Energía de ignición adaptada al material para el corte de acero y aluminio

Energía de ignición adaptada al material para el corte de acero y aluminio

Compendio

A plasma cutter is sold as a universal tool and most of them fire one ignition pulse for every material. The gap breakdown voltage moves only 2.5% between air on steel and nitrogen on aluminium, while the energy the pilot arc needs moves 30 to 55%. This article works the Paschen relation, the ionization lower bound, the material factors and the ablation penalty of a fixed recipe, then gives the recipe table that separates the two.

1. Introducción

A plasma cutter is sold as a universal tool, and the physics of ignition is not universal. Mild steel, aluminium, stainless and coated plate differ in gas requirement, surface condition, melting temperature and breakdown statistics at the torch gap. Most machines fire the same pulse for all of them and cover the mismatch with energy.

The cost lands twice. Energy aimed at the hardest material scorches the consumables on every easy start, and energy aimed at the easiest one misfires on hard plate. Shops cutting mixed batches report 20 to 40% higher consumable costs than single-material shops running the same machine model, and the fixed ignition recipe accounts for almost all of it.

Automated cells carry the problem further. Material changes arrive unattended and often, so an ignition that cannot follow the material code costs cell utilisation instead of a few consumables.

This article separates the two things a recipe has to set: the voltage that clears the gas breakdown statistics, and the energy that establishes and transfers the pilot arc.

2. Mecanismo central

2.1 Breakdown voltage from the gas, not the workpiece

The workpiece enters the ignition physics through the gas chosen for the cut, and the gas sets the breakdown field.

V_brk = (B × p × d) / (ln(A × p × d) − ln(ln(1 + 1/γ)))

where V_brk is the gap breakdown voltage, A and B are the Townsend coefficients of the gas, p is the effective pressure, d is the gap and γ is the secondary emission coefficient of the cathode.

For a 1.8 mm gap at atmospheric pressure, so p × d = 137 Torr·cm, air with A = 15 and B = 365 returns 8.20 kV. Nitrogen with A = 12 and B = 342 returns 8.00 kV. The two cutting gases differ by 2.5% in breakdown voltage, which is far less than the difference in what they cost to ionize.

2.2 The ionization lower bound

The minimum energy for a pilot arc follows from the charge needed to ionize a critical volume of gas.

E_ion = ε_ion × n_e × V_critical

where E_ion is the energy spent creating ion pairs, ε_ion is the effective cost per ion pair, n_e is the electron density in the channel and V_critical is the volume that must be ionized. Nitrogen costs 36 eV per ion pair and air costs 32 eV, so the two differ by 13%.

For a 1 mm³ channel at 1 × 10¹⁹ m⁻³ the lower bound is 58 nJ in nitrogen and 51 nJ in air. Both are negligible against the 0.235 J a real stage delivers, so almost all of the ignition energy goes into gas heating, electrode heating and radiation. That is why the deposition efficiency sits near 1.4 × 10⁻⁸ and not near unity:

E_ign = E_ion / η_deposition

where E_ign is the energy the stage must deliver and η_deposition is the fraction of it that creates ion pairs. Air on steel runs at 2.2 × 10⁻⁷ and nitrogen on aluminium at 1.6 × 10⁻⁷, which turns the 13% ionization difference into the 55% energy difference the recipe table has to cover.

2.3 Material factors

The recipe scales a nominal energy by one factor for the gas and one for the surface.

E_ign(material) = k_gas × k_surface × E_nominal

where E_ign(material) is the energy for that recipe, k_gas accounts for the gas mixture and k_surface for the state of the workpiece. Against mild steel in air at 1.00, stainless in nitrogen measures 1.30, aluminium in nitrogen 1.55, and the same aluminium with a heavy oxide layer 1.87.

2.4 Voltage margin and stored energy

The voltage term carries the margin over the breakdown statistics, and the energy term carries the material factor.

V_ign = V_brk × (1 + margin)

where V_ign is the voltage the gap needs to see, V_brk is the breakdown voltage of the gas and margin is the statistical allowance, set at 20 to 25% from measured breakdown data and not from experience. Applied to 8.20 kV and 8.00 kV that gives 10.0 kV and 9.8 kV, both of which the same 10.71 kV gap voltage covers.

E = ½ × C × V²

donde E es la energía de pulso almacenada, C es la capacitancia del tanque y V es la tensión de carga. Con C en nanofaradios y V en kilovoltios, el resultado está en milijulios, por lo que 4.7 nF a 10 kV devuelven 235 mJ y no 235 J.

Because the two materials need nearly the same gap voltage, the material factor belongs in the capacitance and not in the charge voltage. Stepping the tank from 4700 pF to 6100 pF and 7300 pF at a fixed 10 kV returns 0.235 J, 0.305 J and 0.365 J while the gap voltage stays at 10.71 kV in all three cases.

2.5 What a fixed recipe costs on the easy material

Running one recipe means running the hardest one everywhere. The arc-root heat flux rises with the delivered energy at a constant root area, and the ablation rate follows the surface temperature through the exponential from the vaporization law.

A tank raised to 1.87 times the steel energy lifts the arc-root surface about 50 K, and at 3400 K that temperature step multiplies the ablation rate by 1.41. On a consumable set that costs 30 to 60 USD the arithmetic lands inside the 20 to 40% premium that mixed-material shops report.

2.6 Droop and pulse shape

The recipe also has to hold its shape across the pre-charge range.

ΔV = I × Δt / C

where ΔV is the tank droop during the pulse, I is the discharge current, Δt is the interval and C is the capacitance. Amperes with microseconds and nanofarads return kilovolts, which is the unit trap in this stage. At 300 A held for 0.264 µs a 4.7 nF tank would fall 16.9 kV, more than it holds, so the discharge runs on resonance and the capacitance sets the pulse length, not the droop.

Fig. 1 — One 10 kV pre-charge serves three materials; the tank capacitance steps 4700 to 7300 pF to set the ignition energy.

3. Puntos débiles de la ingeniería

Pain Point 1: Fixed hardware with a fixed energy

A machine with one tank capacitor and one pre-charge level cannot deliver a material-dependent profile, and adapting the firmware alone changes nothing. The hardware needs a pre-charge span of at least 1.5 to 1 and a way to step the tank capacitance, or the recipe table has nothing to write to.

Pain Point 2: Material misidentification

Operators mislabel plate and nesting software sends the wrong code. The ignition then applies a recipe for a different material, and the burn or the misfire is blamed on the machine. After the pilot arc strikes, the arc voltage and the sustain behaviour carry the signature of the real material, so a comparison against the recipe window catches the mismatch before the cut starts.

Pain Point 3: Capacitor drift making the table stale

If the tank capacitance moves with temperature or pulse count, the delivered energy for a given setpoint moves with it. A ±30 ppm/°C coefficient holds capacitance inside 0.15% across a 50 °C swing, which leaves most of the ±2% repeatability allowance for everything else in the loop. A general-purpose Class 2 dielectric does not hold that band, and the recipe table stops describing the machine.

Pain Point 4: Over-energizing the easy materials

A single recipe has to be set for the hardest material in the shop. The 1.87 factor that oxidised aluminium needs applies to every mild steel start as well, which multiplies the ablation rate by 1.41 and spends consumable life on work that needed none of it.

4. Pasos del diseño

Strategy 1: Give the hardware the range the recipes need

Provide a pre-charge setpoint that spans at least 1.5 to 1, from 9.0 kV to 13.9 kV, and a switched tank bank that steps the capacitance in 1400 pF increments. Use an N4700 Class I part in every step so the delivered energy is a known function of the setpoint.

Strategy 2: Set voltage from the gas and energy from the material

Take the charge voltage from the breakdown voltage of the cutting gas plus a 20 to 25% margin, and take the energy from the material factor through the capacitance. On one 10 kV pre-charge that gives 4700 pF, 6100 pF and 7300 pF for steel, stainless and aluminium, with the gap voltage held at 10.71 kV in all three cases.

Strategy 3: Check the declared material against the arc signature

Compare the measured arc voltage and sustain behaviour with the expected window for the declared recipe. A mismatch beyond tolerance should abort the cut and raise a flag instead of proceeding with a wrong-energy burn. The same measurement catches a worn electrode that has moved the arc voltage for reasons unrelated to the material.

Strategy 4: Cap the energy at the minimum that guarantees breakdown

Derive the cap from measured breakdown statistics per gas and per material instead of from experience, and store the 20 to 25% margin in the recipe table next to the value it belongs to. A cap written as a single global number cannot follow the material and ends up set for the worst case.

Strategy 5: Track consumable life per material

Log strikes and consumable sets against material codes. Re-tune each recipe's energy and sustain duration every quarter from that log, using the condition-monitoring approach that monitoring literature applies to capacitors, so the recipe follows the wear instead of a calendar.

Strategy 6: Make the material change cost no cycle time

Let the operator or the nesting software select the recipe in one action, and reconfigure the pre-charge setpoint, the tank step and the duration timer within 50 ms. The gate drive and the diodo de recuperación rápida in the output stage must be specified for the fastest recipe in the table, not for the average one.

Parámetro del componente	Valor típico	Condición de prueba
Tensión nominal (HVCT8G serie)	10 kV DC	alto potencial 1.5× durante 60 s
Capacitance steps	4700 / 6100 / 7300 pF ±5%	1 MHz, 1 V rms
Pre-charge span	9.0 kV to 13.9 kV, 1.5 to 1	for recipe headroom
Coeficiente de temperatura	±30 ppm/°C	−40 °C a +125 °C
Capacitance drift over 50 °C	dentro del 0.15%	against the ±2% repeatability
Descargo parcial	por debajo de 5 pC	a tensión nominal
Capacidad de dv/dt de pulso	por encima de 50 kV/µs	a tensión nominal

5. HVC N4700 (T3M) capacitors for a recipe-driven ignition

A recipe table describes the machine only while the tank keeps its value. HVC builds the N4700 (T3M) Class I family for that duty, with the resistencia de película gruesa that completes the divider in the setpoint loop, from a 6,000 m² owned factory in Dongguan across a 1 kV to 100 kV range.

Seis parámetros que importan aquí

• Energy that follows the setpoint. Tight capacitance tolerance and Class I stability make delivered energy a known function of the recipe value, so the table stays valid between the bench and the shop floor.
• Temperature independence. A ±30 ppm/°C coefficient keeps the recipe accurate from a cold morning start to a hot afternoon shift, and from one season to the next.
• Coverage of the whole recipe span. Ratings from 1 kV to 100 kV carry the pre-charge range for thin-sheet steel starts and heavy aluminium piercing from one qualified family.
• Loss that stays out of the delivery. A dissipation factor below 0.1% leaves the stored energy available to the arc instead of the capacitor body, and keeps the tank step from changing the pulse shape.
• Consistency over years. Partial discharge below 5 pC and the Class I formulation keep the recipe commissioned in year one accurate in year five, with no ageing drift to absorb into the margin.
• Proven where energy reproducibility is a production requirement. The 40 kV 441-series doorknob capacitors replaced first-tier brand parts in high-voltage field-service programs, and a referencia cruzada contra un condensador de película shows the drift a recipe table would otherwise have to absorb.

Entrega y soporte

Standard parts ship in 4 to 8 weeks and matched recipe banks in 6 to 10 weeks. HVC applications engineering supports capacitance stepping, pre-charge range selection, energy-versus-setpoint measurement and material recipe validation. Send the material list, the gas mixtures and the consumable data, and HVC returns a recipe bank selection sheet with tolerance data and PD reports within five working days.

6. Preguntas frecuentes

Q1: Do different materials need different ignition voltages?

Barely. The workpiece affects the ignition stage through the gas chosen for the cut, and air against nitrogen moves the breakdown voltage from 8.20 kV to 8.00 kV at a 1.8 mm gap, which is 2.5%. What changes by 30 to 55% is the energy the pilot arc needs, and that belongs in the capacitance.

Q2: Why not just raise the charge voltage for hard materials?

Because it changes the wrong thing. Raising the charge from 10 kV to 13.7 kV multiplies the energy 1.87 times, but it also lifts the gap voltage 37%, so the arc root sees a hotter attachment on every strike. Stepping the capacitance to 8800 pF delivers the same energy at the same 10.71 kV gap voltage.

Q3: How much does the ionization energy itself account for?

Almost nothing. Nitrogen costs 36 eV per ion pair against 32 eV for air, so the ionization lower bound for a 1 mm³ channel is 58 nJ against 51 nJ. Against a 235 mJ ignition pulse that is a deposition efficiency near 2 × 10⁻⁷, so nearly all the energy goes into gas and electrode heating.

Q4: How much does a fixed recipe cost on easy material?

The aluminium factor of 1.87 on the steel recipe lifts the arc-root surface about 50 K, and at 3400 K that step multiplies the ablation rate by 1.41. On a 30 to 60 USD consumable set, across a year of mixed batches, that lands inside the 20 to 40% premium that mixed-material shops report.

Q5: How do I catch a wrong material code?

Compare the measured arc voltage and the sustain behaviour against the window stored with the recipe. A mismatch beyond tolerance means either the declared material or the electrode condition is wrong, and both are worth stopping for before the cut proceeds and the burn is attributed to the machine.

Q6: How stable does the tank capacitance have to be?

Good enough that the recipe table still describes the machine. A ±30 ppm/°C coefficient holds capacitance inside 0.15% across a 50 °C swing, which leaves most of a ±2% repeatability allowance free. A general-purpose Class 2 dielectric moves far more than that over the same swing.

7. referencias

1. E. Kuffel, WS Zaengl, J. Kuffel, Fundamentos de ingeniería de alta tensión, 2ª ed., Newnes, 2000.
2. Mohan, N., Undeland, TM y Robbins, WP, Electrónica de potencia: convertidores, aplicaciones y diseño., 3.ª ed. Hoboken, NJ: John Wiley & Sons, 2003.
3. Aviso 2 de MIL-HDBK-217F, Predicción de la fiabilidad de los equipos electrónicos, 1995.
4. IEC 60384-1:2021, Condensadores fijos para uso en equipos electrónicos - Parte 1: Especificación genérica, 6ª ed. Ginebra: IEC, 2021.
5. IEC 60384-9:2024, Condensadores fijos para uso en equipos electrónicos – Parte 9: Especificación seccional – Condensadores fijos de dieléctrico cerámico, Clase 2, Ed. 5.0. Ginebra: Comisión Electrotécnica Internacional, 2024.
6. B. Wang, J. Meng y P. Zhao, Monitorización del estado de envejecimiento de los condensadores electrolíticos de aluminio en variadores de velocidad., IEEE Transactions on Power Electronics, vol. 37, no. 4, pp. 4564-4574, 2022. https://doi.org/10.1109/tpel.2021.3121813.
7. V. Belyaev, Modelo de descarga superficial de un aislante en vacío, en Actas del ISDEIV. 18º Simposio Internacional sobre Descargas y Aislamiento Eléctrico en Vacío, págs. 151-154, 1998. https://doi.org/10.1109/deiv.1998.740596.
8. Impulsando el futuro: Cómo los condensadores cerámicos multicapa de alto voltaje impulsan la eficiencia en la electrónica moderna., Murata, https://www.murata.com/en-eu/s/blog/powering-the-future-how-high-voltage-mlccs-drive-efficiency-in-modern-electronics.html .
9. Estados Unidos 6,760,215, Condensador con alto umbral de ruptura por tensión, DF DeVoe (relacionado con Presidio), concedido en 2004, https://www.freepatentsonline.com/6760215.html .
10. GB/T 17702-2021, 电力电子电容器 (Condensadores para electrónica de potencia). Pekín: 国家市场监督管理总局, 2021.
11. Equipo de ingeniería de aplicaciones HVC, Condensador de cerámica de alto voltaje Recursos técnicos, https://www.hv-caps.com