Installation Faults That Keep Getting Blamed on Components

There's a pattern that comes up repeatedly on trade forums, in Gas Safe audits, and in honest conversations between engineers. A boiler keeps faulting. A circuit keeps tripping. A new component goes in. The fault comes back. Another engineer is called out. Another part goes in. Nobody goes back to the beginning and checks whether the job was installed correctly in the first place.

These are four real scenarios, anonymised, that illustrate how installation and commissioning errors get repeatedly misdiagnosed as component failures.

1. The Vaillant "fan fault" that was a wiring fault

A Vaillant boiler on a repeat call-back, fault code pointing at the fan. First visit: fan replaced. Fault returned within weeks. Second visit: PCB replaced. Fault returned again. By this point the customer is understandably furious and the costs are piling up.

The actual cause was that the boiler had no permanent live. Every time the room thermostat or programmer cut the switched live, the boiler lost all power and went through a cold boot cycle. The controls then interpreted the fan's power-up behaviour during that cold start as a fault condition.

Modern Vaillant boilers, and most condensing boilers, need a permanent live to maintain the PCB's memory, run frost protection, and keep the pump anti-seize function working. Without it, you're also adding hours to the boiler's run time as it goes through a full ignition sequence from cold every single time the stat calls for heat, rather than modulating from standby. That shows up as higher gas bills, which the customer also notices.

Check the wiring diagram. Permanent live to L. Switched live to the correct terminal. It's on every installation manual.

2. Reversed flow and return killing condensing efficiency

A condensing boiler installed and running, but the customer complains the house never gets warm and the bills are higher than expected. The boiler isn't faulting. Nothing obvious on a visual inspection. A parts swap on the pump is tried because the flow rate looks low.

The flow and return were plumbed in reverse.

On a condensing boiler, the return is the cooler pipe. The flue gas heat exchanger uses that cooler return water to drop the flue gas temperature below the dew point and recover latent heat. That's how you get the 90%+ efficiency figures and the white plume from the flue. When the flow and return are swapped, you're feeding hot water into the return, the heat exchanger never gets cool enough to condense, and you lose the efficiency advantage entirely. The boiler runs in non-condensing mode without ever flagging a fault.

At commissioning, check your flow and return temperatures with a clip-on thermometer. The return should be noticeably cooler than the flow, typically 20 to 30 degrees cooler on a properly set up system. If they're nearly equal, or the "return" is hotter than expected, something is wrong.

3. 50V phantom voltage on a neutral, sent round as a "cable fault"

An electrician tests a fused spur that's been added to a ring final. The neutral reads around 50V to earth. The spur is dead at the socket. The assumption is a broken or high-resistance neutral in the cable, so a rewire of the spur is quoted.

The cable was fine. The reading was capacitive coupling, often called phantom voltage, induced from an adjacent live conductor in the same conduit or buried beside it. A standard voltage tester or cheap multimeter with a high-impedance input will pick this up and display a plausible-looking voltage that isn't there under any load.

The quick check is to put even a small load across the suspect neutral and earth, a simple resistor, a neon indicator, or a cheap plug-in lamp. Real voltage will hold. Phantom voltage collapses immediately to near zero. A low-impedance voltage tester (a "wiggy" style tester) will also show zero on a phantom because it loads the circuit when measuring.

This one is common enough that BS 7671 and the GN3 guidance specifically reference it. If you're seeing unexpected voltages on neutrals, load the circuit before you condemn the cable.

4. Explosive ignition, blamed on the gas valve

A boiler with a heavy "woof" on ignition, sometimes loud enough to shake the case. The gas valve is replaced. The ignition module is replaced. The fault persists. The customer starts talking about a different boiler altogether.

Explosive ignition, where gas accumulates in the combustion chamber before the spark fires, usually comes down to one of three things: pre-purge time not set or too short (common on some older boilers when the PCB has been changed and defaults restored), electrode gap out of specification, or air-to-gas ratio wrong because the gas pressure or injector size wasn't checked at commissioning.

On a newly installed boiler, it often traces back to nobody checking the gas inlet pressure or burner pressure at commissioning. If the pressure at the appliance is high and the combustion settings haven't been set correctly for the site conditions, you get too much gas relative to air on light-up and a small explosion rather than a clean ignition.

The Benchmark commissioning checklist exists specifically to catch this. It requires you to record inlet pressure, burner pressure, and flue gas readings at commissioning. If it was filled in properly at installation, you have a baseline. If it wasn't filled in at all, that's often the first indicator that the commissioning wasn't done properly.

The pattern

In every one of these cases, a thorough commissioning record would have either prevented the fault or made diagnosis faster. Recorded inlet pressure at commissioning means you can compare against a later reading. Recorded flow and return temperatures tell you whether the boiler was condensing on day one. A completed Benchmark document tells you what the electrode gap and gas rate were set to.

When you inherit a fault on a job you didn't install, the first thing to do is go back to the commissioning documentation, if it exists. If it doesn't, that's your starting point for the investigation.

Good records protect the engineer who did the job right. They also quickly expose the job that wasn't done properly. Either way, they save everyone time.