The conference-room AHU supply-air temperature swings between 15 and 30 °C in a 30-minute period, while the room temperature also overshoots. The reported heating coil operates at 80/70 °C and often only 10–30% valve duty; after a valve change from 10% to 20%, the leaving-air temperature takes about 15–20 minutes to settle. That delay can make a room loop keep opening the valve before the first change has reached the room. Trend the complete signal chain before changing the valve, pump, coil, or controller settings.
Which signals show where the temperature swing begins?
First determine whether the room loop is reacting to a real room-temperature error, a misleading sensor value, or a delayed response from the coil. Trend the room temperature, room setpoint, heating command, valve position feedback (if available), coil entering- and leaving-air temperatures, and supply-air temperature together. Include hot-water supply and return temperatures and, where available, coil water flow and differential pressure. Use a common time base so that a valve command can be compared with the temperature response it produces.
| Signal | Source | Wrong-value symptom |
|---|---|---|
| Room temperature | Room sensor and BMS trend | A biased, poorly located, or unrepresentative reading can make the controller request heat while the occupied space is already warming, or stop heat before the room reaches setpoint. |
| Supply-air temperature | AHU discharge sensor | A reading that disagrees with a calibrated independent measurement can send the air-temperature loop after a false error; unstable actual air temperature points to the coil, water circuit, or control response. |
| Valve command and actual position | BMS output and actuator feedback or physical position check | A changing command with a stationary valve suggests an actuator, linkage, wiring, or output problem. A valve that moves but produces an inconsistent coil response directs attention to flow, valve sizing, or sensor placement. |
| Coil entering/leaving air and water temperatures | Coil-side sensors and field measurements | Unexpected or inconsistent temperature differences can indicate sensor error, poor mixing/stratification, inadequate or variable flow, or a coil response that is slower than the loop assumes. |
Check sensor placement and calibration before interpreting a trend. A single sensor can miss stratification: air temperature at one point may not represent the air leaving the full coil face or the air delivered to the room. Compare readings at representative points with calibrated instruments, and record where each measurement is taken.
How can coil delay make the room loop overshoot?
The sequence described is a delayed feedback problem. The BMS sees room temperature below setpoint and increases heat demand by opening the valve. The hot-water flow and coil metal then affect leaving-air temperature; that warmer air must travel through the duct and enter the room before the room sensor registers the change. If this full response takes 15–20 minutes at the coil and longer at the room, the controller can issue additional opening commands while the first command is still producing heat. When the accumulated heat reaches the room, the temperature can exceed setpoint.
Closing the valve does not immediately remove heat stored in the coil or duct, and the room itself also responds with delay. The same loop can therefore continue heating after the command has fallen. A cycle of delayed opening, overshoot, delayed closing, and undershoot can produce repeated swings. The reported 15–30 °C supply-air variation in 30 minutes is large enough to check both the control sequence and the physical measurements rather than treating the issue as room-loop tuning alone.
Low valve duty by itself does not prove that the coil is oversized. It could reflect low heating load, a high water temperature, excessive available flow, a valve with poor authority or oversized capacity, sensor problems, unstable water pressure, or a loop responding to a slow process. Coil capacity depends on air flow and entering-air temperature as well as water flow and water temperatures. Compare actual operating conditions with the coil selection data before concluding that replacement is needed.
What should be checked before changing the controls?
Establish the design and actual operating conditions. The coil supplier’s selection should identify airflow, entering-air temperature, desired leaving-air temperature, and water temperatures; compare these with site measurements and the intended room load. Also confirm whether the conference room is internal or has significant envelope load, and how ventilation air is supplied. A ventilation-only heating load can be a small fraction of the cooling load, which makes selecting a stable water coil at the required low output more difficult.
Inspect the control valve and hydronic circuit under representative operation. Confirm the actuator follows the BMS command, the valve is installed and configured for the intended flow direction, and the flow at full commanded opening matches the design requirement. Check actual water flow and pressure conditions rather than inferring them from valve position. Other equipment connected to the same water supply can alter pressure and flow as it operates.
Ask the AHU or coil supplier to check the coil capacity at measured conditions and, if needed, calculate the water temperatures and flow needed to deliver the desired air temperature. A pump or revised valve arrangement may be appropriate only after the circuit schematic and design flow are understood. A schematic should show the coil, valve, pump, bypasses, sensors, water supply/return, and control outputs. Without it, “constant volume, variable temperature” is not a complete description of the hydraulic arrangement.
How should the BMS and water circuit be tested?
- Trend the signals listed above through a stable operating period and at least one heating demand change. Mark the time of each command change and compare it with valve movement, coil leaving-air response, supply-air response, and room response.
- Verify each temperature sensor against an independent measurement at its installed location. Check for loose or poorly located sensors, and compare more than one point across the coil discharge if stratification is suspected.
- Check valve command versus actual stem position and measure or otherwise verify water flow at representative valve positions, including full opening if the design permits. Record water supply/return temperature and circuit pressure conditions at the same time.
- Compare measured airflow, entering/leaving air temperatures, water temperatures, and flow with the coil selection. Have the supplier review the selection if actual conditions differ from design or if the coil output cannot be controlled over the required range.
- After correcting any measurement, actuator, or hydraulic fault, make one controlled change at a time. Allow the measured process response to appear in the trend before making another change; do not tune from rapid command snapshots when the coil and room response are delayed.
When tuning is necessary, tune the inner supply-air control loop before relying on a slower room-temperature loop to drive coil output. Confirm what each loop controls: a room loop may set a supply-air target or directly command a valve, and those strategies have different interactions. Use the BMS sequence and trend to identify the actual arrangement. Select controller response from observed process lag and documented controller behavior; do not copy gains or timing values from an unrelated AHU.
When is a pump or constant-flow arrangement appropriate?
A dedicated pump can stabilize coil flow only when the piping and valve arrangement actually maintain the intended flow through the coil. Adding a pump without checking valve placement, bypasses, available pressure, and the control sequence can create unwanted differential pressure or fail to deliver the desired flow. Have the designer or supplier review the complete schematic and establish the required flow and operating point before installing or resizing a pump.
A constant-volume, variable-temperature concept separates the heat-transfer flow through the coil from the temperature control function, but it requires a suitable mixing or bypass arrangement and a control strategy designed for that circuit. A two-way valve that throttles coil flow does not become constant-volume simply because a pump is added. Validate the proposed arrangement by measuring coil flow across the valve’s operating range and checking that the leaving-air temperature responds predictably.
If the coil is oversized for the actual load, replacement is one option, not the first diagnosis. The supplier may find that lowering the coil’s hot-water supply temperature, using a correctly selected valve, or changing the circuit arrangement provides controllable output without replacing the coil. Any water-temperature change must be evaluated against other loads on the system and the required AHU air-side duty.
How do you verify the correction and avoid repeat swings?
Repeat the synchronized trend after each correction under comparable operating conditions. The valve should follow its command, the coil leaving-air temperature should respond in a repeatable direction, and supply-air temperature should settle without continued command escalation. Then observe the room response long enough to include its full delay. Confirm that a room-temperature correction does not trigger more valve movement before the previous action has had time to register.
Compare the measured supply-air and room trends with their control targets and the AHU’s design requirements. Verify sensor readings independently, check that the coil receives the intended water flow, and confirm the BMS sequence is acting on the correct sensor. If the output still swings, separate a measurement fault from a control-loop issue and from unstable hydraulics by checking each link in the trend: room demand, controller output, valve movement, water condition, coil air temperature, and room response.
Which recurring mistakes keep this AHU unstable?
- Calling the coil oversized from low duty alone: Low duty is a symptom, not a coil-capacity calculation. Compare the coil selection with actual airflow, entering-air temperature, water temperatures, and required load.
- Adding a pump before measuring flow: Determine whether the fault is low flow, varying flow, valve authority, or another circuit interaction. Review the piping arrangement before changing pump operation.
- Retuning while the process is still responding: The reported 15–20-minute coil response means repeated adjustments can stack up. Use trends and wait for the relevant response instead of chasing each delayed reading.
- Assuming one temperature represents the whole coil: Check sensor location and air mixing; a point sensor may conceal stratification or a nonuniform discharge temperature.
- Changing water temperature without checking shared loads: A coil supply-temperature change can affect other equipment on the same water system. Confirm system constraints with the responsible designer or supplier.
What should you ask about AHU coil temperature swings?
How do I tell whether an AHU heating coil is oversized?
Compare the coil selection with measured airflow, entering-air temperature, water supply and return temperatures, water flow, and required heating load. Low valve duty alone does not establish oversizing.
How do I check whether the room loop is reacting too quickly?
Trend room temperature, setpoint, valve command and position, coil leaving-air temperature, and supply-air temperature on one time base. Compare each command change with the delayed coil and room response before altering controller settings.
How do I know whether a pump will stabilize coil output?
Measure coil flow and review the complete piping schematic, including valve and bypass arrangement, before adding a pump. Verify that the proposed circuit maintains the intended flow and gives the valve a controllable effect on temperature.
When should I stop adjusting the BMS and escalate?
Stop repeated tuning if sensor readings, valve movement, or water flow cannot be verified, or if the trend still shows unstable output after measured faults are corrected. Escalate the coil selection and hydronic schematic to the AHU/coil supplier or system designer for a capacity and circuit review.