A trace-heated hot run that stays stubbornly lukewarm has three suspects hidden under the lagging: the controller, the temperature sensor, or the heating cable itself. Diagnose them in that order, most accessible first, before you cut open a metre of insulation chasing a fault you cannot yet see.
Electric trace heating — heat tape, or self-regulating heating cable — does the job a secondary return loop does elsewhere. On long runs where a recirculating loop is impractical, the cable lies along the pipe under the lagging and replaces the heat the water loses, holding the line at the maintenance temperature your risk assessment sets. When it works, you forget it is there. When it fails, the pipe drifts toward room temperature, into the band where Legionella multiplies most readily, and because the cable, the sensor and often the controller output are all buried, nothing tells you until an outlet reads cool [1][3].
Treat the whole run as one temperature-maintenance asset with a short list of failure points, and work them in order.
Confirm the run is genuinely cool, not the reading
Rule out a bad measurement first, the same discipline as any Troubleshooting temperature problems in plumbing systems job: a calibrated probe, the point your scheme names, the run time it specifies. With trace heating, add one move the recirculation case does not need. Take a surface temperature at two or three points along the lagged pipe, not just at the tap. A healthy trace-heated line holds a fairly even temperature end to end; a failed one shows a gradient falling away from the controller end — your first clue that the cable, not the outlet geometry, is the problem.
Start at the panel: power and the controller
The cheapest fault to find is no power, so check it first.
Is the circuit live? A tripped breaker, a blown fuse, or — very commonly with trace heating — a tripped RCD will leave the cable dead. A trace-heating RCD rarely trips for nothing: water reaching a damaged cable, a splice kit or an end seal is the classic cause, so a tripped RCD is both the symptom and half the diagnosis. Reset it once to confirm. If it trips again, stop and treat it as a cable fault, not a nuisance trip.
Then the controller or line thermostat. Is it actually calling for heat? Check the set point and the mode. A frequent and quiet reason for “trace heating not working” on the hot water side is someone having turned the controller down, or dropped it into a frost-protection-only mode over summer, with nobody reassessing the run against the risk assessment afterwards. The hardware is fine; the instruction changed.
Suspect the sensor before the cable
If power is present and the controller is calling for heat but the pipe stays cold, the sensor is the next suspect — and a far more likely culprit than the cable.
The sensor, usually an RTD strapped to the pipe, tells the controller what the line is doing. Displace it into warm air, let it drift, or site it beside another hot service, and it reports a pipe hotter than it really is. The controller believes it, cuts the cable, and the run goes cold while the panel looks perfectly content. The check is direct: compare the temperature the controller reports for that sensor against a calibrated surface reading taken at the sensor’s actual location. A meaningful gap between the two is your answer. Sensor faults are cheap to put right and never require condemning the cable, which is exactly why you rule them out before reaching for the insulation knife.
Read what the cable is drawing
Self-regulating heating cable changes its output with pipe temperature: the colder the pipe, the more current it pulls; the warmer, the less. That behaviour gives you a test without removing any lagging. Clamp a meter on the cable’s supply while the run is cool and the controller is calling for heat.
Zero current on a circuit that should be working hard means an open fault — a broken cable core, a failed feed, or a dead splice — and the cable is your problem. Current present but the pipe still will not climb points the other way: either the cable is overwhelmed by heat loss it was never sized to cover, or a length of it has partially failed. Read the current as evidence of which way to go, not a number to act on blind — what counts as healthy depends on the cable type, length and pipe temperature.
Do not forget the lagging
A perfectly healthy cable can still lose the battle if the insulation around it is missing, crushed or wet. Heat loss then outruns the cable’s output and the pipe runs cool regardless of what the controller is doing. Wet lagging is the worst case: it draws heat away and can damage the cable and trip the RCD into the bargain. The cable and its insulation are one system, not two, which is why Pipe insulation and its role in Legionella control matters as much to a trace-heated run as the cable rating does. Before you condemn any cable, confirm the lagging over it is continuous, dry and the right thickness.
From symptom to confirmed cause
Use this to get from the lukewarm outlet in front of you to a cause you have proved, rather than guessing under the lagging. Work the most accessible cause first, confirm it with the check, then act.
| Symptom | Most likely cause | The check that confirms it | The action |
|---|---|---|---|
| Whole trace-heated run cool; controller display dark | No power — tripped breaker, blown fuse or tripped RCD | Inspect the supply at the panel; note whether the RCD has tripped | Restore supply; if the RCD re-trips, treat as cable/water-ingress and call an electrician |
| Run cool; controller powered but not calling for heat | Set point lowered, or controller left in summer/frost mode | Read the set point and mode against the risk assessment’s maintenance figure | Restore the correct set point and mode; record the change |
| Controller calling for heat; pipe cold; panel “looks happy” | Sensor reading high — displaced, drifted, or near another hot service | Compare the controller’s sensor temperature with a calibrated surface reading at the sensor | Re-seat, relocate or replace the sensor; re-verify the run |
| Power and call present; cable draws no current | Open cable fault — broken core, failed feed or dead splice | Clamp the cable supply with the run cool and calling for heat | Isolate, have an electrician insulation-resistance test, then repair or replace |
| Cable draws current but the run still will not reach temperature | Wet or missing lagging, or an undersized/partly failed section | Inspect the insulation for damp and gaps; read current against the cable’s expected behaviour | Replace wet/missing lagging first; if it persists, have the cable assessed against the heat loss |
| Temperature falls steadily along the pipe from the controller end | A failed length of cable midway, or a long unheated tail | Take surface readings at intervals to locate where it drops away | Pinpoint and repair the dead section, or extend/redesign the heated length |
When to escalate, and to whom
Some of this is yours at the panel; some is not. An insulation-resistance (megger) test on the cable, and any repair or replacement of cable, splices or end seals, is electrical work for a competent person, not something to improvise on a live circuit. Bring them in the moment a circuit re-trips or a clamp meter says the cable is dead.
There is a control point here too, not just an electrical one. While a trace-heated run sits cool it is in the Legionella growth band, so the interim response matches any temperature-maintenance failure: feed it into the risk assessment, and step up flushing and monitoring on the affected outlets until heat is restored, at the frequency your scheme and a competent person set rather than a fixed calendar [2][4]. If a run trips or drifts repeatedly, the honest question is whether trace heating suits that pipe at all, or whether a secondary return would hold temperature more reliably — the trade-off behind Hot water recirculation systems: ensuring consistent temperatures. And because the failure stays invisible until an outlet reads cool, the strongest preventive move is to bring the controller’s temperature and alarm output into Using BMS for automated temperature control, so a cold run flags the same day instead of at the next monthly round.
On the figures, and the limits of this guide
This is general guidance to help you find a fault, not electrical design advice, legal advice, or a substitute for a competent electrician working on the cable. The maintenance temperature a trace-heated run should hold is the one your site-specific risk assessment sets. HSE’s hot and cold water guidance describes the familiar reference points — hot water reaching roughly 50°C at the outlet, stored around 60°C, cold kept below about 20°C where practicable — as guidance to apply through that assessment, not as fixed numbers to dial in blind [1]. Do not size, set or condemn a cable on figures copied from a web page; the correct values come from the cable manufacturer’s data and the people qualified to read it.
FAQ
Can I just turn the trace-heating thermostat up to fix a lukewarm run?
Only if the controller and sensor are healthy and the run is simply set too low. If the cable is dead, the sensor is reading high, or the lagging is wet, raising the set point changes nothing, because the cable is already being told to run or cannot keep up. Self-regulating cable also has a ceiling output it will not exceed however high you turn the dial. Confirm power, sensor and current first, and take the target temperature from the risk assessment.
Does trace heating remove the need for a secondary return loop?
It is an alternative way to do the same job — maintaining temperature on a run where a recirculating return is impractical — not an upgrade that removes a duty. Both aim at the same maintenance temperature, and neither removes the need to flush little-used outlets, monitor sentinel points and keep records. If a trace-heated run keeps failing, weighing it against a return loop is a fair design question rather than an admission of defeat.
Why does my trace heating keep tripping the RCD?
Repeated tripping almost always means water has reached somewhere it should not — a damaged cable, a poorly made splice, or a failed end seal — or the cable insulation has broken down. It is a fault to find and fix, not a trip to keep resetting. Isolate the circuit and have an electrician insulation-resistance test the cable to locate the breakdown before re-energising.
What to do at the panel today
Open the controller and read two things: the set point against your risk assessment’s maintenance figure, and whether the unit is calling for heat. Then take a calibrated surface reading at the sensor and compare it with what the controller thinks that sensor is seeing, and clamp the cable supply to check whether current is flowing. Those three readings, five minutes at the panel, tell you whether you are chasing the controller, the sensor or the cable — before anyone cuts a single length of lagging.
Sources
[1] HSE, “Hot and cold water systems”. https://www.hse.gov.uk/legionnaires/hot-and-cold.htm [2] HSE, “Legionnaires’ disease. The control of legionella bacteria in water systems - Approved Code of Practice and guidance (L8)”. https://www.hse.gov.uk/pubns/books/l8.htm [3] HSE, “Legionnaires’ disease: Technical guidance (HSG274)”. https://www.hse.gov.uk/pubns/books/hsg274.htm [4] HSE, “Testing and monitoring your water system for legionella”. https://www.hse.gov.uk/legionnaires/testing-monitoring-water-system.htm