Loading…
Loading…
TO-263-5 · XL4015
XLSEMI · TO-263-5 · 5A Buck Converter
Price on Request
Get a competitive quote within 24 hours. Volume discounts available. Shenzhen stock ships in 1-2 business days.
The XL4015 is XLSEMI's 5 A, 180 kHz, 36 V step-down converter and the silicon inside the popular blue 5 A buck modules used for CNC controllers, 3D printers, LED drivers and automotive accessories. It is a fixed-frequency PWM buck with an internal P-channel power MOSFET and internal frequency compensation, so the external bill of material is an inductor, a Schottky catch diode, an input capacitor, an output capacitor, the feedback divider and a 1 µF bypass on the VC pin tied to VIN. Input range is 8 V to 36 V operating with a 40 V absolute maximum, output is adjustable from 1.25 V to 32 V against a 1.25 V feedback threshold, maximum duty cycle is 100% and minimum dropout is only 0.3 V, which means the part can hold a 12 V output from a 12.5 V battery — something fixed-frequency bucks with a minimum off-time cannot do.
The datasheet numbers are unambiguous and worth quoting to purchasing teams who confuse this part with the 3 A LM2596: continuous output current is 5 A, the internal switch current limit is 7 A, the P-channel MOSFET has 60 mΩ typical and 80 mΩ maximum on-resistance, quiescent current is 2.1 mA typical and 5 mA maximum, and the oscillator runs at 180 kHz nominal with a 144 kHz to 216 kHz spread. Efficiency is specified at 87% typical for 12 V in to 5 V out at 5 A and 93% typical for 24 V in to 12 V out at 5 A, with up to 96% claimed at lighter loads and higher step-down ratios. When the output is shorted the switching frequency folds back from 180 kHz to 48 kHz, which limits dissipation during a fault, and thermal shutdown and current limit are built in. Junction range is -40 °C to +125 °C with a 30 °C/W junction-to-ambient figure for the TO263-5L package on a proper copper area.
Thermal design is the honest limitation. At 24 V to 12 V and 5 A the converter passes 60 W and, at 93% efficiency, dissipates about 4.2 W; with 30 °C/W that is a 126 °C rise, which exceeds the junction limit unless the tab is bolted to a heatsink or poured copper with vias and there is airflow. Practical experience with 5 A modules matches the arithmetic: they will deliver 5 A only with a heatsink, and 3 A to 4 A continuously on a bare board. Plan for the worst case at your maximum ambient, derate the inductor to at least 6 A saturation, and keep the switch-node loop (input capacitor, IC, diode) tight to control EMI at 180 kHz with 5 A of chopped current.
Commercially, the XL4015E1 ships in TO263-5L with 800 units per tape-and-reel and the E1 suffix denoting Pb-free RoHS compliance. It is one of the most cloned power ICs in Shenzhen because module demand is huge, and clones usually fall into three categories: smaller die that cannot sustain 5 A (they thermal-shutdown or fold back at 2 A to 3 A), parts with a switch limit well below 7 A, and devices whose oscillator sits far from 180 kHz, which detunes the reference magnetics. Incoming inspection should measure the switching frequency, load to 5 A for fifteen minutes while logging case temperature and output regulation, and verify the feedback threshold at 1.225 V to 1.275 V. Our MOQ is 100 pcs loose or 800 pcs on tape-and-reel, shipping from Shenzhen in one to two business days; authorised XLSEMI factory lead time is six to ten weeks. Assembled modules with screw terminals, a heatsink and an adjustable divider are also available in packs of 10.
Order XL4015E1 with the matching 6 A-rated inductor, a 60 V Schottky and a heatsink or defined copper area on the same BOM line — the chip itself is rarely the failure point, and specifying the magnetics and thermal path up front avoids the 3 A surprise at first-article inspection.
Both are 5-pin TO-263/TO-220 bucks with similar part counts, but the XL4015 is rated 5 A continuous with a 7 A switch limit and 60 mΩ internal MOSFET, while the LM2596 is rated 3 A with a 4.5 A typical peak limit. The LM2596 switches at 150 kHz with TI-grade documentation, a -40 °C to +125 °C junction rating and a 40 V input; the XL4015 switches at 180 kHz with a 36 V input and far better current capability per dollar. For a 3 A rail with tight documentation requirements choose the LM2596; for 4 A to 5 A on a cost-sensitive board choose the XL4015 and budget heatsinking.
No. At 24 V to 12 V and 5 A the device dissipates about 4.2 W at 93% efficiency, and with 30 °C/W junction-to-ambient that is a 126 °C rise — beyond the 125 °C junction limit before you count ambient. On a well-poured 2 oz copper area with vias you might reach 3 A to 4 A continuously; for a genuine 5 A you need the tab on a heatsink or forced airflow, which is exactly why the commercial modules ship with a bolted aluminium block. Derate for your maximum ambient, not the bench.
Inductor: 33 µH to 47 µH for typical 12 V to 5 V conversions, rated for at least 6 A saturation with low DCR and a ferrite or iron-powder core; undersized magnetics saturate at peak current, collapse inductance and push the device into its 7 A limit. Diode: a Schottky rated at 1.2 times the input voltage and 1.2 times the load current — for a 36 V rail use a 60 V part such as MBR560, 60SQ045 or SS56; never a slow rectifier, whose reverse recovery will spike the switch node and cause instability. Output capacitor: 470 µF to 1,000 µF low-ESR aluminium plus ceramics for high-frequency bypass.
The output ceiling is set by duty cycle, not just the divider: with a 36 V input the practical maximum is about 32 V, and near the ceiling the internal switch and diode drops eat the margin, so a 34 V target will not regulate. Audible whining usually means the converter is in foldback or hiccup — check for an output short, an overloaded rail, an inductor saturating at peak current, or an input supply that current-limits during startup. The datasheet notes the frequency drops to 48 kHz during a short, which is audible in the magnetics; measure with a scope on the switch node before changing the divider.
Yes. VC is the internal voltage-regulator bypass and the datasheet specifies 1 µF from VC to VIN in a typical application. It decouples the internal gate-drive and bias supply; omitting it or routing it far from the pin shows up as erratic duty cycle, extra switch-node ringing and a converter that will not start reliably at high input voltage. Place the 1 µF ceramic within a few millimetres of pin 4, together with the input bulk capacitor (220 µF/50 V class) that the datasheet assumes for its specified measurements.
Three measurements catch almost everything. First, switching frequency on the switch node: genuine parts sit within 144 kHz to 216 kHz; clones are often 40% low. Second, current capability: load to 5 A for fifteen minutes while logging output voltage and case temperature — remarked or smaller die fold back below 3 A or exceed the temperature rise predicted from the 60 mΩ to 80 mΩ switch resistance. Third, feedback threshold: it must regulate to 1.225 V to 1.275 V, since the output divider ratio depends on it. Physically, look for laser-etched XL4015E1 marking, a firmly bonded tab and matte tin leads; re-tinned pulls look shiny and show solder wicking.
MOQ is 100 pcs loose for samples or 800 pcs on tape-and-reel (XL4015E1). Shenzhen stock ships in one to two business days and authorised XLSEMI factory lead time is six to ten weeks. Assembled 5 A buck modules are stocked in packs of 10, which is often faster than building the first prototypes.
Submit an RFQ — we reply within 24 hours with stock, pricing, lead time and packaging options.
Request a Quote →