Humidistat and Dehumidification Control Setup

Purpose

This standing instruction accepts a humidity control on a measured change in relative humidity over a stated run time at a stated outdoor condition. A humidifier that energizes is not a humidifier that works, and a dehumidify mode that drops the blower speed is not a dehumidify mode that is safe.

Both directions have a protective gate set before the first run, and both gates get skipped. On the humidify side the gate is condensation: indoor air at a fixed humidity meets a window whose inner surface gets colder as the weather does, so a setpoint that was comfortable in November soaks the sills in January. On the dehumidify side the gate is the coil: the mechanism is running less air across the same coil so the air leaves colder and drier, and pushed too far that is a coil below freezing with an iced face and a compressor being fed liquid.

Safety actions that gate this procedure

• Open and lock the equipment disconnect and prove dead before any hand goes into the cabinet or any accessory terminal is landed, live-dead-live on a known live source before and after per NFPA 70E-2021, 120.5, with work practices at 29 CFR 1910.333(b)(2), written for a qualified person under 1910.332.
• A humidifier puts a water line beside energized equipment. Route the supply line and the drain so a leak or a weep runs away from the cabinet and the electrical connections, never over them, and confirm a shutoff you can reach without tools before the line is charged.
• Wear a personal CO monitor for any run in this procedure that fires a burner.

Scope

Covers setting up and accepting humidification and dehumidification control on residential and light-commercial systems: accessory terminal logic, outdoor reset, airflow reduction, and the measured runs that accept the setup.

Does not cover sizing or selecting a humidifier or dehumidifier, or duct connection design. Does not cover indoor air quality assessment, owned by its own SOP. Does not cover refrigerant charge or coil cleaning, which change the same numbers this procedure measures and must be correct before it starts.

Roles and handoffs

Procedure

1. Measure and write the starting condition before you configure anything. Acceptance: indoor humidity and temperature taken at the return with an instrument of known calibration, outdoor temperature and humidity, and a named list of the coldest interior surfaces - north windows, exterior-wall closets, skylights. Wrong looks like starting from the customer's account of dryness, which is a comfort report and not a measurement. Stop rule: no working instrument and the visit becomes a return with one; a humidity setup with no starting number cannot be accepted later. Hazard: none here, instrument work in occupied rooms.

2. Read the accessory terminal's logic and rating out of the equipment manual before you land a conductor. Acceptance: whether the terminal is an output or an input, whether it is 24 VAC or a dry contact, and for a dehumidify terminal whether the board expects the contact open or closed to call, all transcribed with the document identifier. Wrong looks like a normally-open control landed on a board expecting normally-closed, which gives a system that dehumidifies always or never and reads as a failed board. Stop rule: a terminal you cannot define in the manual is not used. Hazard: none here, desk work.

3. Set both protective gates before any accessory runs. Acceptance: the humidifier control's outdoor reset relationship entered from the control manufacturer's reset table and written out as a schedule of setpoint against outdoor temperature; and the dehumidify airflow reduction set to the value the equipment manual states, never deeper. Wrong looks like a fixed winter setpoint with no reset, or an airflow reduction chosen because more felt better. Stop rule: a control with no reset capability on a house with cold single-glazed windows gets a written manual schedule and a customer instruction, not a set-and-forget number. Hazard: these values command the equipment. Deeper airflow reduction drives the coil colder, so the manual's figure is a ceiling and step 5 measures what it actually produced.

4. Land the accessory and prove it energizes on a real call, measured at the accessory. Acceptance: the accessory's own terminals reading the commanded state under load when the system calls, and reading nothing when it does not. Wrong looks like a solenoid confirmed by hearing a click, which does not distinguish an energized valve from one passing no water. Stop rule: a high-impedance reading that collapses under load is not a call arriving; that goes to the low voltage control wiring SOP. Hazard: a live measurement in an open cabinet, so meter and leads rated for the circuit, one hand out, and the water line already routed clear of the electrical side.

5. Run and measure: relative humidity at the return, over a stated run time, at a stated outdoor condition. Acceptance: a start and end humidity reading at the same point with the elapsed time and the outdoor temperature both recorded; and on any dehumidify run, evaporator saturation temperature measured and held above 32 F. Wrong looks like a reading taken ten minutes after start-up, which measures the instrument settling and not the building. Stop rule: saturation at or below 32 F and the dehumidify airflow reduction is backed off to the manual's figure or shallower before the run continues; you do not run a coil below freezing to make a humidity number. Hazard: gauges are connected before start-up on any run that measures saturation, and no protective circuit is bypassed to keep a run going.

6. Walk the cold surfaces after the humidify side has held its setpoint through a full cold night. Acceptance: no liquid water or frost on any surface from the step 1 list, checked at the coldest hour available. Wrong looks like a hallway hygrometer reading a comfortable number while the north bedroom window runs water onto the sill. Stop rule: condensation anywhere on the list and the reset schedule drops a step and the run repeats; do not accept on the average reading. Hazard: none here, a walk through occupied rooms; if a sill is already wet, tell the occupant rather than wiping it.

7. Restore, prove what you disturbed, and hand the schedule over in writing. Acceptance: panels refitted, water connections checked dry under pressure, the blower door interlock proved by opening the door and confirming drop-out, a full cycle run with no lockout, and the reset schedule plus the window rule left with the customer on paper. Wrong looks like a verbal handover, after which the first cold snap produces a call about wet windows and nobody knows what the setpoint was. Stop rule: any accessory that does not drop out when the system stops is left disconnected and noted. Hazard: this puts power, water and motion back with the customer nearby - doors fastened, stand to the hinge side of the panel, and nobody in the blower compartment before the first call.

The record this produces

A setup record filed to the equipment:

• Starting condition: return humidity and temperature, outdoor temperature and humidity, instrument used, and the cold-surface list
• Accessory terminal logic and rating, with the document identifier
• The reset schedule as entered, setpoint against outdoor temperature, row by row
• Dehumidify airflow reduction as set, and the manual figure it came from
• The step 4 measurement at the accessory, high impedance and under load
• The run: start and end humidity at the return, elapsed time, outdoor temperature, and saturation on any dehumidify run
• Step 6 result, surface by surface, and step 7 verification

The cold-surface list is the field that makes the record worth keeping. Next winter's complaint will name a specific window, and this list tells the next tech whether that window was already the worst one in the house when the setup was accepted.

Worked pass: bypass humidifier with outdoor reset, on a two-stage gas furnace

Step 1: outdoor 20 F. Return air reads 68 F and 22 percent relative humidity on a hygrometer checked against a reference that morning. Cold-surface list written: two north bedroom windows, a skylight over the stairwell, and the exterior-wall closet in the front bedroom.

Step 2: the furnace manual defines the humidifier terminal as a 24 VAC output energized with the blower on a heat call, and defines the dehumidify terminal's expected contact state. Document number recorded. Only the humidify side is in use on this system.

Step 3: the humidifier control's own reset table is entered as a written schedule of setpoint against outdoor temperature. At 20 F outdoor the table's setpoint is 40 percent, and that is what goes in.

Step 4: cabinet opened under the electrical gate, humidifier landed, heat called. At the humidifier's own terminals the reading is 25.6 V under load with the blower running, and near zero with the system off. Water flow confirmed at the drain.

Step 5: run measured over 24 hours at an outdoor average of 20 F. Return humidity goes from 22 percent to 31 percent at the same measurement point, so 31 minus 22 leaves a 9 point rise. The control is still calling at the end of the period, which says the house has not reached the 40 percent setpoint rather than that the humidifier has failed.

Step 6 FAILS. At the coldest hour, both north bedroom windows carry liquid water along the lower third and a band of frost at the very bottom edge of the glass. The skylight is dry and the closet wall is dry. Stop rule taken: the setup is not accepted on the 31 percent return reading, and the reset schedule drops a step.

The mechanism is worth stating because it decides which knob moves. The glass surface is the coldest thing in the room and gets colder as the outdoor temperature falls, so the failure is not that 31 percent is high in the abstract, it is that the dew point of the room air is above that particular glass surface temperature at 20 F outdoor. Adding heat to the room does not fix it and raising the setpoint makes it worse; the setpoint has to fall with the weather, which is what the reset schedule exists to do.

Re-run: schedule dropped one row so the 20 F setpoint is 30 percent. After a second 24 hour period at an outdoor average of 19 F, return humidity reads 29 percent, the control cycles off rather than calling continuously, and both north windows are dry at the coldest hour with no frost at the edge.

Step 7: panels on, water connections dry under pressure, interlock proved by pulling the blower door with the system running. The reset schedule is printed and left at the equipment with a copy for the customer, along with the one rule that matters to them: water on a window means the setpoint is a row too high for the weather, not that the humidifier is broken.

Checking this pass: step 5's acceptance asks for start and end humidity at the same point plus elapsed time and outdoor temperature, and both runs print all four with the subtraction shown. Step 6's acceptance is surface by surface from the step 1 list, and the failing run reports which two surfaces failed and which two passed rather than a summary. No evaporator saturation figure appears because no dehumidify run was performed on this system, and that is recorded as not applicable rather than left blank.

References

• Equipment manufacturer's installation manual, the authority on accessory terminal logic and rating and on the airflow reduction figure
• Humidifier or dehumidifier control manufacturer's manual and its outdoor reset table, from which the schedule in step 3 is transcribed
• NFPA 70E-2021, 120.5 and 29 CFR 1910.333(b)(2) with 1910.332, for de-energizing and proving dead before landing the accessory in step 4
• See related: the indoor air quality assessment visit SOP; the sensor calibration and offset record SOP for the instrument used in step 1; the low voltage control wiring troubleshooting SOP