A SCADA database of thousands of points is built by copying rows. Copy errors — two points on one address, a description not edited, Hi above HiHi, an analog value in the coil table — load without complaint and appear during point-to-point testing or, worse, in operation. Checking the list first finds them in seconds.
Standards followed
Modbus Application Protocol V1.1b3 (data tables and addressing)IEC 60870-5-101/104 (information object addresses, common address)ISA-18.2 / EEMUA 191 alarm management principles (alarm limits and normal states)
What it can do
- Reads CSV or a copy from Excel and finds the columns from their headings: tag, description, device, protocol, address/IOA, common address, type, unit, EU low/high, LoLo/Lo/Hi/HiHi, normal state, deadband.
- Duplicate tag names; duplicate addresses per device, common address, protocol and data table; empty addresses.
- Missing descriptions, duplicated descriptions (copied rows), units missing or spelt differently, tag names with spaces or breaking your naming rule.
- Engineering range and alarm limits: low ≥ high, limits out of order, limits outside the range, large deadbands; trip and alarm signals without a normal state.
- Modbus: analog values in the coil or discrete-input table; IEC 104: IOA outside 1–16 777 215.
Example problem
The sample tag list: 20 points for a gateway (IEC 104) and a meter (Modbus).
- Q01_CB_POS appears twice; Q01_PROT_TRIP and Q02_PROT_TRIP share IOA 2002.
- Q02_I_A has Hi 1800 above HiHi 1600; T1 oil temperature HiHi 160 is above its 0–150 range.
- AUX_FREQ (analog) is at 00021, a coil address; kV is also written KV.
5 errors, 7 warnings and 3 notes, each with its line — fixed before the database is loaded and tested.
The validator checks the list for consistency. It cannot know whether an address matches the device's real register map — confirm that during point-to-point testing.