Three different things get called "electrical CAD"
The confusion starts with the category. Three families of software are routinely lumped together.
General 2D drafting. LibreCAD and FreeCAD's Draft workbench are drawing engines. They know lines, arcs, layers and dimensions. They do not know that a circle with a cross is a luminaire, or that a cable has an ampacity. You supply every convention yourself, usually through blocks.
Schematic and diagram editors. QElectroTech is the most cited open source option here. It ships symbol libraries, draws conductors between elements, and can generate a basic conductor list or a title block. It is genuinely useful for control-panel schematics and simple single-line diagrams.
Electronic design automation. KiCad is frequently recommended in forum threads asking for "electrical CAD", and it is the wrong tool for building wiring. It is excellent at PCB schematics and netlists, where the output is a board, not a distribution board.
Only the second family overlaps with the daily work of an electrical designer, and even then only partly.
The realistic shortlist
| Tool | Licence | Best at | Not built for |
|---|---|---|---|
| QElectroTech | GPL, free | Schematic diagrams, symbol libraries, conductor numbering | Load calculations, panel schedules, standard checks |
| LibreCAD | GPL, free | 2D drafting, plan redraws, DXF exchange | Anything electrical-specific — no symbols, no data model |
| FreeCAD (+ Draft, Arch) | LGPL, free | 3D and parametric work, BIM-adjacent tasks | Fast plan-and-schematic production; steep learning curve |
| KiCad | GPL, free | PCB schematics, netlists, footprint libraries | Building wiring, consumer units, cable sizing |
| Dia / Inkscape | GPL, free | Quick sketches, annotating a PDF plan | Engineering data, quantities, revision of objects |
The table is not a ranking. LibreCAD is a fine program; it is just a pencil, and a pencil has never produced a bill of materials.
Where open source is genuinely strong
Open source tools win in three situations, and they win clearly.
Drawing control schematics. QElectroTech handles terminal strips, contactor coils and cross-references well enough for panel builders who already know their conventions. The symbol editor is straightforward, and the file format is XML — scriptable if you have someone who scripts.
Working without a licence budget. For a student, a workshop, or a small firm in a country where a seat of commercial software costs several monthly salaries, a GPL tool is not a compromise on principle. It is the only option that exists.
Long-term access to your own files. XML and DXF are readable. You can archive a project and open it in ten years without a subscription. This is a legitimate engineering argument, not a sentimental one.
There is also the audit argument: if a calculation matters, you can read the source, or pay someone to. With commercial software you accept the vendor's arithmetic on trust.
Where it stops
The gaps are consistent across every open source option, and they are not cosmetic.
No project model. An electrical design is not a drawing; it is a set of connected objects — a room, a circuit, a breaker, a cable, a load — with relationships that must stay consistent. Open source drafting tools store geometry. Rename a circuit on the plan and nothing else updates, because nothing else knows the circuit exists.
No panel schedule. The schedule is a report over the circuit data: way number, description, rating, cable, load, spare capacity. If the data lives only as text on a drawing, the schedule is retyped by hand, and retyped schedules drift from drawings.
No load or cable calculation. Demand factors, diversity, voltage drop over the actual route, derating for grouping and ambient temperature. Some of this can be done in a spreadsheet — and in many small firms it is — but the spreadsheet does not know when a circuit changes.
No rule checking. An open source drawing environment will happily let you put a 32 A breaker on a 1.5 mm² conductor. Nothing in the file knows the pair is wrong. Compliance checking is the single largest gap, because it is the part of the job where mistakes become expensive.
No plan import. Converting a PDF survey or architect's drawing into usable geometry is a data problem, not a drawing problem. Open source tools expect you to trace.
Support and pace. Mailing lists and forums answer questions well and slowly. If a release breaks your file format the week a tender is due, "the community is helpful" is not a schedule.
The cost argument, honestly
"Free" describes the licence, not the project. The real cost of an open source CAD seat is the setup and the glue: block libraries built to your standard, a spreadsheet for calculations, a second spreadsheet for the schedule, and the hours spent keeping the three in agreement. That glue is invisible on an invoice and very visible in a deadline.
For a firm doing two small projects a year, the glue is cheap. For a firm doing two a week, it becomes a part-time job — which is the point at which a tool with a data model starts paying for itself.
The reverse also holds. If your work is 90% control-panel schematics with heavy custom symbology, a mature open source editor plus your own library may beat a general web tool, because the general tool will not model your conventions either.
A hybrid workflow that works
The practical answer for most small teams is not "open source or commercial". It is a split by document type.
- Draw schematics where the symbol library fits. QElectroTech or an existing commercial seat.
- Keep the plan, circuits and panel in one model. That is where consistency earns its money — see how a single-line diagram and a panel schedule are generated from the same circuit list rather than drawn twice.
- Run checks before export. Deterministic rules catch breaker-versus-cable mismatches, missing RCD protection and overfilled panels long before an inspector does; the 41-rule check describes the family of rules and, just as importantly, what they cannot see.
- Export DXF/PDF for whoever needs a drawing. The deliverable is a file, not a workflow. Nobody downstream cares which editor produced it.
A browser tool fits this split because it removes the installation step: a live demo opens in a tab, and the free tier is enough to test whether the model-based approach suits your projects before you commit to anything.
When open source is the right call
Choose open source when the deliverable is a drawing and the calculations are done elsewhere by people who own them; when the budget is genuinely zero; when you need scriptable file formats; or when you are building a custom symbol system that no vendor will model for you.
Do not choose it expecting a panel schedule, a cable calculation or a compliance check to appear. Those are separate engineering problems, and no GPL drafting program has solved them.
The middle path — open source for schematics, a model-based tool for the plan, circuits, schedule and checks — is unglamorous and it is what most small firms actually end up doing. If you want to see the second half of that path, pricing explains the free and paid tiers, and the documentation covers the import and export formats. For the wider landscape of commercial and free options, the electrical CAD software guide is the place to start.
Sources
- QElectroTech — official project site and documentation: qelectrotech.org
- LibreCAD — official project site: librecad.org
- FreeCAD — official documentation, Draft and Arch workbenches: wiki.freecad.org
- KiCad — official documentation: docs.kicad.org
- IEC 60364 series, Low-voltage electrical installations (scope reference): webstore.iec.ch
This article compares tooling, not professional advice. Verify requirements against the edition of the standard enforced in your jurisdiction.
No. It is a competent open source schematic editor with symbol libraries and conductor numbering. It does not produce panel schedules from circuit data, run load or voltage-drop calculations, or check a design against a standard. For schematic drafting alone, it is a serious tool.
You can draw an electrical plan in either. You will build your own symbol library, and the drawing will contain no engineering data — no circuit list, no ratings, no quantities. Suitable for small jobs and for redrawing an existing plan; painful as a production workflow.
No. KiCad is for printed circuit boards. Its symbols, footprints and netlists describe copper on a substrate, not cables in a wall. It is frequently recommended in forums because the question was about "electrical" in the electronics sense.
The licence is free. The total cost includes building symbol libraries, maintaining spreadsheets for calculations and schedules, and the engineering time spent reconciling documents that do not share a data model. For low project volume that cost is small; at higher volume it is not.
Not in any meaningful way. A drafting program has no rule engine and no model of a compliant installation. Any checking is done by a person reading the drawing, which is exactly the process that misses errors at 11 pm before a submission.
Design it in the browser
Turn the plan into circuits, a panel schedule, a single-line diagram and a calculated bill of materials. Start free or open the demo.
Open demo See pricing