Point Contact Transistor Base Common Reference Terminal
# Use the Base as the Common Reference Terminal: The Sequence Stops Building and Starts Operating
Step 16 sealed a device with three external leads and nothing yet telling anyone how to connect them. Every step before this one built, verified, or protected a physical object; this is the first step whose subject is a circuit, not a component. Three terminals come out of the cartridge — emitter, base, collector — and a circuit needs two loops, one carrying the input signal and one delivering the output, which means one of the three terminals has to be shared between both loops. Step 17 is the decision that it will be the base, and that decision is not a wiring convenience. It is forced by the asymmetry this series has been building into the emitter and collector since Step 1.
The emitter and collector were never interchangeable, and that history rules out using either of them as the shared terminal. Step 1 chose N-type germanium specifically so that holes, not electrons, would be the species injected and collected. Step 13 placed the collector relative to an already-seated emitter, not the reverse, because the emitter's injection has to exist before the collector's collection means anything. Every step since has treated the two point contacts as asymmetric: one forward-biased to inject, one reverse-biased to attract what survives the crossing. A shared reference terminal has to sit outside that asymmetry, available to both loops without already being committed to one carrier role or the other. The base is the only terminal in the device that was never asked to inject or collect anything — its entire function since Step 5 has been to spread current across the whole body with as little resistance as possible, which is exactly the kind of neutral, low-impedance node a shared reference needs to be.
Choosing the base as common is also what makes the device's actual gain mechanism work, even though the device does not amplify current in the ordinary sense. The emitter operates forward-biased into a low-impedance junction; the collector operates reverse-biased into a junction that presents a much higher impedance to the same current. A point-contact device built this way does not generate current gain by itself — the current reaching the collector is, if anything, a fraction of what the emitter injects, not a multiple of it. What the common-base configuration delivers instead is power gain: the same current, encountering a far higher impedance at the output than at the input, produces a far larger voltage swing there, and power is the product of the two. This is why the configuration matters as much as the physical device — a correctly built point contact wired the wrong way delivers no useful amplification at all, regardless of how well Step 14's forming pulse worked.
Nothing about the device itself changed between Step 16 and this step; only the description of how to connect it did. The physical terminals, the formed point contacts, the base electrode's low spreading resistance — all of that was finished and sealed before this step began. Step 17 adds no new material and performs no new process operation on the device at all. It specifies, for the first time, the one circuit topology that respects the roles every earlier step assigned to each terminal, which is why this step marks a real boundary in the sequence even though it leaves the device itself completely untouched.
| Step | Process operation | Input | Output | Specification | Constraint |
|---|---|---|---|---|---|
| 17.1 | Identify which of the three terminals carries no carrier-selective role of its own | Sealed device with three terminals from Step 16 | Candidate common terminal | Terminal whose function since fabrication has been spreading current, not injecting or collecting carriers | Choosing a terminal already committed to injection or collection forces that role to also serve as a neutral reference, which it cannot do cleanly |
| 17.2 | Confirm the base terminal's identity against the device's provenance record | Candidate common terminal from 17.1, provenance from Step 2.6 and Step 15.6 | Confirmed base terminal | Terminal traced back to the broad, low-resistance electrode established in Step 5 and Step 6 | Connecting the wrong lead as "base" on a mislabeled or miswired unit defeats every later step's bias assumptions |
| 17.3 | Establish the base as the single node shared between the input loop and the output loop | Confirmed base terminal from 17.2 | Common-base circuit topology | Base node electrically common to both the emitter-base loop and the collector-base loop | A topology that routes the emitter or collector into the shared node instead reintroduces the asymmetry this terminal was chosen to avoid |
| 17.4 | Verify the base connection presents low enough impedance to serve as a shared reference without becoming a signal path of its own | Common-base topology from 17.3 | Verified low-impedance common node | Base impedance small relative to the impedances the emitter and collector loops will operate at | A base connection with unexpectedly high impedance can let signal leak between the two loops through the node meant to keep them independent |
| 17.5 | Document the finalized common-base configuration against the device's complete build record | Verified topology from 17.4 | Documented common-base circuit | Configuration recorded against every provenance record carried forward since Step 2.6 | Without this record, later measurements in Steps 18 through 22 cannot be tied back to a specific, confirmed circuit configuration |
Step 17 does not change what the device is; it decides what the device is allowed to become once current flows through it. Every physical decision this series made — N-type germanium, a conditioned contact face, a broad low-resistance base electrode, two asymmetric point contacts — was made in anticipation of a circuit this step is the first to actually specify. The base was never going to be a free choice among three equivalent options; it is the only terminal that was built, from Step 5 onward, to do nothing carrier-selective at all, which is precisely what a shared reference terminal has to be.