A 3.3 V rail that loses 0.7 V to a diode is not losing a rounding error — it is giving up 21% of its voltage before the load even sees it. On a 5 V rail the same diode takes 14%. Those two numbers are why the “just grab any diode” habit stops working the moment you build anything powered at logic voltages, and why the choice between silicon, Schottky, and germanium is one of the few places in a low-voltage design where a single component decision measurably changes how the whole rail behaves. This article compares the three materials on the dimensions that actually decide a low-drop design — forward drop, leakage, reverse voltage, temperature behavior, and cost — then gives a selection path for 3.3 V and 5 V rails.
The 21% Loss Problem on a 3.3 V Rail
Start with the arithmetic, because it sets the stakes. A silicon diode conducts with a forward voltage drop of roughly 0.7 V at moderate current. Feed a 3.3 V rail through it and the usable voltage entering the load is 3.3 − 0.7 = 2.6 V. The diode has consumed 21% of the rail. The same part on a 5 V rail costs 14%. A Schottky diode at roughly 0.3 V drops the loss to 9% on the 3.3 V rail and 6% on the 5 V rail; a germanium diode near 0.2 V brings the 3.3 V loss down to about 6%.
Those percentages translate differently depending on what the rail feeds. For a load that regulates its own input, the diode drop shows up only as waste heat and efficiency loss. For a rail that feeds a fixed-voltage load directly, the drop is a budget problem — the diode narrows the headroom between the supply and the load’s minimum voltage, and the whole design tightens to compensate. Many 3.3 V designs fail not because the supply is weak but because the forward drop of a single diode consumes the margin the rest of the circuit was counting on.
The knee point matters as much as the headline number. Diode forward voltage is not a constant; it moves with current and temperature, and the relationship is exponential rather than a fixed “0.7 V” tag. The physics behind that curve — why a low-current reading can show 0.48 V and a heavily loaded one 0.9 V — is the subject of the forward voltage physics guide. For selection, the practical consequence is that you must compare materials at the current your rail actually carries, not at the datasheet’s reference current.
Three Materials, Three Drop Voltages: A Comparison Table
The honest comparison is five-dimensional, because the forward drop is the headline feature but not the only one that decides a working rail.
| Material | Typical VF (moderate current) | Leakage at 25 °C | Practical reverse voltage | Temperature sensitivity | Typical use |
|---|---|---|---|---|---|
| Silicon (PN) | 0.6–0.8 V | Low (nA–µA) | High (tens of V and up) | −2 mV/°C approx. (see curve) | General rectification, high-voltage rails |
| Schottky | 0.25–0.45 V | Higher (µA–mA) | Lower (Varies by design, volts to tens of V) | Drop falls with heat; leakage rises sharply | Low-drop, high-efficiency, low-voltage rails |
| Germanium | 0.2–0.35 V | Highest for normal junction types | Low (typically a few V or less) | Markedly temperature-sensitive | Signal detection, legacy low-voltage circuits |
Two patterns stand out. Schottky and germanium both win the drop battle, but they pay in different currencies. A Schottky’s advantage is thinner — it trades leakage and reverse-voltage headroom for forward drop, which is exactly why a Schottky on a 3.3 V rail can be the right choice while the same part on a 12 V rail would be the wrong one. Germanium’s low drop is genuine but it carries the worst leakage and the lowest practical reverse voltage of the three, which is why it survives mostly in signal diodes and specialty detection instead of power rails.
The table also hides the within-family spread that the datasheet makes visible. Two common parts illustrate the gap. A BAT54 is a small-signal Schottky rated for modest current; a 1N5819 is the classic low-drop power Schottky. Both sit in the same Schottky column, yet their VF at a given current and their useful current range differ enough that picking between them without the curve is a guess. Germanium sits in the opposite situation: the material’s low drop is well documented, but stock is thin and most modern 3.3 V and 5 V rails reach the same result more cheaply with a Schottky, leaving germanium to signal detection and legacy repair work. The correct habit is to treat the table as a starting scale and the datasheet comparison guide as the actual decision tool, since test current and temperature are part of every VF claim.
Where Low VF Pays for Itself: OR-ing, Reverse Polarity, and Low-Voltage Buck
Low forward drop is a design feature, not a virtue by itself. Three circuit patterns make the difference big enough to matter.
OR-ing or diode-combined supplies is the clearest case. When two rails are diode-ORed so either can supply a load, the diode sits in the main power path permanently, and its drop is a continuous loss at full load current. A 0.4 V difference between silicon and Schottky in an OR-ing stage on a 1 A, 3.3 V rail is 0.4 W of continuous dissipation and a 12% rail difference — the sort of number that moves battery life and thermal design. Reverse-polarity protection has the same profile: the protection diode carries the whole rail current normally, so its drop is pure cost until the fault it guards against ever happens. The reverse polarity protection guide is the practical context for this pattern in an entry-protection role.
Low-voltage buck converters are the third, more subtle win. The conversion path in a buck is lossy before the diode is involved, so a 0.2–0.3 V diode saving appears small against the converter’s total loss — but it is a saving that scales with current and never goes away, and on rails already tight on efficiency it is often the cheapest last 2% available. Where the design really needs every millivolt, the low-VF Schottky families are the parts optimized for exactly this duty, and the low VF Schottky guide covers their datasheet behavior and package options.
Two Hidden Costs: Schottky Leakage and Low Reverse Voltage
Every material comparison that stops at the forward drop is incomplete, because the two most attractive low-drop materials carry costs that show up only in specific operating conditions.
Schottky leakage is the first. A Schottky diode conducts more reverse current than a silicon PN junction of similar size; the number is typically orders of magnitude higher, and it climbs steeply with temperature. On a rail that floats at low current or sits in standby, leakage becomes a continuous drain that the forward-drop saving has to justify. The leakage measurement guide shows why the semiconductor type matters as much as the individual part, and why a “10 µA at 25 °C” claim can become tens of milliamps in a hot enclosure. In battery-powered or standby-heavy designs, this margin can matter more than the 0.2 V you saved.
Low reverse voltage is the second hidden cost. Schottky diodes deliberately use a low barrier, and the barrier that gives them a low forward drop limits how much reverse voltage the junction can hold off. Datasheet reverse-voltage ratings are lower, and derating rules — the margins you must keep between the working rail and the rated limit — are stricter. On a 3.3 V or 5 V rail this rarely binds, but the same diode family is sold for higher-duty work, and applying a low-drop part outside its reverse-voltage envelope is a real thermal and reliability hazard that the rectifier voltage ratings guide spells out.
A Quick Selection Flow for 3.3 V and 5 V Rails
The decision reduces to four questions, in order.
First, what is the rail voltage? Below roughly 5 V, low forward drop begins to matter; above it, reverse-voltage headroom and robustness usually dominate. Second, what current and temperature does the rail actually see? The drop comparison must be done at the real current and a realistic junction temperature, not at a datasheet reference. Third, does the design tolerate leakage in standby or in hot conditions? If the rail idles or spends time elevated, Schottky’s leakage tax may outweigh its drop saving. Fourth, is the device the only protection against reverse current, or is it a series loss in a path that already has protection? The answer decides whether the diode is being chosen for drop or for reverse-blocking duty.
The table omits a fourth candidate that increasingly matters at higher power: a SiC Schottky barrier diode keeps the low forward drop of a Schottky at much higher voltage and temperature ratings. It costs more and carries its own leakage profile, so it rarely wins a 3.3 V hobby rail, but the SiC device category is the right place to look when the same low-drop requirement appears at 400 V or in a hot enclosure where a conventional Schottky would derate into unusability. Running those four questions in order on a 3.3 V logic rail usually lands on a Schottky for the main path and a silicon part for the reverse-blocking stage; on a 5 V rail with standby, the same flow often favors silicon and saves the Schottky for the OR-ing stage where it earns its keep continuously. Germanium appears mostly when the application is signal detection or a legacy circuit rather than a power rail drawing amps. When you are ready to browse actual parts, the Schottky rectifier category and the silicon rectifier category cover both families, and the Schottky selection guide walks the same material trade-off through a full part-selection process.