Electrical Pulse Forming
# Electrical Pulse Forming: Capacitor-Discharge Dynamics, Micro-Alloy Thermal Balance & the Shockley Hook
Electrical pulse forming was the one process step that turned a pair of ordinary point contacts into an amplifier — everything before it (crystal growth, CP-4 polish, base stud soldering, whisker chiseling) only built two mechanically sound, electrically unremarkable metal-semiconductor junctions. As mechanically placed, those two junctions behave as nothing more than back-to-back point-contact diodes, with current gain well below unity. Forming's entire job was to deliver, in a few microseconds, enough localized energy to melt and re-freeze a microscopic volume directly beneath the collector whisker — converting a plain rectifying contact into a regenerative four-layer structure capable of current gain greater than one. Getting there required controlling three things precisely: how fast a charged capacitor could dump its energy, how much of that energy actually reached the melt zone instead of a current-limiting resistor, and what micro-structure the resulting melt-and-refreeze cycle left behind.
---
## 1. Why the Energy Budget Is Fixed Before the Pulse Ever Fires
A charged capacitor delivers a discharge whose voltage and current both decay exponentially with a time constant $\tau = RC$ set entirely by the series resistance and capacitance chosen *before* the pulse is triggered:
Total energy stored in the capacitor, and therefore the absolute ceiling on what the pulse can deliver no matter how long it is allowed to run, is fixed the instant the capacitor finishes charging:
for typical production values of $C\approx 0.5\,\mu\text{F}$ and $V_0 \approx 80\,\text{V}$. This is the central reason a capacitor discharge, rather than a continuous current supply switched on and off by a timer, was the forming circuit's design choice: a continuous supply's total energy delivery depends on exactly how long a switch stays closed, which is a far harder quantity to control to microsecond precision than a capacitor's charge, which is fixed and finite the moment it stops charging. The forming process is, by this design, self-limiting — there is a hard ceiling on deliverable energy regardless of any timing imprecision in the trigger circuit.
---
## 2. From Deposited Energy to Melt-Pool Radius
Only a portion of $E_{\text{pulse}}$ actually reaches the collector tip; the rest is dissipated across the series current-limiting resistor, whose value was chosen specifically to keep peak current — and therefore instantaneous power at the tip — within a range that melts a confined micro-volume rather than vaporizing or mechanically destroying the whisker itself. The energy that does reach the tip is deposited into a volume small enough, and briefly enough, that heat conduction away from the melt zone during the pulse itself is limited — an approximately adiabatic thermal problem over the pulse's short duration, so the melt radius follows an energy balance against the latent and specific heat needed to raise that volume to germanium's melting point:
Solving for $r_{\text{melt}}$ given $E_{\text{tip}}$ on the order of a fraction of a millijoule and germanium's known density, specific heat, and latent heat of fusion yields the observed melt radius of roughly $2$–$5\,\mu\text{m}$ — small enough to stay confined beneath the already-sharpened whisker tip from the previous process step, and large enough to create a continuous, electrically connected regrown region once it cools.
---
## 3. The Shockley Hook: Why Melting and Refreezing Produces Gain Above Unity
As the melt-pool cools, two competing effects determine the regrown region's final doping: rapid quenching generates lattice defects that act as thermal acceptors, while phosphorus carried in from the phosphor-bronze whisker itself simultaneously supplies donor atoms — the balance between them leaves a thin p-type shell immediately surrounding an n⁺ core at the collector tip, a micro-structure with four distinct layers (n-bulk / p-shell / n⁺-core, with the whisker metal itself as a fourth contact) where a simple point contact previously had only one junction. This structure provides a feedback path ordinary point contacts lack: holes injected from the emitter and arriving at the collector accumulate at the p-shell's space-charge barrier, and that accumulated charge *lowers* the barrier seen by electrons trying to leave the n⁺ core back into the metal whisker — pulling in more electron current than the arriving hole current alone would account for. The result is a current gain assembled from three multiplicative factors:
where $\gamma$ is emitter injection efficiency, $\beta$ is the base transport factor, and $\alpha^*$ is the collector's own barrier-lowering multiplication — the term that exists *only* because forming created the hook structure in the first place, and the reason $\alpha$ for a point-contact transistor can exceed unity when no single ordinary p-n junction ever could.
---
## Electrical Pulse Forming's Place in the Point-Contact Process Flow
| What Forming Controls | What It Depends On From Earlier Steps |
|---|---|
| Peak current density at the tip | Whisker tip contact area set by the point-chisel step |
| Melt-pool confinement | Whisker sharpness and base stud's low thermal resistance (heat sink path) |
| Hook-structure doping balance | Phosphorus content of the phosphor-bronze whisker material itself |
| Pulse self-limiting behavior | Capacitor energy $\tfrac12 CV_0^2$, fixed before any timing variation can matter |
Read electrical pulse forming through a *stored-energy-not-applied-power* lens rather than a *zap-it-and-see* lens: every design choice in the forming circuit — capacitor size, charge voltage, series resistance — exists to fix $E_{\text{pulse}}$ to a known value in advance, so that the melt-pool radius and the resulting hook structure's doping balance come out the same way on every single unit, converting what could have been an uncontrolled spark into the one process step that actually makes a point-contact transistor amplify.