittent connection, a loose connector, a failed surge device or a breaker that was never properly reset.

  Sudden total loss is almost always a connection or a protective device, not a degraded component.

  • × Starting with the most accessible component instead of the source-to-load path
  • × Resetting a breaker without asking what tripped it

• 'Low power' on an otherwise working system

  String-by-string current comparison under similar irradiance isolates it in under thirty minutes.

  Shading, soiling, mismatch or one underperforming string all read as generic underproduction at the meter.

  • × Jumping to inverter replacement
  • × Comparing against nameplate rather than a condition-corrected expectation

• Nuisance tripping of residual-current or earth-leakage protection

  Insulation-resistance testing plus visual inspection of terminations and cable routes, correlated with weather.

  Moisture ingress, damaged insulation and incorrect earthing arrangements trip protection that is working correctly.

  • × Replacing the protective device
  • × Testing at midday when the fault only appears after overnight rain

Commissioning, service and as-built records

The three documents every region expects. Printable, and written to be read by someone who was not there.

The paperwork, filled in the way it will be audited.

• Commissioning sheet

  Proves the system was tested to spec, under stated conditions, on a stated date.

  • · Site and system identification
  • · Array and string configuration as built
  • · Isolation and prove-dead record
  • · Measured values with irradiance and temperature
  • · Protection settings and grid parameters applied
  • · Defects, deviations and outstanding items
  • · Technician declaration and signature
  Open the printable document →

• Service report

  Records the reported symptom, what was measured, what was decided and what was verified.

  • · Reported symptom in the customer's words
  • · Conditions on arrival
  • · Checks performed and results
  • · Diagnosis and the evidence supporting it
  • · Work carried out and parts used
  • · Verification after repair
  • · Advice given and follow-up required
  Open the printable document →

• As-built notes

  Captures what is actually installed once the design met the roof, so the next technician is not re-surveying.

  • · Array layout with string routing and module counts
  • · Isolator and disconnect locations
  • · Cable types, routes and protection
  • · Earthing and bonding arrangement
  • · Equipment serials and firmware versions
  • · Deviations from the approved design, with reasons

Safety-fail scenario labs

Graded labs from the curriculum where an unsafe selection ends the attempt. Work these until the order is automatic.

Graded scenario labs — a safety-critical error still fails outright.

• Isolating a storage system in the right order

  Storage holds energy independently of sun and grid. The order is the assessment.

• Isolation fault — finding the string without becoming the path

  Weather-correlated isolation faults reward patience and punish shortcuts.

• Capstone stage 1 — arrival and safety

  A safety failure resets the stage rather than letting you carry on.

Shared diagnostic core — practice questions

Method and safety questions phrased the way each region's examiners phrase them.

Exam-style questions with the reasoning behind every answer.

Practise with Maya

Core method mock — 45 minutes

A timed check on the shared core before you sit a region mock. Method and safety only; no regional codes.

Sit it against the clock and read the score report honestly.