Managing Virtual Devices

The Virtual devices list shows everything configured for the project: its name, the metric it produces, and the expression behind it. You can search it, toggle any one on or off from the list, and open an entry to edit or delete it.

Showing the expression in the list is more useful than it sounds. It is the fastest way to spot two Virtual Devices doing nearly the same thing, or one whose formula is not what its name implies.

Editing

Editing opens the same page used to create, with the current configuration filled in. You can change the name, the metric name and unit, the expression, the variables, the schedule and the timestamp settings.

Use Test expression again after any change, even one that looks cosmetic. Renaming a variable changes what the expression refers to, and an expression that no longer matches its variables will not produce what you expect.

Editing changes future output. It does not recalculate values already produced, so a corrected expression leaves the earlier readings as they were.

Check what depends on it first

This is the part worth slowing down for. A Virtual Device's output can be plotted on dashboard cards, used as the metric behind alert thresholds, and taken as an input by another Virtual Device. There is no view that lists those references for you.

Before changing an expression, its unit, or its variables, find out where the output is used:

1. Check other Virtual Devices for variables pointing at it.
2. Check dashboards in the project for cards plotting it.
3. Check alerts for thresholds set on it.
4. Consider whether anyone outside the project reads a report or export built on it.

The first one is the easiest to miss and the worst to get wrong, because a change propagates silently through everything built on top.

Changing the unit is the case that causes real trouble. An alert threshold set at 80 does not move when the value underneath it switches from centimetres to millimetres — the threshold stays where it is and quietly starts meaning something else.

Disabling instead of deleting

A Virtual Device can be disabled, either from its own page or with the toggle in the list. It stops calculating and stops producing new values, while its configuration and the values it already produced stay in place.

Disable when an instrument is offline for maintenance, when an expression is under review, or when you want to stop output without losing the setup. Re-enable when you are ready. It is the reversible option and usually the right first move.

Anything charting or alerting on a disabled Virtual Device stops receiving new values, including any Virtual Device built on top of it. From a dashboard that looks exactly like an instrument that stopped reporting, so tell whoever reads it or expect the question.

Deleting

Deleting removes a Virtual Device from the project.

Do the dependency check first. A dashboard card or alert built on something that no longer exists is a problem discovered at the worst moment, usually by someone who did not make the change.

If you are unsure whether something still matters, disable it and wait. A Virtual Device nobody misses after a month is safe to remove; one that generates a question in a week was load-bearing.

Keeping the list usable

Projects accumulate these, particularly where several people build them. Two habits keep the list readable: name things so the list is scannable without opening anything, and disable what is no longer used rather than leaving it running.

A calculation still producing values that nobody reads is not free — it is a number available to be charted or alerted on by someone who assumes it is maintained.