English

简体中文

English

简体中文

Appearance

How to Select a Rectifier Diode: Selection Points for Schottky and Fast Recovery Diodes ​

Bottom line: the selection order for a rectifier diode is first fix the device category by voltage class and switching frequency, then fix the specification by current and loss, and finally fix the margin by thermal design. Schottky wins on "no reverse recovery, low forward drop", but its voltage ceiling is low and its leakage current is large; fast recovery wins on "high voltage rating", at the cost of reverse recovery loss. Using these two families in the wrong position — for example a Schottky at a high-voltage rectification point, or a general-purpose rectifier at a secondary running at tens of kHz — is the most common selection failure on adapter, energy-storage and charging products.

First, Answers to the Three Most Frequently Asked Questions ​

Q: What really distinguishes a Schottky from a fast recovery diode?

The essential difference is reverse recovery. A Schottky is a majority-carrier device and has no reverse recovery process at turn-off — only junction-capacitance charge and discharge; a fast recovery diode has stored charge that must be removed at turn-off, producing reverse recovery current. This difference directly determines that Schottky suits high-frequency low-voltage use and fast recovery suits high-voltage use. On forward drop, Schottky is typically 0.35~0.6V and fast recovery typically 0.7~1.0V.

Q: How much does reverse recovery loss actually matter?

In every cycle, when the diode has to go from conduction to blocking, the reverse recovery charge Qrr is discharged, and the corresponding loss is roughly Qrr × reverse voltage × switching frequency. The higher the frequency and voltage, the less negligible this loss becomes — which is why a general-purpose diode is fine for line-frequency rectification while a flyback secondary or the freewheeling point of a synchronous Buck must use Schottky or ultrafast recovery, otherwise the diode itself heats up first.

Q: Where are diodes used in products such as energy storage and BMS?

Far more widely than the word "rectification" suggests: secondary rectification in flyback auxiliary supplies, freewheeling or synchronous rectification in DC-DC, reverse-polarity protection at the battery output, reverse-current blocking in MPPT, clamping in drive and sampling loops, and so on. Every position has different voltage, current and frequency conditions, so the device category and derating strategy differ too — this article gives configuration advice position by position.

Step 1: Fix the Device Category by Voltage Class and Switching Frequency ​

This is the fork in the road during selection. The capability boundaries of four common rectifier families:

Device categoryReverse recoveryForward drop (typical)Voltage range (typical)Suitable scenario
Schottky diodeNone (junction-capacitance charge / discharge only)0.35~0.6VSilicon Schottky generally ≤100VLow-voltage high-frequency rectification, freewheeling, reverse-polarity protection
Silicon carbide SchottkyAlmost noneRises steeply with current650V and aboveHigh-voltage high-frequency (PFC, PV boost)
Ultrafast recovery diodeAbout 25~75ns0.7~1.0VSeveral hundred V to over 1000VMedium / high voltage, relatively high-frequency rectification
Fast recovery diodeAbout 100~500ns0.7~1.0VSeveral hundred V to over 1000VMedium / high voltage, rectification at modest frequency

The table gives category-level typical ranges; within the same category there is large variation by process and part number, so specific parameters are subject to the manufacturer datasheet.

Two rules of thumb:

  • Rectification point below 100V with switching frequency above tens of kHz: prefer Schottky. Zero reverse recovery means the Qrr term is absent from switching loss, and a difference that is invisible at low frequency gets amplified into the dominant heat source at tens of kHz.
  • Voltage beyond the Schottky range: look at frequency. Above tens of kHz prefer ultrafast recovery; at line frequency or low frequency, fast recovery or even a general-purpose rectifier is enough — a general-purpose diode has reverse recovery time in the microsecond range, which does not matter at line frequency but is a disaster in a switching supply.

Also watch the leakage current behavior of Schottky: reverse leakage current is markedly larger than in a PN-junction diode and rises exponentially with junction temperature. Leakage loss in high-temperature environments and the potential for thermal runaway (see the misconceptions section below) are a check that Schottky selection must perform separately.

Step 2: Fix the Specification by Current and Loss ​

Diode loss is made up of three main parts:

  • Forward conduction loss: P ≈ VF × Iavg. Note that Iavg is the average rectified current converted from the waveform, not the output current — for example the flyback secondary current is a triangular wave whose peak may be 2~3 times the average, so selection must check the converted average and the peak separately (the average maps to loss, the peak maps to the dynamic rise of VF and to the maximum rating).
  • Reverse loss: reverse voltage × reverse leakage current. For Schottky this term rises significantly at high temperature and cannot be ignored in high-temperature applications.
  • Reverse recovery loss (PN-junction types only): Qrr × Vr × f. The higher the frequency and voltage, the more critical it is.

Checklist when fixing the specification:

  1. Reverse voltage rating: check against the peak reverse voltage in the circuit (including leakage-inductance spikes and bus ripple troughs), and leave more than 20% derating. Checking only against the average or nominal voltage is one of the main causes of blown devices.
  2. Average forward current: compare against the datasheet definition conditions (usually average rectified current at a specified case temperature) and the actual thermal conditions — datasheet ratings are mostly numbers obtained under ideal cooling.
  3. Peak forward current: cover transient conditions such as startup, surge and short circuit.
  4. Junction temperature: derate against a Tj maximum of 150°C (for most devices); in engineering practice the continuous operating junction temperature is often held within 80~105°C to guarantee lifetime and thermal-runaway margin.

Step 3: Thermal Design Sets the Margin ​

Thermal accounting for a diode is the same arithmetic exercise as for a MOS: loss × thermal resistance = temperature rise. Schottky in particular must be calculated for the high-temperature condition — VF drops slightly with temperature (good for efficiency), but leakage current rises exponentially with temperature; when the two effects combine, the worst point is often not full load at room temperature but full load at high temperature. Key points:

  • Use the datasheet RθJC / RθJA together with the actual cooling conditions to calculate steady-state temperature rise; do not simply apply rules of thumb such as "TO-220 handles 10A";
  • For pulsed conditions, use the transient thermal impedance curve (Zth) rather than steady-state thermal resistance;
  • In layout, keep rectifier devices away from heat-sensitive parts such as electrolytic capacitors and MCUs — Schottky leakage is temperature-sensitive, and thermal coupling amplifies the problem.

Selection Flowchart ​

mermaid
flowchart TD
    A["Define operating conditions<br/>rectification point voltage / average current / switching frequency"] --> B{"Reverse voltage<br/>within the ≤100V class?"}
    B -- "Yes" --> C["Schottky diode"]
    B -- "No" --> D{"Switching frequency"}
    D -- "Line frequency / low" --> E["Fast recovery / general-purpose rectifier"]
    D -- "Tens of kHz and above" --> F["Ultrafast recovery or<br/>silicon carbide Schottky"]
    C --> G["Derate check against peak reverse voltage<br/>including spikes and ripple troughs"]
    F --> G
    E --> G
    G --> H["Convert average and peak current<br/>calculate conduction / reverse / recovery loss"]
    H --> I{"High-temperature full-load junction temperature<br/>≤ derating target?"}
    I -- "No" --> J["Larger package / lower Rθ /<br/>switch to synchronous rectification or a MOS solution"]
    J --> H
    I -- "Yes" --> K["Confirm the periphery by circuit position<br/>snubber network / layout thermal isolation / parallel current sharing"]
    K --> L["Measure: reverse voltage waveform +<br/>case temperature + efficiency"]

Typical Configuration by Circuit Position ​

Circuit positionMain operating conditionRecommended device categoryKey check points
Flyback / forward secondary rectificationHigh-frequency high-current pulsesSchottky (low voltage) or ultrafast recovery (high voltage)Peak current conversion, reverse spike snubbing
DC-DC freewheeling / synchronous rectification fallbackHigh frequency, continuous conduction at light loadSchottky or synchronous MOS body diodeDead time, light-load reverse conduction
Battery / power supply output reverse-polarity protectionContinuous high currentSeries Schottky, or P-MOS / ideal diodeContinuous loss of drop × current
PV input reverse-current blocking (MPPT)High voltage, continuousSchottky (low-voltage systems) or contactor / MOS solutionStatic loss and drop budget
Auxiliary supply, drive loop rectificationSmall current, high frequencyUltrafast recovery or small-signal SchottkyEffect of reverse recovery on switching waveforms

Voltage and current conditions differ greatly from product to product at each position, so the table above only indicates the category direction; point-by-point parameters are subject to the overall design and the manufacturer datasheet.

Reverse-polarity protection deserves more detail: a 0.5V drop at a 10A load is 5W of continuous loss, and in a low-voltage high-current system that is a painful figure. So for small-current applications a series Schottky is the lowest-cost option; for high-current applications prefer a P-MOS / N-MOS or ideal-diode controller solution, which pushes the loss from watts down to hundreds of milliwatts.

Common Misconceptions ​

  • "Selecting the Schottky voltage rating by the nominal voltage is enough": it must be checked against the reverse peak (including leakage-inductance spikes and ripple troughs) and derated. The actual reverse voltage at a flyback secondary is often much higher than the output voltage.
  • "Schottky has no reverse recovery, so any high frequency is fine": junction-capacitance charge / discharge loss rises linearly with frequency, and at very high frequency this term is likewise significant — just much smaller than the Qrr loss of a PN-junction diode.
  • "Leakage current is small enough to ignore": at room temperature it is indeed small, but Schottky leakage rises exponentially with junction temperature, and high temperature at full load is the trigger condition for thermal runaway — it must be checked under the worst-case condition.
  • "Paralleling solves everything": diode VF has a negative temperature coefficient, so parallel devices naturally drift toward unequal current sharing (the device with the lower VF takes more current). Current-sharing design or same-batch matching is required — paralleling is not simply "divide the current by 2".
  • "Only conduction counts as loss": at a high-frequency rectification point, reverse recovery loss can exceed conduction loss, so selecting on VF × I alone underestimates the total.

FAQ ​

Q1: What is the difference between Schottky and fast recovery diodes, and how do I choose for a rectifier circuit? ​

The core difference is reverse recovery and forward drop: Schottky has no reverse recovery process (only junction-capacitance charge and discharge), with a forward drop typically of 0.35~0.6V, suiting low-voltage, high-frequency secondary rectification and freewheeling; fast recovery diodes have reverse recovery time of several hundred nanoseconds and a forward drop typically of 0.7~1.0V, but can be made with high voltage ratings, suiting rectification above 200V or at line frequency and high voltage. Rule of thumb: if the rectification point is below 100V and switching frequency is above tens of kHz, prefer Schottky; if the voltage exceeds the silicon Schottky range (generally not above 100V — go to silicon carbide Schottky or fast recovery beyond that), choose between fast recovery and ultrafast recovery by frequency. Specific parameters are subject to the manufacturer datasheet.

Q2: Why does a Schottky diode always run hot or even burn out? ​

Four common causes: first, insufficient reverse voltage margin — the Schottky reverse voltage ceiling is generally not high, so it must be checked against the peak (including leakage-inductance spikes and ripple troughs) rather than the average, with more than 20% derating; second, leakage-current thermal runaway — Schottky leakage is large and rises exponentially with junction temperature, so when cooling is inadequate the leakage grows, loss grows further, and a vicious cycle forms; third, the actual average current is underestimated, since the average rectified current converted from duty cycle and waveform is often higher than the rated output current; fourth, the loop lacks a snubber network, so a reverse spike breaks the device down. Suggested troubleshooting order: measure the reverse voltage peak first, then case temperature and junction-temperature margin, then check the current waveform.

Q3: Can a synchronous rectification MOSFET completely replace a Schottky diode? ​

In most medium- and low-voltage, high-current applications it can, with clear benefit: the forward drop can be brought down to tens of millivolts, more than an order of magnitude lower than Schottky, so efficiency improves significantly. But three points need attention: first, drive and control complexity — a dedicated controller or synchronous rectification logic in the main controller is required, and at light load there is a risk of false turn-on and reverse conduction; second, standby and light-load efficiency are not necessarily better, since dead time and switching loss eat part of the gain; third, reliability design — when control fails, conduction normally falls back to the body diode, so the diode path still has to be checked. Synchronous rectification is therefore already mainstream in efficiency-sensitive products such as portable power stations and adapters, while Schottky is still common in small-current or highly cost-sensitive applications.

Q4: What diode should be used for reverse-polarity protection of a battery or power supply output, and how is the drop loss calculated? ​

The traditional approach is a series Schottky diode for reverse-polarity protection — simple and reliable — but the loss equals forward drop × load current: for example a 0.5V drop at a 10A load is 5W of continuous loss, which is a heavy price in low-voltage high-current applications. There are two improved options: one is to replace the diode with a P-MOS (high side) or N-MOS (low side) for reverse-polarity protection, where the drop is set by the on-resistance and the loss can be reduced to hundreds of milliwatts; the other is an ideal-diode controller plus MOS, which combines reverse-polarity protection with low loss and low quiescent consumption. Which one to choose depends on the load current and the allowable drop budget: for small current, the series Schottky is the lowest-cost option; for high current, prefer the MOS solution.

Contact Us ​

For assistance with rectifier and freewheeling device selection — point recommendations across Schottky / fast recovery / ultrafast recovery, reverse-polarity protection scheme calculations, and sample and measurement data support, contact us.

Shenzhen Intek Technology Co., Ltd — electronic components distributor and system solutions provider Tel / WeChat: 136-3264-8484 Address: 4F, Building 2, Jingwei Center, No. 309 Pingan Avenue, Pinghu Street, Longgang District, Shenzhen, China Web: www.intek.vip

Disclaimer: This article is technical popular-science and industry-exchange content, provided for reference only, and does not constitute selection, procurement or other commercial decision advice. Device parameters mentioned (forward drop, reverse recovery time, voltage range, junction temperature, etc.) are category-level typical ranges; within the same category there is large variation by process, part number and test conditions, so actual selection is subject to the manufacturer latest datasheet. Loss, derating and thermal design figures are common engineering references, and specific projects are subject to the overall product specification and customer acceptance criteria.