Room Too Hot or Too Cold Investigation

Purpose

This standing instruction resolves a one-room complaint to a measured cause before anybody touches a damper.

The single-room call is the most commonly guessed job in residential HVAC. A tech opens the register wider, closes two others, and leaves. Half the time the room improves for a week and the complaint moves to whichever room lost the air. The reason is that the room's temperature is the output of four separate things, and three of them cannot be corrected by moving air around. Measure which one is failing and the fix stops being a negotiation with the ductwork.

Safety actions that gate this work

• Loose fill insulation of unknown type is not disturbed. Vermiculite in particular may contain asbestos, and attic loose fill in older housing can carry it. Do not walk in it, blow it, or sweep it aside to reach a duct. Where it must be moved, it is sampled first, and any respiratory protection used is selected under a written program meeting 29 CFR 1910.134.
• Attic work in cooling season is the real injury risk on this call. Walk on structural members only, carry two lights, work in short cycles, and tell someone in the house each time you go up and come down.
• Hood readings at ceiling registers put you on a ladder holding a hood in both hands. Set the feet level, keep your belt buckle inside the rails, and have the hood handed up rather than carried while climbing.

Scope

Covers investigation of a temperature complaint confined to one room or one small group of rooms on residential and small light-commercial forced-air systems, through to a ranked disposition with measured evidence behind each item.

Does not cover whole-house underperformance, which is a system-level problem and starts at the static pressure diagnostic and airflow verification SOPs. Does not cover the balancing work itself, owned by the register and grille balancing visit SOP, or the correction of a defective branch, owned by the flex duct inspection SOP. Zoning design and equipment replacement are out of scope.

The four causes, and the measurement that separates them

Two of these can be present at once, and usually are on a bonus room or an addition. That is why the procedure measures all four rather than stopping at the first one that looks guilty: correcting delivery on a room whose load is also wrong gets the customer part of the way and then reads as a failed repair.

Roles and handoffs

Procedure

1. Confirm the complaint with two simultaneous readings rather than the customer's account. Acceptance: room air temperature and the temperature at the thermostat location taken at the same minute with the same instrument at breathing height, plus the mode, run time and outdoor conditions. Wrong looks like comparing a reading taken now against one the customer took last night. Stop rule: if the two agree within 2 degrees F, the complaint is about something other than air temperature and you find out what before proceeding. Hazard: none physical; keep the instrument out of direct sun and off exterior walls.

2. Verify the system-level numbers before investigating the room. Acceptance: TESP and total measured airflow from their own SOPs, dated this visit, with total airflow within 10 percent of its target. Stop rule: a system short of its own airflow target is a whole-house problem and this room investigation is premature; go fix that first, because every room reading you take now will change when it is fixed. Hazard: as those SOPs state, power off at the disconnect for any drilling.

3. Establish what the room is entitled to. Acceptance: a required CFM for the room taken from a room-by-room load calculation and stated as a share of total system airflow. Where no load calculation exists, an area-weighted interim is allowed, labeled interim on the ticket. Wrong looks like balancing to area share on a room with an exposure the rest of the house does not have, which underfeeds every bonus room ever built. Stop rule: a room over a garage, under a roof deck, or on a west wall does not get an area-weighted target as a final answer. Hazard: none physical; a desk step.

4. Measure delivered airflow at every register in the room and sum them. Acceptance: a hood reading at each register with any correction the hood manufacturer publishes applied, a summed figure, and a comparison against the requirement with a gate of within 20 percent. Wrong looks like reading one register and multiplying. Stop rule: below the gate, delivery is an active cause and the branch gets traced before anything is balanced. Hazard: ladder work at ceiling registers per the safety block.

5. Measure the room's load drivers rather than assuming they match the design. Acceptance: surface temperature on each exterior boundary and on an interior partition taken within the same few minutes, attic insulation depth over the room, glass area and orientation. Wrong looks like a knee wall behind a bonus room with no insulation and no one checking, which shows up as a large surface temperature spread. Stop rule: a boundary reading far off the interior partition is a load finding and goes in the ranking even though it is not an HVAC repair. Hazard: reaching that knee wall usually means the attic, so walk on structure, and do not disturb unknown loose fill.

6. Rule the pressure and control causes in or out with numbers. Acceptance: closed-door room pressure in Pa against the 3 Pa criterion, every damper position on the branch recorded as found, and the thermostat-location temperature compared against the room at the same minute. Wrong looks like a room that only complains with the door shut being written up as a duct problem. Stop rule: a room over the pressure criterion routes to the return air deficiency assessment SOP before any airflow is added. Hazard: brief the occupants before you start closing doors.

7. Rank the active causes, correct, re-measure and restore. Acceptance: a ranked list with a number behind each item, and after any correction, delivered CFM re-measured against a requirement re-run for the corrected condition, plus the room-to-reference temperature spread re-read after a stated run time. Stop rule: if the load figure changed because envelope work was done, the requirement is re-run rather than compared against the old one. Hazard: this step puts the system back in service, so panels on, interlock proven by opening the door on a running blower, and every register left in the position recorded on the ticket.

The record this produces

• Room and reference temperatures at the same minute, with mode, run time and outdoor conditions
• System TESP and total airflow from this visit
• The room's required CFM, its source, and whether it is a load calculation or a labeled interim
• Delivered CFM at each register, summed, and the percentage of requirement
• Surface temperatures by boundary, insulation depth, glass area and orientation
• Closed-door pressure in Pa and every damper position as found
• Thermostat location and its temperature against the room
• The ranked causes, and post-correction figures against a re-run requirement

The advisor uses the ranking to keep envelope work visible instead of selling only the part the shop can invoice. The next tech reads the as-found damper positions rather than discovering that someone moved them. The customer's copy carries the causes that were not corrected, which is what prevents a partial fix being remembered as a failed one.

Worked pass: bonus room over a garage, running about 8 degrees F warm in cooling

Step 1: room 82 F, thermostat location in the downstairs hall 74 F, both at 2:40 pm, same instrument, 96 F outdoors, 25 minutes of run time. The 8 degree spread is real.

Step 2: TESP 0.58 in w.c. against 0.50 rated, total measured airflow 1,450 CFM on a 4-ton system against a target of 1,600, which is 91 percent and inside the 10 percent gate. The system is not the problem, so the room investigation proceeds.

Step 3: the house is 2,200 sq ft conditioned and the room is 320 sq ft, so area share is 320 divided by 2,200, about 14.5 percent, which would be 0.145 times 1,450, roughly 210 CFM. The room-by-room load calculation puts it at 22 percent of the sensible load because of the garage floor, the knee walls and west glass, so the requirement is 0.22 times 1,450, which is 319 CFM. The area-weighted figure is recorded as what it is, an interim that would have underfed this room by a third.

Step 4 FAILS. The two registers read 88 CFM and 74 CFM, summing to 162 CFM. Against the 319 CFM requirement that is 162 divided by 319, about 51 percent, and the gate needed 0.80 times 319, or 255 CFM. Stop rule taken: delivery is an active cause, nothing is balanced, and the branch is traced. It is a 6 in flex run about 38 ft long with two hard bends and a compression at a truss, which hands off to the flex duct inspection and correction SOP.

Step 5: the knee wall behind the room reads 104 F surface while an interior partition in the same room reads 78 F at the same minute, a spread of 104 minus 78, which is 26 degrees F. There is no insulation behind that knee wall. Load is a second active cause.

Step 6: closed-door pressure 1.2 Pa, inside the 3 Pa criterion, so pressure is ruled out. Branch damper found fully open and recorded. Thermostat is downstairs in the hall and cannot see this room at all, which is recorded as a control finding rather than corrected today.

Step 7 ranking, with numbers: delivery at 51 percent of requirement first, load at a 26 degree F boundary spread second, control third. After the branch is corrected to 8 in, shortened and properly supported, and the knee wall is insulated, the load calculation is re-run and the room's share falls to 16 percent, so the requirement becomes 0.16 times 1,450, or 232 CFM. Delivered re-measures at 265 CFM, which is 265 divided by 232, about 114 percent, inside the 20 percent gate in the high direction. Room-to-reference spread after 30 minutes of run time reads 2 degrees F.

Checking the pass against the sections above: all four causes named in the table were measured rather than assumed, and two came back active, which is exactly the combination the table warns about; the requirement in step 3 is a load-calculation figure and the area-weighted number is printed beside it and labeled interim; the failing step is step 4 and its stop rule sends the branch out rather than balancing around it; and step 7's re-measurement is compared against a requirement re-run for the corrected envelope, not against the original 319.

References

• ACCA Manual J for the room-by-room load that sets a room's required airflow, and Manual T for register selection and throw, in the editions your jurisdiction or program has adopted
• Flow hood manufacturer's documentation for the correction factors and the back-pressure limits of the instrument
• 29 CFR 1910.134 for the written respirator program governing any protection used where insulation is disturbed
• See related: the static pressure diagnostic SOP, the airflow measurement and CFM verification SOP, the return air deficiency assessment SOP, and the flex duct inspection and correction SOP