تخميد الرنين الطفيلي الثانوي في محولات إشعال البلازما

تخميد الرنين الطفيلي الثانوي في محولات إشعال البلازما

الملخص

A plasma ignition transformer is a distributed resonant network, and its secondary parasitics decide whether the gap sees a clean breakdown pulse or a burst of narrow spikes. This article derives the layer and self resonances, the modal superposition, the damping branch and the ring-down criterion of a 1 to 1.5 transformer, then gives four pain points, six design steps and the capacitor data that keeps the damping on frequency.

1. مقدمة

The high-voltage ignition transformer is the least understood part of a plasma igniter. On the schematic it steps the tank voltage up to the level that breaks down the torch gap. In the machine it is a distributed resonant network: inter-turn capacitance, layer capacitance, leakage inductance, and winding-to-core and winding-to-ground capacitance all interact with the tank and the cable.

Every one of those resonances shapes the ignition pulse. Some help the breakdown, most of them hurt it, and the gap between an igniter that strikes reliably and one that misfires at random shows up in the management of the secondary parasitics, not in the nominal turns ratio.

Service engineers know the failures. A transformer ringing at a parasitic frequency delivers its energy at the wrong moment or in a burst of narrow spikes that break the insulation between turns instead of the gas gap. Field returns include secondary winding burnout, coupling capacitor failure and torch cable insulation damage.

The standard remedy, a damping resistor added by judgement, either changes nothing or removes the ignition pulse. This article puts the secondary resonances, the damping branch and the acceptance test into numbers.

2. الآلية الأساسية

2.1 The layer resonance

Each layer of the secondary presents series leakage inductance and parallel-plate capacitance to the neighbouring layer, and the two form a parallel resonance.

f_par = 1 / (2π × √(L_leak_s × C_layer))

where f_par is the layer resonance, L_leak_s is the secondary leakage inductance and C_layer is the layer-to-layer capacitance.

An 18 mH secondary leakage with 260 pF of layer capacitance puts that mode at 73.6 kHz. Its characteristic impedance follows the same two values: 8.3 kΩ. The mode is not a small correction on the ignition pulse. It carries enough energy to ring the winding for several cycles.

2.2 The series resonance of the winding

The winding as a whole resonates against its own self-capacitance to core and ground.

f_ser = 1 / (2π × √(L_sec × C_sec))

where f_ser is the series resonance of the secondary, L_sec is the secondary self-inductance and C_sec is the total secondary self-capacitance.

A 1.2 H winding with 45 pF of self-capacitance resonates at 21.7 kHz and shows 163 kΩ at that frequency. That mode carries the turns ratio and the ignition energy, so its Q has to stay high. Torch cable capacitance adds to C_sec directly: 4 m of cable at 50 pF/m takes the self-capacitance to 245 pF and the series mode down to 9.3 kHz.

2.3 Mode superposition at the gap

When the tank fires, the energy excites every mode at once and the secondary voltage becomes a sum.

v_sec(t) = Σ V_k × e^(−α_k t) × sin(2π f_k t + φ_k)

where v_sec(t) is the secondary voltage, and V_k, α_k, f_k and φ_k are the amplitude, decay rate, frequency and phase of mode k. The 15 kV ratio mode and a 35% layer contribution that peaks in the same instant add to 20.3 kV, so both the insulation and the damping capacitor see more than the nominal secondary voltage.

2.4 Where the damping has to act

Damping has to reduce the parasitic Q without touching the ratio mode, and the two modes present very different impedances.

R_damp = Q_target × √(L_mode / C_mode)

where R_damp is the resistance that sets the mode to Q_target at the mode frequency, and L_mode and C_mode are the values that define that mode.

The layer mode needs 2.1 kΩ in the loop for Q 4, and the ratio mode needs more than 1.6 MΩ to stay at Q 10. A plain resistor across the secondary cannot be both. A series R-C branch can: 1.5 kΩ with 1.5 nF presents 2080 Ω at 73.6 kHz and 5200 Ω at 21.7 kHz, which returns Q 4 for the parasitic mode and Q 31 for the fundamental.

2.5 Decay per cycle and the acceptance test

The amplitude of a damped mode falls by a fixed ratio per cycle, set by Q alone.

A_n = A_0 × e^(−π n / Q)

where A_n is the amplitude after n cycles, A_0 is the initial amplitude, n is the number of cycles and Q is the quality factor of the mode.

At Q 10 the mode still holds 39% of its amplitude after three cycles, and it never reaches the 10% acceptance level inside the ignition window. At Q 4 it falls to 9.5% in three cycles. The 10% in three cycles criterion therefore sets the target at Q 4.1 or lower, which is tighter than the Q 2 to 5 band often quoted.

2.6 Energy, loss and droop in the damping branch

Three checks bound the branch: stored energy, dissipation and droop.

E = ½ × C × V²

where E is the stored pulse energy, C is the capacitance and V is the voltage. The unit pair matters: with C in nanofarads and V in kilovolts the result comes out in millijoules, so 2.09 nF at 15 kV returns 235 mJ and not 235 J.

P = I² × R

where P is the dissipation, I is the branch current and R is the branch resistance. Amperes squared with milliohms return milliwatts, so the 220 A ring current through 1.5 kΩ reads as 72,600,000 and means 72.6 kW at the peak of the ring.

ΔV = I × Δt / C

where ΔV is the droop across the branch capacitor, I is the current, Δt the interval and C the capacitance. Amperes, microseconds and nanofarads return kilovolts, so 220 A for 120 ns takes the 1.5 nF branch capacitor down by 17.6 kV, which is why that part needs a 20 kV rating and not a 15 kV one.

Fig. 1 — The 73.6 kHz secondary mode falls to 9.5% of its initial amplitude in three cycles at Q 4, and stalls at 39% at Q 10.

3. نقاط الضعف الهندسية

Pain Point 1: Damping added without a measurement

A resistor placed across the secondary with no knowledge of the modal frequencies either overdamps the ratio mode and cuts ignition energy or leaves the parasitic mode untouched. The window between those outcomes is narrow: 16.6 kΩ to 41.6 kΩ across the secondary for the layer mode, against 1.6 MΩ needed to keep the ratio mode at Q 10.

Pain Point 2: Voltage concentration in the hot-end turns

A high-frequency mode distributes voltage unevenly along the winding. With a distribution parameter of 6 the first 10% of the turns take 45% of the applied voltage, and the gradient at the hot end runs six times the uniform value. A turn designed for 16.7 V then sees 100 V on every parasitic ring, and the insulation degrades until it flashes over.

Pain Point 3: The torch cable retunes the secondary

Cable capacitance adds to the secondary self-capacitance, so a machine reconfiguration moves the series mode. Going from 2 m to 4 m of cable takes the self-capacitance from 145 pF to 245 pF and the series mode from 12.1 kHz to 9.3 kHz, which can place it on the repetition rate of the ignition burst and make the transformer over-ring.

Pain Point 4: The secondary mode couples back to the primary

Secondary oscillations reach the primary through inter-winding capacitance and appear as spikes on the switch node and as false edges in the control electronics. The fault starts in the transformer, shows up as an EMC and reliability problem in the drive stage, and survives every fix applied at the switch.

4. خطوات التصميم

Strategy 1: Characterise before damping

Measure secondary impedance against frequency with the secondary open and again with it shorted, and record the frequency, amplitude and damping of every mode. The 21.7 kHz ratio mode at 163 kΩ and the 73.6 kHz layer mode at 8.3 kΩ are the two values the damping design needs.

Strategy 2: Tune the damping to the mode, not to the transformer

Use a series R-C branch of 1.5 kΩ and 1.5 nF, which presents 2080 Ω at 73.6 kHz and 5200 Ω at 21.7 kHz. That returns Q 4 for the parasitic mode and Q 31 for the ratio mode. Verify both numbers on the measured impedance curve before and after fitting the branch.

Strategy 3: Grade the winding against the high-frequency gradient

Use layer-by-layer voltage grading with progressive pitch or interleaved layers, and keep the first turns from the hot end away from the core. Hold the inter-turn margin with a partial-discharge test at 120% of the peak parasitic voltage, which is 24.3 kV on a 20.3 kV secondary peak.

Strategy 4: Set the cable interaction on purpose

Add a small tuning capacitor at the secondary so the combined secondary and cable resonance lands at a known frequency inside the ignition band, then damp that mode with the same branch. A cable change then moves a frequency the design already knows instead of re-tuning onto a dangerous harmonic.

Strategy 5: Use a capacitor with predictable high-frequency behaviour

The damping capacitor carries 220 A of ring current and sees 17.6 kV of droop per event, so it needs both a 20 kV rating and a stable capacitance. An N4700 Class I part holds the branch on frequency, which keeps the damping calculated at commissioning valid at the end of a shift.

Strategy 6: Validate with a worst-case ring-down test

Excite the secondary with a pulse generator, capture the ring-down, and confirm that the parasitic mode falls to 10% of its initial amplitude inside three cycles and that the ratio mode still delivers the design peak voltage into a simulated gap load. Repeat the test with the transformer at 85 °C and with the longest cable the machine will use.

معلمة المكون	القيمة النموذجية	شرط الاختبار
الجهد المقنن (HVCT8G سلسلة)	10 كيلو فولت العاصمة	جهد عالي 1.5× لمدة 60 ثانية
مواسعة	4700 بيكوفاراد ±20% (472 ميجا)	1 ميجاهرتز، 1 فولت RMS
المقاومة المكافئة التسلسلية عند 1 ميجاهرتز	أقل من 50 ملي أوم	-
الاستقراء الذاتي	أقل من 8 نانوهيدروجين	تباعد طرفي 40 مم
معامل درجة الحرارة	±30 جزء في المليون/درجة مئوية	40 درجة مئوية إلى +125 درجة مئوية
التفريغ الجزئي	أقل من 5 بيكو درجة مئوية	عند الجهد المقدر
قدرة نبض dv/dt	أعلى من 50 كيلو فولت/ميكروثانية	عند الجهد المقدر

5. HVC N4700 (T3M) capacitors for secondary damping

The damping branch holds its frequency only if the capacitor holds its value. HVC builds the N4700 (T3M) Class I family for that duty, with the مقاوم ذو طبقة سميكة for the branch and the مقاومة أخذ العينات for the divider, from a 6,000 m² owned factory in Dongguan across a 1 kV to 100 kV range.

ستة معايير مهمة هنا

• Damping capacitance that stays on frequency. A ±30 ppm/°C coefficient and a dissipation factor below 0.1% keep the tuned branch at the mode frequency from a cold start to a hot module, so the Q measured at commissioning is the Q in service.
• Voltage rating matched to the parasitic peak. Ratings from 1 kV to 100 kV let the branch capacitor survive the 20.3 kV superposition peak and the 17.6 kV of droop in one event, with margin above the nominal secondary voltage.
• No partial discharge of its own. A discharge level below 5 pC keeps the damping parts from adding a second source of ageing to a winding that already runs at 100 V per turn on every ring.
• Low self-inductance. A self-inductance below 8 nH keeps the branch from introducing a new resonance at the mode frequencies it exists to damp, which is the failure a wirewound or long-lead part brings.
• Stability across pulse count. The Class I formulation resists ageing, so the damping does not drift over millions of ignition pulses, the drift that returns in مكثف الفيلم designs as a loss of attenuation.
• Proven in field-service replacement. The 40 kV 441-series doorknob capacitors replaced first-tier brand parts in high-voltage field-service programs, and a إشارة الصليب against those parts shows the drift a damping design would have to absorb.

التوصيل والدعم

Standard parts ship in 4 to 8 weeks and matched damping sets in 6 to 10 weeks. HVC applications engineering supports impedance characterisation, mode identification, damping-branch calculation and ring-down test setup. Send the transformer turns ratio, secondary leakage, layer capacitance and cable length, and HVC returns a damping selection sheet with PD data and pulse-life reports within five working days.

6. أسئلة مكررة

Q1: Where does the 73.6 kHz mode come from?

From the secondary leakage inductance resonating with the layer-to-layer capacitance. With 18 mH of leakage and 260 pF of layer capacitance the mode sits at 73.6 kHz and shows 8.3 kΩ. It carries real energy, so it rings the winding for several cycles after every ignition pulse.

Q2: Why does a plain damping resistor fail?

Because the two modes need resistances 800 times apart. The layer mode reaches Q 4 with 2.1 kΩ in the loop, while the ratio mode needs more than 1.6 MΩ to stay at Q 10. A single resistor either overdamps the ratio mode and cuts ignition energy or leaves the parasitic mode ringing.

Q3: How do I choose the R-C branch values?

Set the branch magnitude at the parasitic frequency to the resistance that gives Q 4, which is 2.1 kΩ at 73.6 kHz. Splitting that between resistance and reactance gives 1.5 kΩ with 1.5 nF. The same branch then presents 5200 Ω at 21.7 kHz and leaves the ratio mode at Q 31.

Q4: What decay should the ring-down test show?

The parasitic mode must fall to 10% of its initial amplitude within three cycles. Amplitude decays as e to the minus pi n over Q, so Q 4 gives 9.5% at three cycles and Q 10 gives 39%. That criterion sets the target at Q 4.1 or lower.

Q5: Why do the hot-end turns see so much voltage?

A high-frequency mode does not distribute voltage uniformly along a winding. With a distribution parameter of 6 the first 10% of the turns carry 45% of the applied voltage and the local gradient reaches six times the uniform value. A turn rated 16.7 V then sees 100 V on every ring.

Q6: What does the cable do to the secondary?

Cable capacitance adds to the secondary self-capacitance. Going from 2 m to 4 m at 50 pF/m takes the self-capacitance from 145 pF to 245 pF and the series mode from 12.1 kHz to 9.3 kHz. If that lands on the burst repetition rate, the transformer over-rings, so set the combined resonance at a known frequency with a tuning capacitor.

7. المراجع

1. إي. كوفيل، دبليو إس زينجل، جيه. كوفيل، أساسيات هندسة الجهد العالي، الطبعة الثانية، نيونس، 2000.
2. موهان، ن.، أنديلاند، ت.م.، وروبنز، و.ب. إلكترونيات الطاقة: المحولات والتطبيقات والتصميم، الطبعة الثالثة. هوبوكين، نيوجيرسي: جون وايلي وأولاده، 2003.
3. إشعار رقم 2 من MIL-HDBK-217F، التنبؤ بموثوقية المعدات الإلكترونية، 1995.
4. IEC 60384-1: 2021 ، المكثفات الثابتة للاستخدام في المعدات الإلكترونية - الجزء 1: المواصفات العامة، الطبعة الرابعة. جنيف: اللجنة الكهروتقنية الدولية، 2021.
5. IEC 60270:2025، تقنيات اختبار الجهد العالي - قياس التفريغات الجزئية بناءً على الشحنة، الطبعة الرابعة. جنيف: اللجنة الكهروتقنية الدولية، 2025.
6. جي. دانتونا و إل. بيرفيتو، تحديد موقع التفريغ الجزئي في مفاتيح التبديل المعزولة باستخدام طريقة توسيع الدوال الذاتية، مجلة IEEE للمعاملات في مجال الأجهزة والقياس، المجلد 68، العدد 5، الصفحات 1294-1301، 2019. https://doi.org/10.1109/tim.2019.2904806.
7. إس إيه بهوميوات، تقييم حالة عزل جلبات المحولات باستخدام تحليل تيار الاستقطاب/إزالة الاستقطاب، في وقائع مؤتمر IEEE الدولي لعام 2004 حول العزل الكهربائي، الصفحات 500-503، 2004. https://doi.org/10.1109/elinsl.2004.1380664.
8. تطوير مكثفات سيراميكية متعددة الطبقات من نوع BME X7R ذات سعة عالية وجهد عالٍ، KEMET Electronics (CARTS USA 2011 Proceedings), https://www.mouser.jp/pdfdocs/2011-03CARTS-DevelopmentofHighCapacitance-HighVoltageBMEX7RMLCCapacitors.pdf .
9. الولايات المتحدة 10,529,487 ، مكثف سيراميكي متعدد الطبقات ذو أقطاب داخلية ذات مواضع حواف مختلفة، شركة تايو يودن المحدودة، منحت عام 2020، https://www.freepatentsonline.com/10529487.html .
10. ГОСТ IEC 61071-2014, المكثفات الإلكترونية السيلوفية (المكثفات لإلكترونيات الطاقة). موسكو: ستاندارتينفورم، 2014.
11. فريق هندسة تطبيقات HVC، الجهد العالي السيراميك المكثفات الموارد الفنيةhttps://www.hv-caps.com