VBP16R90SE
Product Specifications
VBP16R90SE
VBsemi — Discrete & Power Semiconductors
Product Overview
The VBP16R90SE 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
VBP16R90SE: 600 V and 90 A in a TO247 with 18 milliohm
The VBP16R90SE is a single N channel MOSFET in a TO247 body, rated at 600 V drain to source, plus or minus 30 V on the gate, a 3.5 V threshold, 90 A of drain current and 18 milliohm at a 10 V gate. The vendor table lists a superjunction deep trench die. It shares its voltage, current and package with the 24 milliohm part and differs in the resistance and in the column structure.
Eighteen milliohm at 600 V is the lowest resistance in this band, and the difference against 24 milliohm is 25 percent. At 90 A the loss falls from 194 W to 146 W, at 68 A from 111 W to 83 W, and at 45 A from 49 W to 36 W. That is the difference between a marginal heatsink and a comfortable one at 40 A.
The SE suffix marks the deep trench column structure. In this die class it is used to reach a lower resistance without growing the area, because the trench columns allow a tighter cell pitch than a multiple epitaxy stack. The trade is a sharper output capacitance curve, which shows up as a higher displacement current at turn on.
Key data for the VBP16R90SE
| Configuration | Single N channel |
| Package | TO247, through hole, tab on the drain |
| VDS | 600 V |
| VGS | Plus or minus 30 V |
| Vth | 3.5 V typical |
| ID | 90 A |
| RDS(on) at 10 V | 18 milliohm |
| Structure | Superjunction, deep trench |
| Loss at 22 A | 8.7 W |
| Loss at 45 A | 36.4 W |
| Loss at 68 A | 83.2 W |
| Loss at 90 A | 145.8 W |
Thermal headroom that the lower resistance buys
At 45 A the hot loss at 18 milliohm is 73 W, against 97 W at 24 milliohm. On a 0.5 K per watt heatsink at 45 C ambient the junction sits at 118 C against 136 C, so the lower resistance part holds 18 C more margin at the same current.
That margin is what allows a 5 kW stage to run on one part rather than two. At 60 A the hot loss is 130 W and the junction is at 157 C, which is outside, so the ceiling for one part is 45 to 50 A with a good fan and 30 A on a plate.
Where the heatsink is fixed by the enclosure, the lower resistance part is the cheapest way to gain thermal margin, because the alternative is a bigger heatsink with more volume and more cost.
Switching behaviour of a deep trench die
A deep trench superjunction die has a lower output capacitance at high drain voltage and a sharper transition around 50 V. In a hard switched bridge the energy stored in that capacitance is recovered into the opposite switch at turn on, and at 400 V and a 2 nF equivalent the stored energy is 0.16 mJ per transition, which is 16 W at 100 kHz.
That number is why the switching frequency is the first decision on a board that uses this part. At 40 kHz the capacitance loss is 6 W, at 100 kHz it is 16 W, and at 200 kHz it is 32 W. The resistance saving against the 24 milliohm part is 20 W at 45 A, so at 200 kHz the lower resistance is entirely consumed by the capacitance.
The gate charge also rises with die area, and a 90 A die needs 2 to 3 A of peak gate current for a fast transition. A weak driver turns the part into a linear resistor for 100 ns per edge, and that is 20 W of loss at 100 kHz on its own.
Where this part fits
Bridgeless power factor stages from 3 to 5 kW, EV onboard chargers at 6.6 and 11 kW in a two part bank, server power supplies at 3 kW with a 400 V bus, and high current battery test equipment. In a charger the part runs at 60 to 100 kHz and the capacitance loss is part of the budget.
In a welding inverter at 20 kHz the capacitance loss is negligible and the resistance is everything, so this is the right part for that application at 60 to 80 A peak.
For a three phase 400 V input the class is 900 V, and this part is not usable. Where a product family spans both line voltages, the supplier corrective action page covers how a failure on one voltage class is closed out at the source before the other class ships.
Where the part is used in a two part bank, the sharing between the two is set by the source connection rather than by the die. Two TO247 parts screwed to the same heatsink with separate source leads to a common bus share within 10 percent at 40 A, because the positive temperature coefficient of the die pushes current away from the hotter part. A shared source trace with unequal length can hold a 25 percent split, and the hotter part then runs 30 C above the other. The BOM consolidation page covers how a bank like that is treated on one bill of materials.
The sourcing side of this part is covered on the supplier corrective action. For a deep trench line in a new stage, that page covers how a failure is closed out at the source.
Questions we get on this code
Is 18 milliohm at 25 C? The vendor table does not state the temperature. For a superjunction die at this voltage the hot figure is roughly 1.8 to 2 times the printed value, so size the heatsink at 33 to 36 milliohm.
Can it replace the 24 milliohm part in the same holes? Physically yes, and the loss falls. The gate charge rises, though, so the driver and the gate resistor have to be checked before the swap is signed off.
What is the body diode recovery? Not printed. At 45 A in a hard switched bridge it is significant, and a fast recovery grade is the safer choice where the bridge runs above 50 kHz.
Is a snubber needed? At 400 V and 45 A with a 20 ns transition, the loop inductance should be under 15 nH. Above that, a resistor capacitor snubber across the switch or a ferrite bead on the drain is the usual fix.