A pre-biased transistor is a small-signal BJT with its bias resistor built into the package, and it exists to solve a specific board problem: driving a transistor directly from logic without a separate resistor, a capacitor, or a spare layout trace. The package looks like an ordinary transistor, but the datasheet hides two resistors — typically a 10 kΩ and a 47 kΩ pair — that change how the part is read, biased, and driven. Misreading those internal resistors is the easiest way to burn the part or starve the logic it serves. This article explains why the resistors are inside, decodes the part number, and shows what the internal bias actually means for the output current a logic drive can produce.
The Resistor Pair Inside: R1/R2 and the Two Reasons They Exist
The internal resistor pair serves two jobs, and both come from the transistor’s role in a logic drive. The first resistor (R1, typically the larger value) sets the base bias from the collector supply, so the part can be driven single-ended without a separate base path. The second (R2, smaller) provides a shunt to ground that keeps the base at a defined level when no drive is present — the transistor defaults to off, which is what a logic output expects from an unbiased input.
The two-reason framing explains the pair: R1 feeds a defined standby bias, and R2 holds the base quiet when the drive is absent. Without R2 the part could drift on with leakage or noise; without R1 the logic input would need its own resistor in the rail. The pre-biased transistor explainer on the blog covers the family role, and the digital switching transistor article places it in the logic-drive family this package belongs to.
The physical consequence is that the part is not a plain BJT plus a free resistor — it is a biased BJT with known, fixed operating points. The bias sets the collector current at a chosen rail voltage, and the fixed pair means the designer gives up the freedom to tune the bias with a custom external resistor. That trade is the subject of the final section, where unbaking the resistor externally becomes the upgrade path.
Reading the Code: How the 10k/47k Pair Is Marked into the Part Number
The resistor pair is usually marked directly in the part number or the family code, and reading the code is the first step in using the part. A common convention carries the two resistor values in the number: a “47k” style suffix or a code position that maps to the R1/R2 pair. The part number’s digits tell you the two internal resistors at a glance, saving a datasheet lookup for the ten most common code values.
The discipline transfer: never assume the internal pair from the package shape — two physically identical-looking parts can carry very different R1/R2 values, and the drive current follows the resistor values, not the look. A part marked with a larger R2 biases “quieter” and needs a stronger drive to switch; a part with a smaller R2 biases “hotter” and switches with less input but burns more in standby.
The decoding table for the common codes lives in the transistor family documentation, and the general-purpose NPN and PNP profiles show the un-biased counterparts the code is compared against. Reading the code replaces guessing, and it is the single most useful habit for working with these parts. A quick rule for the bench: when the marking is faint or unfamiliar, treat the part as un-verified until the internal pair is confirmed against the family table, because a misread bias pair drives the wrong load and fails in exactly the silent way this article’s current math is meant to catch.
Output Current Reality for Logic Drive
The internal bias sets a hard ceiling on the output current a pre-biased part can deliver from a given logic rail, and the ceiling is often lower than the beginner expects. The collector current at the bias point is approximately the bias voltage divided by the resistance presented by the R1 path through the transistor gain, and it is set by the internal resistors, not by the logic drive alone.
The drive-limits table below shows the pattern:
| Logic rail | Internal bias | Typical bias collector current | Useful load limit |
|---|---|---|---|
| 3.3 V | 10k/47k pair | Low (sub-mA) | Light logic loads |
| 5 V | 10k/47k pair | Modest | Small relays, LEDs |
| 5 V | 47k-driven variant | Lower | Signal-level only |
| 12 V | Larger R1 | Higher bias | Heavier loads |
The table is the “reality” of pre-biased parts: they are signal-drive components, comfortable switching light loads and logic-level outputs, not power stages. The practical reading is that a pre-biased transistor is selected for its bias point as much as for its transistor gain — two parts with the same transistor can drive very different loads purely because their internal resistor pairs differ, which is why the datasheet line for the internal pair deserves the same attention as the gain number. A pre-biased part asked to switch a heavy relay coil will simply not develop the collector current — the internal bias is a signal-level design. The relay drive sizing method is the companion that applies the current math for the heavier job.
Selecting a Pre-Biased Part for 3.3 V Logic vs 5 V Logic
Selection across 3.3 V and 5 V logic returns two different answers because the rail voltage sets the available bias current. On a 3.3 V rail, the internal bias has less voltage to work with, so the useful collector current is small and the part should drive only the lightest logic loads — an LED, a small signal relay. On a 5 V rail, the bias voltage is higher and the same part develops more usable collector current, extending the load range.
The honest selection check is the current math at the actual rail. Compute the bias collector current from the part’s internal resistors and the rail voltage, compare it to the load current the design must switch, and reject the pre-biased part if the load exceeds the bias capability — that is the point where the design needs either a different candidate or the external-bias upgrade in the final section.
The small-signal transistor family lists the pre-biased and general-purpose variants side by side, and the SOT-23 bipolar profile covers the dominant small-signal package these parts ship in. The selection is a two-number decision — bias current from the rail, load current required — and the pre-biased part wins only when the first comfortably exceeds the second.
A worked selection makes the two-number decision concrete. A design must drive a 15 mA LED indicator from a 5 V logic rail with the smallest footprint. The candidate pre-biased part lists a 10k/47k internal pair; at 5 V the bias through R1 develops roughly 0.1–0.2 mA of collector bias, which through a typical hFE of 50–100 gives a usable output in the few-mA-class, just below the 15 mA the LED needs. The check fails: the part’s internal bias cannot develop the load current. The redesign has two options — a second pre-biased part with a hotter internal bias, or the regular BJT with the external resistor pair tuned so the rail delivers the 15 mA with margin. Matching the part to the rail and the load, rather than grabbing the first pre-biased package in the family, is exactly the selection habit this article is building.
The board-level verification is the last step before trusting the selection. With the part wired, drive the logic input and confirm the collector actually switches the load to the expected current — a pre-biased part that was selected on a wrong rail assumption will visibly fail to drive, and the fix is the current math on the real supply voltage, not a bigger package. The digital switching transistor profile and the SOT-23 bipolar package article are the two published resources that complete the family picture the selection reads against.
Upgrade to a Regular BJT and Bake the Resistor Externally for Custom Bias
The upgrade path for a pre-biased part that does not quite drive the load is not a bigger package — it is the regular BJT with the bias baker externally. Removing the internal pair and placing an external R1/R2 on the board gives back the design freedom the pre-biased package traded away: the external pair is tuned to the exact rail, load, and temperature margin the design needs.
The trade is layout cost. The external resistors consume board area, add a layout requirement, and undo the “no separate resistor” convenience that made the pre-biased part attractive. The decision is the standard packaging-versus-performance trade: the pre-biased part wins when the signal-drive fits and the board is tight; the regular BJT with an external pair wins when the current math needs tuning beyond the fixed internal values.
The conversion is measurable and reversible, which is why the transistor category is read with both the pre-biased and the general-purpose families in view. The BJT test method verifies either package out of the drawer, and the logic-MOSFET upgrade path completes the ladder when even the externally biased transistor is no longer the right tool for the load.