VBMB165R42SFD
Product Specifications
VBMB165R42SFD
VBsemi — Discrete & Power Semiconductors
Product Overview
The VBMB165R42SFD from VBsemi is a discrete power semiconductor delivering efficient switching and power conversion. VBsemi specializes in discrete MOSFETs, power transistors, and protection devices at competitive cost for power supplies, motor control, and consumer electronics.
Superb Automation partners with leading brands including Qorvo, TI, ADI, XILINX, NVIDIA, Intel, ST, MACOM, Renesas, and Coilcraft to provide complete BOM and supply-chain solutions.
Key Features
- Low Rds(on) for minimal conduction losses in power switching
- Fast switching speed with low gate charge for high-frequency operation
- Avalanche-rated for robust operation under inductive load conditions
- High junction temperature rating for demanding environments
- Logic-level gate drive compatible for simplified design
- RoHS and AEC-Q101 qualified options available
Key Specifications
| Vds / Vce | V |
| Id / Ic | A |
| Rds(on) / Vce(sat) | mΩ |
| Qg | nC |
| Technology | — |
| Package | — |
| Configuration | — |
| Mounting | — |
Applications
Request Quote — In Stock, 100% New & Original, Full Warranty
VBMB165R42SFD: 650 V, 0.056 ohm and a 90 ns body diode in the isolated TO-220F
The VBMB165R42SFD is a 650 V N channel super junction MOSFET in the TO-220 Fullpak body, the isolated tab version of the three lead package. The product summary on page one gives two numbers: 650 V at the maximum junction temperature and 0.056 ohm at a 10 V gate. Everything else a buyer needs sits further down, and two of those rows matter more than the headline: a maximum on resistance measured at 13.5 A rather than at the 42 A continuous rating, and a body diode that recovers in 90 ns.
The sheet names server and telecom power supplies, switch mode supplies, power factor correction and lighting ballast as the target circuits. Those four have one thing in common: the switch works against a 380 V to 400 V rail that carries a surge when the load drops, and the part has to hold 650 V while it does. A 600 V device in the same socket leaves 200 V of margin on a 400 V bus, which is thin once a motor or a long cable adds ringing.
The Fullpak body changes the thermal story more than the electrical one. The tab is moulded, so the part bolts to a heatsink without an insulating washer, and the price is a junction to case path that is worse than the exposed tab parts of the same family. On a board where the switch runs at 2 W this does not matter. At 20 W it decides whether the part survives the first thermal cycle.
The parameter table in full
| Drain to source voltage at maximum junction temperature | 650 V |
| Gate to source voltage, continuous | plus or minus 30 V |
| On resistance at 10 V gate and 13.5 A, 25 C | 0.056 ohm typical |
| Gate threshold at 250 microamp | 2.5 V minimum, 4.5 V maximum |
| Continuous drain current, case at 25 C and 100 C | 42 A and 25 A |
| Pulsed drain current, repetitive | 125 A |
| Single pulse avalanche energy, 28.2 mH and 10 A start | 1600 mJ |
| Maximum power dissipation, case at 25 C | 210 W |
| Linear derating factor | 1.67 W per C |
| Junction to case and junction to ambient | 0.57 and 62 C per watt |
| Operating junction and storage range | minus 55 to plus 150 C |
| Drain source voltage slope, junction at 125 C | 50 V per ns |
| Reverse diode voltage slope | 15 V per ns |
| Input, output and reverse transfer capacitance | 4500, 80 and 4 pF |
| Effective output capacitance, energy and time related | 63 and 213 pF |
| Total, gate source and gate drain charge | 98, 15 and 19 nC |
| Turn on delay and rise time at 520 V and 8 A | 18 and 24 ns typical, 25 and 55 ns maximum |
| Turn off delay and fall time, same conditions | 80 and 12 ns typical |
| Gate input resistance | 0.8 ohm |
| Continuous and pulsed source drain diode current | 42 and 125 A |
| Body diode forward drop at 8 A | 1.5 V maximum |
| Body diode recovery time, charge and peak current | 90 ns, 5.8 microcoulomb and 30 A |
| Forward transconductance at 13.6 A | 5.6 S |
| Peak soldering temperature for 10 s | 260 C |
Two thermal rows that decide the real current
The 210 W dissipation row assumes a case held at 25 C. The row that binds on a real board is 62 C per watt from junction to ambient, so with a 150 C junction limit and a 25 C room the part may shed 2.02 W. Through 0.056 ohm that is 6.0 A, and at a 50 C ambient the same arithmetic gives 1.61 W and 5.37 A.
Run the numbers from the other end and the gap widens. The 42 A continuous rating dissipates 98.8 W in the channel at 25 C, which needs a case below 94 C through the 0.57 C per watt junction to case path. That is a heatsink the size of the package, not a copper pour. A customer who reads 42 A on page one and puts the part on a 20 by 20 mm pad has sized the switch for the wrong thermal boundary.
The derating factor closes the argument. 1.67 W per C is simply 210 W divided by the 125 C allowed rise, so any loss above 2 W in still air has to leave through the tab. When a request arrives with a 4 W loss on this part and no heatsink, the honest answer is a different body or a different part, not a copper change on the board.
What the switching rows say about a hard commutated stage
Gate charge is 98 nC typical at a 10 V gate and a 520 V drain, which is high for a 0.056 ohm super junction device. At 100 kHz the drive costs 0.098 W, small against the 2 W the board can remove, so the gate is not the limit. The internal gate resistance of 0.8 ohm is low enough that an external 4.7 ohm resistor sets the turn on shape, and the 18 ns turn on delay with a 24 ns rise says the die switches fast enough that the loop inductance of the board, not the part, decides the ringing.
The diode row is where this part differs from its family. 90 ns of recovery, 5.8 microcoulomb of charge and 30 A of peak reverse current at 8 A forward and 400 V describe a diode that a hard switched bridge or a continuous conduction PFC can tolerate. The 600 V members of the same family print 475 ns and 5.8 microcoulomb, and the 800 V members print 325 ns. Anyone choosing between them for a 400 V bus reads the recovery row first and the on resistance second, because in continuous conduction the diode loss is often larger than the conduction loss.
Peak reverse current of 30 A against a 42 A forward rating is not a soft recovery diode, and a phase node with 15 V per ns of allowed diode slope will ring against any board that puts the return path under the die. The 50 V per ns forward slope belongs to the same story: the part tolerates a fast rail, and the board has to keep the overshoot inside 650 V.
Where this part still earns its place
A 3 kW server supply at 65 kHz, a telecom rectifier, an HID ballast and a continuous conduction PFC all sit inside the envelope. The isolated tab is the practical advantage: on a shared heatsink with several switches, the Fullpak removes the insulating pad and the associated 1 to 2 C per watt, and it removes the risk of a pad pierced by a burr. For equipment that has to pass a hipot test on the heatsink, that is worth more than the 15 percent on resistance the exposed tab version would give.
Against the 600 V VBMB16R20SFD in the same isolated body, this part trades 0.175 ohm and 475 ns of recovery for 0.056 ohm and 90 ns, and it adds 50 V of headroom. Against the 800 V VBMB18R15S it gives up 150 V of margin and keeps a diode two generations faster. The comparison has to be made on the diode and the thermal pair, not on the current column, because all three sheets print 20 A or more at a 25 C case.
Where a customer is replacing an older trench part of the same voltage, the useful gain is the recovery and the gate charge together. Our power management cross reference work keeps those two figures on file for every alternate we approve, so a proposed replacement is checked against the diode and the drive rather than against the summary page.
Questions we answer before a reel is booked
First, the avalanche row. 1600 mJ is printed against a test of 28.2 mH and a 10 A start, but half of 28.2 mH times 100 A squared is 1410 mJ, so the printed energy is larger than the energy the stated test can store. Our reading is that the row came from a different inductor, and we do not use it to justify a snubberless flyback. Ask for the test report before you rely on it.
Second, the gate. A threshold window of 2.5 V to 4.5 V means this is not a logic level device. A 5 V driver runs it in the linear region, and the threshold climbs as the die cools, so a cold start is the worst case. Drive it from 10 V, and check the 0.8 ohm internal resistance against the external resistor so the switching times match the sheet.
Third, the second source. Super junction 650 V parts at 0.056 ohm come from several suppliers in the same three lead body with different gate charge, different diode recovery and different threshold. We hold those three on file per approved alternate, and the incoming inspection plan checks the incoming lot against them. The pin to pin comparison starts from the isolated tab and the lead spacing, since a Fullpak will not drop into a footprint cut for an exposed tab part.