CCA Control Mode

This page explains the background and internal logic of CCA mode. For setup details, see the Configuration Wizard and CCA Mode UI Configuration Entities.

What CCA Mode Is Trying to Do

CCA mode is designed for concrete core activation cooling and other very slow cooling systems with a lot of thermal mass.

These systems behave differently from a normal room thermostat:

  • they react slowly
  • they store cooling in the building structure
  • they need to act before the room temperature becomes uncomfortable

Because of that, CCA mode does not control directly from the current room temperature like the PI controller does. Instead, it uses the weather forecast to estimate how much cooling should be built up in advance.

Main Inputs

CCA mode uses three main inputs:

  • Cooling enable signal
    • A binary sensor, switch, or input boolean tells the controller whether cooling is currently allowed.
  • Daily weather forecast
    • The configured weather entity provides forecast highs and lows for the coming days.
  • Runtime tuning settings
    • Forecast thresholds, update interval, output limits, charge target scale, forecast response strength, and thermal storage persistence shape how strongly the controller reacts.

High-Level Model

The controller works with three closely related internal values:

  • Charge Target (Before Scaling)
    • The forecast-based demand level before the configured charge-target scale is applied.
    • Internally, this is the controller’s heat score.
  • Charge Target
    • The raw output target derived from the forecast after scaling.
  • Current Charge (Est.)
    • The controller’s estimate of how much cooling is already stored in the building core.

The basic idea is simple:

  • hot forecasts increase the target
  • stored cooling reduces the need for more output
  • the final output is limited so it cannot jump too much or exceed configured bounds

How Forecasts Become Cooling Demand

For each forecast day, CCA mode looks at:

  • the forecast high temperature
  • the forecast low temperature

From those two values it builds a normalized demand score:

  • hotter days increase the score
  • warmer nights also increase the score
  • the scores are averaged across the configured forecast horizon

This is why the two forecast threshold settings matter:

  • Hot Day Threshold decides when daytime highs start to count as a cooling signal.
  • Warm Night Threshold decides when nighttime lows start to count as a cooling signal.

How Charge Target Is Calculated

Once the forecast-based demand is known, the controller converts it into a Charge Target.

The most important setting here is:

  • Charge Target Scale
    • Raises or lowers the overall target level for the same forecast.

In practice:

  • higher scale means the controller aims for more cooling overall
  • lower scale means it aims for less

How Stored Cooling Is Taken Into Account

CCA mode does not assume every new forecast requires starting from zero. It keeps a running estimate of how much cooling is already stored.

That estimate is shown as:

  • Current Charge (Est.)

On each automatic control step, the controller:

  1. starts from the previous estimated charge
  2. adds the effect of the previous automatic output
  3. subtracts the effect of forecasted heat demand

This is where the two high-level tuning sliders come in:

  • Forecast Response Strength
    • Changes how strongly forecast heat pushes the controller toward more cooling.
  • Thermal Storage Persistence
    • Changes how long stored cooling is assumed to remain effective.

In practical terms:

  • a more responsive setting reacts more strongly to hot weather
  • a more persistent setting assumes cooling stays in the building longer

How Output Is Derived

The controller first calculates a raw need for cooling by comparing:

  • Charge Target
  • Current Charge (Est.)

If the target is above the estimated current charge, more cooling is needed. If the building is already charged enough, less cooling is needed.

After that, the controller applies these limits:

  • Output Step Limit
    • Prevents large jumps between automatic control steps.
  • Output Minimum and Output Maximum
    • Keep the automatic output within the configured range.

If Manual Override is enabled, the published output is replaced with the manual output value, but the controller still keeps tracking its internal automatic state.

Update Timing

CCA mode uses two different time layers:

  • Coordinator heartbeat
    • Runs every minute so Home Assistant can refresh state and countdown information.
  • Update Interval
    • Controls when a new automatic CCA control step is actually allowed.

This means the UI can update frequently even though the main CCA calculation only advances on a much slower schedule, such as every 6 hours.

The sensor Next Update In shows how long remains until the next automatic step is due.

What Happens Between Scheduled Steps

Between scheduled CCA steps, the controller normally keeps using its cached state.

Some setting changes still take effect immediately:

  • manual override changes
  • output min/max clamping of cached output
  • certain forecast-driven tuning changes such as charge-target scale

Those immediate refreshes update the published result without consuming the next scheduled automatic step.

Forecast-Unavailable Behavior

If the weather forecast cannot be used, CCA mode falls back to the configured behavior:

  • Hold
    • Reuse the last automatic output.
  • Shutdown
    • Force the output to 0 %.

The Status sensor shows whether the controller is active, inactive, holding forecast output, unavailable due to forecast problems, or running in manual override.

Reading the Main CCA Sensors

When diagnosing behavior, these sensors usually matter most:

  • Output
    • The value you typically use in automations.
  • Charge Target (Before Scaling)
    • The raw forecast-driven demand level.
  • Charge Target
    • The scaled target the controller is working toward.
  • Current Charge (Est.)
    • How much cooling the controller thinks is already stored.
  • Status
    • Whether the controller is active, inactive, holding output, unavailable, or in manual override.
  • Next Update In
    • When the next scheduled automatic step will happen.

Tuning Strategy

For practical tuning, start in this order:

  1. Charge Target Scale
    • Set the overall cooling level for the building.
  2. Forecast Response Strength
    • Decide how strongly hot forecasts should push the controller.
  3. Thermal Storage Persistence
    • Match how long cooling really lasts in the building.

Then fine-tune:

  • Hot Day Threshold and Warm Night Threshold to define when weather starts to matter
  • Output Minimum, Output Maximum, and Output Step Limit to shape how the automatic output behaves in practice
This site uses Just the Docs, a documentation theme for Jekyll.