Lilienfeld 1925 Preserve Metal Separation
# Preserve Metal Separation: The Clearance That Keeps Control Capacitive
Step 5 fixed how far apart the two current terminals sit from each other; this step addresses a completely different clearance — how far each terminal coating has to stay from the foil itself. The two requirements are easy to conflate but govern different failures. Step 5's gap sets a resistance; this step's clearance, if it fails, does not degrade anything gradually — it replaces the entire control mechanism outright. The whole concept depends on the foil's influence reaching the conducting film only through Step 3's oxide capacitance. A terminal coating that physically touches the foil shorts that capacitive path into a direct ohmic connection, and the device stops being a field-effect control electrode and becomes, at that one spot, just a third current terminal wired to the wrong place.
## 1. The Clearance Has to Survive the Same Tolerance Stack-Up Every Deposition Step Faces
Keeping the terminal coatings physically separated from the foil is not a single measurement — it is a budget that has to absorb every positioning and edge-definition tolerance the deposition process introduces, combined the same way independent tolerances combine anywhere else in a fabricated assembly:
where $d_{\text{nominal}}$ is the clearance the design intends between a terminal's edge and the foil, and each $\delta_i$ is an independent source of positioning error — how precisely the foil was seated in Step 4, how sharply the terminal coating's own edge was defined during deposition, how much the mask or stencil used to apply it could have shifted. As long as $d_{\text{min}}$ stays comfortably positive, direct contact never happens even under the worst realistic combination of these errors; let it approach zero, and an ordinary, unremarkable amount of process variation is enough to bridge the gap on some fraction of finished units.
## 2. Real Diagram: A Budget That Has to Stay Positive
## 3. A Bridge, Even a Small One, Turns a Capacitor Into a Resistor
If the clearance ever does close, the consequence is not a gradual loss of control sensitivity — it is a qualitative change in what kind of circuit element connects the foil to the film. Step 3's oxide capacitance normally couples the control electrode's voltage swings to the film through a purely reactive path; a direct metal bridge introduces a parallel leakage resistance that competes with that capacitive coupling for every signal passing between them:
A genuine metal-to-metal bridge drives $R_{\text{leak}}$ toward zero, and because a zero resistance dominates any finite reactance at every frequency, not just at DC, the oxide capacitance this entire concept depends on becomes irrelevant the instant that bridge forms. This is a meaningfully different failure than the kind of resistive bridging a purely rectifying device might tolerate in degraded form — here, the capacitive control mechanism has no fallback mode at all once $R_{\text{leak}}$ collapses; it is either present or it is gone.
## Real Diagram: A Short Overwhelms the Capacitor at Every Frequency
## Preserve Metal Separation's Place in the Process Lineage
Preserve Metal Separation follows Step 5, Form Two Metal Terminals, whose deposition process is the actual source of the positioning errors this step's clearance budget has to absorb; it precedes Step 7, Deposit Thin Copper, which will cover the terminals, the intervening glass, and the exposed foil edge all at once — making this the last point before that blanket deposition where terminal-to-foil clearance can still be verified directly, rather than inferred after the fact. It is the sixth step of this concept's construction sequence and the one step whose entire purpose is confirming that nothing Steps 3 through 5 did has already quietly undermined the field-effect mechanism the rest of the device is built around.