One schematic, thirteen studies: where Grid Tools stands
What runs today in the schematic workspace, which benchmarks it is checked against, what the free tier covers, what is still a stub — and what to tell us to build next.

Most engineering software asks you to choose: a calculator that does one thing properly, or a platform that does everything shallowly. We are building the third option — a set of calculators that each stand on their own, sharing one project structure, one equipment database and one schematic workspace, so that the cable you sized in one tool is the cable that carries the fault current in another.
This is a status note, not a launch announcement. It says what runs today, what it is checked against, what is still a stub — and it asks you to tell us what to build next.
The shape of the thing
Four standalone calculators are live: #001 Voltage Drop, #002 Short-Circuit, #003 Substation Grounding and #004 Cable Ampacity. Each is a complete tool with its own report. Around them there is now a shared layer:
- Projects — one project tree visible from every calculator, so a run can be saved and reopened from any tool;
- Equipment database — cables, switchgear, transformers, PV modules and inverters, shared by the calculators and the schematic editor;
- Grid Tools — the schematic workspace where the same engines run over a whole network instead of a single circuit.
The calculators answer "is this cable big enough". Grid Tools answers "does this network work" — and that is where most of the recent development has gone.
Grid Tools: what actually runs today

You draw a single-line diagram from a palette of 45 symbols — busbars and cables; LV and MV circuit breakers, disconnectors, fuses, load-break switches, auto-reclosers, earthing switches; CTs, VTs and surge arresters; two- and three-winding transformers, auto-transformers, split-winding and phase-shifting transformers; utility infeed, generators, synchronous condensers, UPS; motors and loads; NGR, NGN and zigzag earthing transformers; shunt, series and split reactors; capacitor banks and filters; SVC, STATCOM, HVDC; PV arrays, PV inverters and PV stations; BESS with its inverter; type-3 and type-4 wind turbines with their step-up transformer; a fault marker and a drawing frame.
Then you tick the studies and press RUN. Thirteen engines are wired in:
| Study | Basis | Runs today |
|---|---|---|
| Power flow | Newton-Raphson | yes |
| Voltage drop | IEC 60364-5-52 | yes |
| Short circuit | IEC 60909 | yes |
| Cable ampacity | IEC 60364-5-52 + IEEE 738 | yes |
| Protection coordination | IEC 60255-151 | yes |
| Arc flash | IEEE 1584 | yes |
| Power factor correction | — | yes |
| Motor starting | IEC 60034 | yes |
| Harmonics / THD | IEEE 519 | yes |
| Grounding | IEEE 80 | yes |
| Stability / short-circuit ratio | IEC 60909 SCR | yes |
| Reliability | IEEE 493 | yes |
| Protection stress-test | 100 faults + 100 switching events | yes |
They are not thirteen independent buttons — they feed each other. Ampacity consumes the per-load currents that voltage drop produced. Protection coordination takes I″k per bus from the short-circuit run. Arc flash takes the fault level and the clearing time that protection just derived. Motor starting uses the Thévenin impedance at the bus. Harmonics uses I_sc for the TDD limit lookup. Grounding takes the grid current. Stability takes S_sc. That chaining is the reason for a single workspace: in a folder of spreadsheets, those hand-offs are where the errors live.
On top of the studies there is the drawing side — sheet frame, project data, display parameters per element — and a report generator: the full study set is written into a report that is saved in the project tree and downloadable as .docx.
What it is checked against
An engine you cannot check is a rumour. Grid Tools ships a benchmark library — canonical networks with published answers:
| Benchmark | Size | Source |
|---|---|---|
| IEEE 9-bus (WSCC) | 9 buses, 9 branches, 3 generators | MATPOWER case9 — Anderson & Fouad (1977) |
| IEEE 14-bus | 14 buses, 20 branches | IEEE Common Data Format |
| IEEE 39-bus (New England) | 39 buses, 46 branches, 10 generators | MATPOWER case39 — Athay, Podmore, Virmani (1979) |
| IEEE 118-bus | 118 buses, 186 branches, 54 generators | IEEE Common Data Format |
| IEEE RTS-96 | 72 buses, 119 branches | IEEE RTS-96 Task Force (1999) |
| IEC TR 60909-4 §3 | LV 400 V test system | IEC TR 60909-4:2000, Figure 6, reference I″k in Table 4a |
| IEC TR 60909-4 §6 | 380/110/30/10 kV test network | IEC TR 60909-4:2000, Figure 16, reference I″k in Table 12 |
Load one, run the study, and a validation card compares the computed bus voltages, angles and branch flows — or the computed I″k3 bus by bus — against the published values, with the tolerance band and the source printed next to each row. The comparison can be exported as its own Word document.
This is the same discipline we applied to the standalone calculators: #004 now reproduces every buried row of IEC 60502-2 Table B.2 within 0,7 % in trefoil, and #002 reproduces the element impedances of the IEC 60909-4 worked example digit for digit. Both of those checks found real errors in our own code before they found agreement — which is the point of running them.
The free tier, precisely
Small networks run free, and "free" means the whole calculation, not a teaser:
- Free — power flow, voltage drop and short circuit, on a network of up to 10 busbars and 12 branches (a branch is anything carrying an impedance between two buses: cables, all transformer types, reactors, HVDC links).
- Any active licence removes the size cap. The calculator you own also unlocks its engine inside Grid Tools: #001 adds power factor correction, #002 adds protection coordination, arc flash, motor starting, harmonics, stability, reliability and the protection stress-test, #003 adds grounding, #004 adds ampacity.
- Past the cap the run is refused, not truncated — a partially solved network is an engineering trap, not a courtesy.
One limitation to state plainly: creating a network needs an account (free). A visitor can open the editor and look around, but the drawing surface only comes alive once there is a project to draw into. The same applies to the benchmark library.
The assistant
There is an assistant docked in the editor, running on a hosted Llama model. It answers engineering questions, and it can act on the schema — add and connect elements, set parameters, run a study, generate the report.

Treat it as a fast pair of hands, not as an authority: everything it produces lands in the same schema and the same engines that print their references, and that is what you check.
What is not there yet
- Substations and Renewables editors are announced on the landing page but currently redirect to Grid Tools. The substation-specific bay templates (single busbar, double busbar, one-and-a-half breaker), VT/CT placement and the IEEE 80 grid tied to the layout are the next iteration.
- Guests cannot draw. The free size envelope is implemented, but you still need to register to open a canvas.
- Validation coverage is uneven. Power flow and short circuit have published benchmarks behind them. Arc flash, harmonics, reliability and the protection stress-test run and produce numbers, but their published-reference validation is still ahead of us — treat them as engineering aids until we publish the comparison, the way we did for #002 and #004.
- The equipment database is being cleaned. Entries without a traceable manufacturer catalogue are being removed rather than left in with a plausible-looking specification.
We would rather tell you this than let you discover it in the middle of a design.
Tell us what to build
This is the part we actually need. The roadmap is not fixed, and feedback from working engineers moves it more than anything else does.
What helps most, in order:
- A case that gives a wrong or suspicious number — with the inputs, the value you expected, and the standard clause or published table you expected it from. This goes to the top of the queue every time. It is also how the last two engine fixes were found: running published tables through our own code and finding it disagreed.
- A missing element or parameter — "your transformer has no zero-sequence data for a YNyn0 with a delta tertiary", "there is no room for the cable's actual screen cross-section". Concrete gaps are easy to close.
- A study you need that is not in the list — insulation coordination, cable pulling tension, transformer inrush, LV selectivity against a specific manufacturer's curves. Tell us the standard you would want it referenced to.
- Reporting requirements — the format your client or authority actually accepts. If a report has to carry a specific title block, revision table or clause list, that is a small change for us and a large one for you.
- The equipment you use — manufacturer, series, and where the published data lives. That is how the database gets trustworthy.
Where to send it: the forum is the best place — it is public, so other engineers see the answer and can add to it, and each category maps to a part of the toolbox. For anything you would rather not post publicly, the contact address on every calculator page reaches us directly.
We read everything, and we say what we did with it. That is also why the guides on this site now carry a hand calculation against a published table in every worked example — if a number cannot be checked by the reader, it is not finished.
Where to start
- Grid Tools — draw a small network and run power flow, voltage drop and short circuit for free.
- #001 · #002 · #003 · #004 — the standalone calculators, each with a worked guide in Useful content.
- Forum — tell us what is missing.