VBA3106N
Product Specifications
VBA3106N
VBsemi — Discrete & Power Semiconductors
Product Overview
The VBA3106N 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
VBA3106N: a dual 100 V N channel in the SO-8 body
The VBA3106N is a dual N channel MOSFET in the SO-8 body, printed at 100 V with 51 milliohm at a 10 V gate and 69 milliohm at 4.5 V per channel, 6.8 A and 5.8 A per channel, 9 nC of gate charge, 600 pF of input capacitance and a threshold window of 1 to 2 V. The data sheet names high frequency boost converter and LED backlight for an LCD television. Two hundred volt class work in an eight pin plastic body is what the part is for: two switches in one package that share a thermal path and a footprint.
The SO-8 body is the old standard and it is still bought because it is cheap to place and easy to inspect. The thermal path is through the lead frame to the pins, not through a pad, so the junction to ambient resistance is 37 K per watt typical on a one inch square board, which is better than the SC-70 and worse than a DFN with a pad. On a four layer board with copper under the body the number improves, and on a two layer board it does not.
The two channels are independent and each has its own gate and source, which makes the part usable as a half bridge with the sources tied externally, or as two separate switches in a boost and a backlight rail. The 100 V rating on both channels means the same code can be used on a 48 V rail and on a 24 V rail without a second approval.
Key data for the VBA3106N
| Configuration | Dual N channel, independent |
| Package | SO-8, eight lead |
| VDS | 100 V |
| Vth | 1 to 2 V at 250 microamp |
| RDS(on) at 10 V | 51 milliohm per channel at 6.8 A |
| RDS(on) at 4.5 V | 69 milliohm per channel at 5.8 A |
| ID at a 25 C case | 6.8 A and 5.8 A per channel |
| Pulsed drain current | 20 A |
| Ciss | 600 pF per channel |
| Gate charge | 9 nC typical per channel |
| Body diode | 0.8 V typical, 1.2 V maximum |
| RthJA | 37 K per watt typical, 50 maximum at 10 s |
| RthJF | 17 K per watt typical, 21 maximum |
| Applications | High frequency boost converter, LED backlight for LCD television |
The 600 pF input capacitance against 9 nC of gate charge is a low ratio compared with the DFN parts in this bank, which tells the buyer the die is small and the gate is thin. A thin gate means a lower threshold, and the 1 to 2 V window confirms it. It also means the gate is more exposed to transients, and the plus or minus 20 V window on the gate is the protection that matters on a 48 V rail where the source can move quickly.
The junction to foot resistance of 17 K per watt typical is much better than the 37 K per watt junction to ambient, which is the signature of an SO-8: the heat leaves through the drain pins into the copper, and the plastic top does nothing. A board with wide drain traces on both sides of the body runs close to 17 K per watt, and a board with thin traces runs at 37 or worse. On a dual part both channels feed the same lead frame, so the sum of the two losses goes through the same pins.
Where a dual 100 V part in an SO-8 is used
The first case is a two stage boost for an LED backlight in a television. One channel is the main switch and the other is the disconnect switch that isolates the LED string when the panel is off. The 100 V rating covers the boost output at 60 to 80 V with margin, and the 51 milliohm at 10 V keeps the conduction loss under 0.5 W at 3 A.
The second case is a half bridge in a 48 V to 12 V converter at 100 to 200 W. Both channels are in one body, the switch node is between them, and the gate drive comes from a controller with a bootstrap for the high side. Here the 100 V rating and the low gate charge matter, and the 20 A pulsed rating sets the current limit for the short circuit protection.
The third case is two separate load switches on a 24 V industrial rail. Each channel feeds a subsystem, the gates are controlled independently, and the body diodes block a reverse feed between the subsystems. The part is chosen because one SO-8 replaces two discrete switches and the assembly cost falls with it.
Drive, dead time and the SO-8 thermal path
At 9 nC of gate charge the drive current at 200 kHz is 1.8 mA per channel, so a controller output drives both without a buffer. In a half bridge the high side gate needs a bootstrap capacitor of at least ten times the gate charge, so 100 nF is the minimum and 1 microfarad is the practical value for a 100 V part where the switch node moves fast.
The body diode recovery of an N channel is fast, and on this part the drop is 0.8 V typical. In a synchronous converter the dead time should be as short as the comparator can blank, because every nanosecond of dead time is a nanosecond of diode conduction at 0.8 V. On a 3 A load that is 2.4 W per microsecond of dead time, which is why a 50 ns dead time matters and a 500 ns one does not work.
Where the same function is bought from two vendors, the vendor consolidation page covers how the codes are grouped and how a second source is kept qualified without carrying two full approvals. The SO-8 outline is the most widely second sourced body in the industry, so the practical difference between vendors is the gate charge and the threshold rather than the footprint.
What to check before this code is bought
The threshold window of 1 to 2 V is low enough that a 4.5 V gate drive is marginal on a part at the top of the window. On a 48 V rail the drive is usually 10 V, which removes the question. On a 12 V rail driven from a logic output, the 69 milliohm column is optimistic for a unit with a 2 V threshold.
The 20 A pulsed rating is modest for a part with two 6.8 A channels, and a half bridge that sees a shorted output will exceed it. The current limit has to act in a few microseconds, and the gate should be slowed with a resistor so that the part survives the interval between the fault and the limit. There is no safe operating area curve printed in the rows we read, so this has to be verified on the bench.
Where the code is compared with other 100 V duals in the SO-8 body, the power management cross reference page sets out which parameters have to match before a substitute is proposed, from threshold and gate charge to the body diode. On this body the pin out is standard, so the swap fits, and the gate charge is what changes the drive loss and the dead time.
The plastic body of an SO-8 absorbs moisture, and a part that has been stored open takes up enough water to crack during reflow. The moisture level and the floor life are not printed on the sheet we read, so a bake before a second reflow is the safe rule rather than an optional one.
Questions we get on this code
Is 6.8 A the current for the whole package? No, it is per channel and it is a 25 C case figure. With both channels loaded the lead frame carries the sum, and the practical continuous current is 3 to 4 A per channel on a good board.
Can the two channels be paralleled for 13 A? They can, and the resistance falls to 25 milliohm, but the threshold window of 1 to 2 V means the sharing is uneven until both are fully enhanced. A small source resistor on each channel is the usual fix.
Is 100 V enough for a 48 V rail? With a clamp, yes. The switch node of a boost or a half bridge can ring to twice the rail with a poor board, so the 100 V rating should be read against the peak at the node rather than against the rail.
Does it need a heatsink? No, but it needs copper. The 17 K per watt junction to foot figure assumes wide drain traces on both sides of the body, and a board with 0.5 mm traces runs at 37 K per watt or worse.