1) What “Reverse Transformer” Means Here

We are not swapping primary/secondary windings. Instead, we flip the energy emphasis: the core path (iron mass shaped as or within the winding) is impulsed, and the output is taken from a capacitive shell around that core. The foil shell acts as C1, indirectly (capacitively) coupled to the core region. The goal is strong displacement-current pulses with reduced counter-EMF back into the driver.

2) Block Diagram at a Glance

3) How It Works (Field Picture)

• Impulse: The blocking oscillator produces steep dI/dt in L, magnetizing the iron path and launching a strong dΦ/dt.
• Displacement pickup: The nearby foil cylinder sees rapid dV/dt via capacitive coupling; a brief displacement-current surge flows in C1.
• Phase quirk: The two foil plates exhibit equal-and-opposite spikes (≈180°). Measuring AC across them can show cancellation.
• Rectify one lobe: Insert a diode so only one polarity passes. The observed result is a distinct ~24 V pulse train at C1.
• Counter-EMF relief: Because the pickup is indirect (capacitive), much of the return EMF doesn’t hammer the transistor the way a tightly-coupled secondary would.

Mentions of “scalar” behavior here refer to common-mode field cancellation at the measurement nodes; keep an open, testing mindset.

4) Parts & Example Starting Values

• 9 V battery, switch, fuse (250 mA–500 mA).
• NPN transistor (e.g., MJE13009, TIP41C, 2N3055) + small heatsink.
• Base resistor R: 470 Ω–2.2 kΩ (start ~1 kΩ; tune for clean oscillation).
• Drive/feedback winding L: 80–200 turns of 26–30 AWG magnet wire on a plastic former around an iron rod/bundle.
• C1 (foil tube):
  • Inner foil sleeve (bond a lead tab before wrapping), 1–2 layers of PVC tape or paper as dielectric, then outer foil sleeve with its own lead tab.
  • Length: 60–120 mm; spacing: 0.2–0.5 mm dielectric. Keep edges taped to avoid corona and shorts.
• Output diode: fast UF4007/FR207 (or Schottky with adequate PIV).
• Scope probes (x10), DMM, insulating hardware.

5) Step-by-Step Build

1. Prepare the core region. Fit an iron rod or tight iron-wire bundle inside a plastic tube/ former. Wind the drive/feedback coil L neatly; leave a center tap if using that topology.
2. Add the foil capacitor (C1). Wrap inner foil sleeve on the outside of the former, tape it fully, then add the outer foil sleeve. Keep a clear insulated gap between sleeves. Bring out two separate foil leads.
3. Wire the oscillator. Connect +9 V to the coil center-tap; one coil end through R to base; the other coil end to collector; emitter to ground. Add a diode clamp/snubber if your transistor runs hot.
4. Add the rectifier. From the C1 “hot” foil lead, go through a diode to your load/test jack; return to the other foil. (Polarity = choose the lobe you want.)
5. Power-on check. Scope CH1 on the collector (yellow), CH2 across the rectified C1 output (blue). Verify: tall flyback spike on CH1; ~20–30 V pulses on CH2.

6) Tuning & Notes

• Cancellation issue: Without the diode you may only read ~1–2 V AC due to 180° anti-phase. Keep the rectifier to select one spike.
• Capacitance matters: Longer/wider foil or thinner dielectric → larger C1 → slower but stronger pulses. Find a sweet spot for your iron/coil.
• Core options: Compare soft iron vs ferrite. Iron gives strong coupling at lower kHz; ferrite reduces eddy losses at higher kHz.
• Transistor stress: If heating, raise R, add an RC snubber across the coil, or reduce supply. Keep leads short.
• Scaling: Multiple C1 “taps” along the core (segmented sleeves) can provide several pulsed outputs; couple each through its own diode.

7) Troubleshooting

• No oscillation: Reverse the feedback leads (swap base/collector coil ends) or change R.
• Very low C1 output: Check for foil short, add/flip the diode, increase spacing symmetry, or adjust oscillator frequency.
• Transistor blows or gets hot: Add snubber/clamp, use a higher-voltage device, or reduce duty cycle.
• Noise on scope: Use x10 probes, short ground leads, and keep the build compact to reduce stray coupling.