Electrical Discharge Machining

§1Cutting with sparks

Electrical discharge machining removes metal not by a cutting edge but by a rapid succession of tiny electric sparks, each melting away a minute crater. The tool and work never touch.

An electrode — the shaped tool — is held a fraction of a millimetre above the workpiece, both submerged in an insulating (dielectric) fluid, and a voltage is applied. When the gap is small enough, the fluid breaks down and a spark jumps, melting and vaporising a speck of metal; thousands of such sparks a second erode the work into the electrode’s shape. Because nothing touches and nothing is cut mechanically, EDM does what no cutter can: it machines the hardest conductive materials — hardened tool steel, carbide — and forms intricate, deep, sharp-cornered cavities with no cutting force to break a delicate shape (§3). It works only on conductive materials, and it is slow, but for hard dies and fine detail it is unmatched.

Contents

§2How the spark erodes

Each spark is a controlled, contained melting event: a discharge across the gap heats a tiny spot to vaporisation, and the fluid flushes the debris away before the next.

The cycle repeats thousands of times a second. Voltage builds across the gap until the dielectric ionises and a spark discharges through it, concentrating enough energy on a microscopic spot to melt and boil away metal from both the work and, to a lesser degree, the electrode (§5). The discharge collapses, the dielectric flushes the molten debris out of the gap and re-insulates, and the next spark forms — often at the next-closest point, so erosion spreads evenly. The dielectric fluid does three jobs: it insulates so the spark only jumps when the gap is right, it flushes the eroded particles clear, and it cools. Control the energy of each spark and their number, and you control how fast and how finely the metal is removed (§6). It is melting by spark, not cutting by force.

Contents

§3No force, no hardness limit

The two defining advantages of EDM both follow from the tool never touching the work: there is no cutting force, and hardness is irrelevant.

Because erosion is by spark, not by a mechanical edge, no cutting force acts on the workpiece or the tool — so a slender, delicate or thin-walled feature that any cutter’s force would deflect or shatter can be machined intact, and a fragile electrode can form a deep narrow slot. And because the metal is melted, not sheared, the material’s hardness does not matter: fully hardened tool steel and cemented carbide cut no slower for being hard, which is exactly the situation that defeats conventional machining. This is why EDM owns the making of hardened dies and moulds — the part can be hardened first and then EDM’d to shape, avoiding the distortion of hardening a finished cavity. The one requirement is that the work conduct electricity; within that, hardness and delicacy cease to be obstacles.

Contents

§4The spark gap and overcut

Since the spark jumps a gap, the cavity EDM cuts is always slightly larger than the electrode — an overcut that must be allowed for by making the electrode undersize.

Example 1 — sizing the electrode

The spark erodes across a small gap all around the electrode, so the finished cavity is bigger than the tool by that gap on every side — typically an overcut of about 0.02 to 0.05 mm per side, depending on the spark energy. To cut a hole to size, then, the electrode is made undersize by the overcut: to leave a 10.00 mm cavity with a 0.03 mm gap, the electrode is ground to 10.00 − 2 × 0.03 = 9.94 mm across. A coarse, high-energy setting gives a bigger gap and overcut (and a rougher wall); a fine, low-energy finishing setting gives a smaller, more predictable gap. Knowing and allowing for the overcut is how EDM holds size — the machined cavity is never the electrode’s exact size, but the electrode’s size plus a gap you have designed in.

Contents

§5Electrode wear

The sparks erode the electrode as well as the work, so the tool slowly wears — a cost that shapes how electrodes are chosen and used.

Because each discharge removes a little metal from both electrodes, the tool gradually loses its sharp corners and length, and that wear transfers error to the cavity if uncorrected. Electrode materials are chosen to resist it: graphite and copper are the usual choices, eroding far more slowly than the steel they cut because the process can be tuned (polarity, energy) so most of the erosion falls on the workpiece rather than the tool. Even so, a roughing electrode is often followed by a fresh finishing one, and multiple electrodes may be used to complete a worn detail. The wear is the price of a contactless process — modest, controllable, but real — and managing it, by material choice and by finishing with a fresh electrode, is part of holding the cavity accurate.

Contents

§6Rate against finish, and wire EDM

As in every machining process, EDM trades speed against finish — set by the spark energy — and a second form, wire EDM, cuts profiles right through the work with a travelling wire.

The energy per spark sets the trade: high current gives big sparks that erode fast but leave a rough surface and a wide gap — roughing; low current gives small sparks that erode slowly but finish finely with a tight gap — finishing. So an EDM job is roughed hard then finished light, exactly as turning is, but with spark energy as the lever instead of feed and depth. Wire EDM is the important variant: instead of a shaped electrode plunging in, a thin travelling wire — like a bandsaw blade — is fed through the work while sparking, cutting a fine, precise profile straight through plate. It cuts hardened punch and die profiles, gears and intricate outlines to great accuracy with a very narrow kerf. Sinker EDM makes cavities with a shaped electrode; wire EDM cuts through-profiles with a wire — together covering the hard, precise, forceless machining conventional tools cannot.

Contents

§7Quick reference

The working core of the page on one card rack.

Principle

spark erosion, no contact

melts metal spark by spark

Advantages

no cutting force

hardness irrelevant (conductive)

Overcut

cavity > electrode by the gap

~0.02–0.05 mm/side → undersize tool

Electrode

graphite / copper · wears slowly

Two forms

sinker (cavities) · wire (profiles)

Contents

Handbook application: from concept to controlled practice

Purpose. This expanded section turns the original page into a practical handbook. It preserves the supplied material and adds a repeatable way to apply, check and review Electrical Discharge Machining. It does not replace a contract, legislation, a controlled standard, competent engineering judgement or specialist advice.

The operating aim is to carry the subject from function and assumptions through design evidence, verification and controlled release. Read the original explanation first, then use the workflow and checks below to convert knowledge into evidence.

Apply Electrical Discharge Machining by beginning with the duty, not the component or software command. Convert the key ideas—spark, machining, cutting, force, overcut—into measurable requirements and interfaces. Record operating and non-operating environments, duty cycle, expected life, loads, energy sources, human interaction and reasonably foreseeable abnormal conditions. When a value is not a project requirement or verified supplier datum, identify it as an assumption or illustrative value.

Create a calculation and evidence trail that another competent person can audit. Every input should carry a source, unit, revision and uncertainty or tolerance where relevant. Every model should state its boundary conditions and limitations. Keep nominal capacity separate from design capacity, and keep verification margin separate from an arbitrary safety factor. If a code or standard governs the work, confirm the applicable edition and contractual status rather than copying a number from a secondary summary.

Design for manufacture, assembly, inspection, operation and maintenance at the same time. A technically valid geometry can still fail because it cannot be fixtured, measured, cleaned, guarded, reached or replaced. Review process capability, datum or reference strategy, tolerance accumulation, access, error-proofing and changeover. Where people interact with plant, apply the hierarchy of controls and consult those who will operate, clean, maintain and recover the equipment.

Plan verification before release. Define the characteristic, method, equipment, sample or test condition, acceptance criterion, record and responsible person. Validation then asks a different question: whether the resulting system is effective and suitable in the intended use context. A passed drawing check or analysis does not by itself validate usability, maintainability or production performance.

Step-by-step operating method

Define the duty. Capture the required function, interfaces, operating environment, life, loads and unacceptable outcomes.
Establish the model. Identify governing principles, units, material or process data, assumptions and uncertainty.
Develop alternatives. Compare feasible concepts against performance, manufacturability, safety, maintainability and cost.
Verify the design. Use analysis, test, inspection or demonstration with acceptance criteria defined before execution.
Release and learn. Baseline the design, control changes, retain evidence and feed operating results into the next revision.

Illustrative design review record

Illustrative values only. Build a one-page record with the required function, input sources, assumptions, governing load or process condition, failure consequences, selected concept, verification method and acceptance criterion. Mark every numerical input as project requirement, verified supplier data, measured value, calculation output or assumption. Review the weakest evidence first. If an assumption can change safety, compliance, interchangeability or capacity, it must be resolved before release rather than buried in a calculation note.

Evidence class	Question	Release expectation
Requirement	What must the design do and under which conditions?	Approved and traceable
Input	Where did the load, property, tolerance or process limit come from?	Source, unit and revision recorded
Analysis	Which model and assumptions connect input to result?	Checkable calculation or simulation
Verification	How will conformity be demonstrated?	Method and acceptance criterion agreed
Validation	Will the solution work for intended users and conditions?	Representative use evidence

Common failure modes and recovery actions

1. Watch for

Starting detailed design before interfaces and operating limits are agreed.

Recovery: Return to the governing definition or requirement and restate the decision in one sentence.

2. Watch for

Using catalogue or typical values as though they were certified project inputs.

Recovery: Separate evidence from assumption, assign an owner and set a date for validation.

3. Watch for

Checking nominal performance while ignoring tolerances, degradation and foreseeable misuse.

Recovery: Run a small counterexample, boundary test, pilot or independent check before proceeding.

4. Watch for

Confusing verification of requirements with validation of user need.

Recovery: Record the consequence, decision and rationale, then update the controlled baseline.

5. Watch for

Releasing drawings or procedures without configuration, inspection and change controls.

Recovery: Escalate when the issue affects safety, compliance, acceptance, material value or an agreed tolerance.

Review checklist

What function and failure consequence govern this decision?
Which inputs are measured, specified, assumed or illustrative?
How will conformity be demonstrated and recorded?
What change would invalidate the current evidence?
Are mandatory requirements distinguished from recommendations and illustrative values?
Are sources, assumptions, units, dates and versions recorded closely enough to reproduce the decision?
Have safety, legal, ethical, stakeholder and operational consequences been considered at the appropriate level?
Is there a named owner and a trigger for review, escalation, change or retirement?

Questions for deeper application

What is the most important distinction a practitioner must preserve when applying Electrical Discharge Machining?