DSC-BRAIN USER GUIDE · USING THE DASHBOARD
Add a device
Adding a device takes seven steps. Each one is a pick from what the brain already knows, and the last step proves the device reacts. A device that fails the proof is saved as unproven, never as working.
● Running on our test kit: 9.10.0 · kit firmware 9.0.0.0 · last report 22 Oct 2026
Status: Live. On by default in brain 9.9.0.
On this page
Where to start
On the Devices screen, choose “Add a device”. The same flow is in Settings, in the Advanced row. The first-run guide uses the same Add entry for new kits.
A fresh card refuses every change until its owner sets a PIN, and adding a device is a change. The first card of the first-run guide reads “Set owner PIN (required)”. If you start before that, the first change you try opens a dialog headed “Set an owner PIN to finish setup”. Looking never needs the PIN. See Connections: how the brain reaches devices.
Two ways in
- A plug or sensor. Pick “What is it” from the device types, which are grouped by kind. The name is filled in from the catalogue. You can change it, up to 60 characters. This is the one field you may type. Choose “Add it”. The brain adds the device and reads the model back. Only then does it say “Added, and the model lists it”.
- A DSC device. Choose “A DSC device: power it now, I’ll listen 2 min”, then “Power it now, listen 2 min”. The hub listens for two minutes. The screen then says what the brain itself heard join: “Heard 1 new device”, or “Nothing heard in 120 s” with a reminder to check the device has power.
Under “A plug or sensor” the screen also tells you what you will be asked next, for example: “You will be asked to place: air intake, air out.” See Devices: classes, ports and placement.
The seven steps
| Step | Label on screen | What happens |
|---|---|---|
| 1 | Pick the device | Add one, or pick a device already in the model. |
| 2 | What kind of device | The brain’s open question about the device type. You pick from its options. |
| 3 | Which product | A pick if the product is in DSC-KitLib. If it is not, choose “Not in KitLib / next”. The device then stays generic, which is fine. |
| 4 | Where it sits | The placement questions for each port, picked from the places in your space tree. |
| 5 | What its readings mean | For each reading, pick what it means. See below. |
| 6 | How it is set up | The remaining open questions about the device, such as facts the brain needs. |
| 7 | Prove it works | A one-tap test. You do the physical action and the brain watches for the reaction. |
Answers go in as picks. After each one the brain reads the model back. The card closes only when the brain no longer lists the question.
What its readings mean
A reading such as “contact” or “power” means nothing until you say what it means. Pick the reading, then pick its meaning. The brain offers seven meanings: Water leak, Motion, Tank full, Door / contact open, Temperature, Relative humidity, and Lamp on (from power draw). Some ask a second question, such as what TRUE means (wet or dry), or the power level above which a lamp counts as on (5, 10, 20, 50 or 100 W).
A suggestion from the product library is only a hint. What you pick always wins, and it is saved as stated by you. Choose “Save meaning”. The screen says “Saved as what you said it means.”
Prove it works
Choose “Start the test”. The screen tells you the physical action, for example “Wet the sensor now” or “Switch the lamp on”. The result stays open until the brain sees the reaction or its window runs out. The windows are 60 or 120 seconds, depending on the meaning.
| Result | What it means |
|---|---|
| Working: the reaction was seen. | The device is saved as working. Only an observed reaction can do this. |
| Unproven: ... | The reaction did not arrive in time, or the test could not be read. The device is saved, but not marked as working. |
This is the same action then reaction rule the rest of DSC uses. A command or a tap proves nothing. The reaction does.
The first-run guide
A new kit gets a short guide with five steps in this order: Site, Spaces, Devices, Prove, Plants. Each step shows “Facts complete n of m”, counted from the brain. If the brain does not answer, the line says “not known” instead of a number. See Site profile for the first step.