Transistor to drive an ignition coil

I want to drive a 6-volt ignition coil with a DC resistance of 3 ohms, from a power supply of 6.2 volts DC at 2 amps. I propose to use a 555 timer as the PWM, with a 50% duty cycle at about 5 to 10 kHz, with a totem pole amplifier. I don't need full power output from the ignition coil. What would be a good power transistor for the output? (I would like to use a MOSFET, but don't have a grounded working mat or wrist strap.)

-
A MOSFETis usable using entirely reasonable care - antistatic equipment availability should not limit your choice. Ability to work well is the main concern. More later maybe. Rushing ... –  Russell McMahon Apr 14 '12 at 21:27

Olin is right about the BJT. For power switching a power MOSFET is often chosen because of it's low $R_{DS(ON)}$, therefore it's low voltage drop, therefore it's low dissipation. Your requirements are not that high, however. 2A at 50% duty cycle is 1A average, a 200mV saturation voltage will cause only 200mW dissipation. Also, while there are logic MOSFETs that are specified at a few volts $V_{GS}$ they usually can only deliver a few hundred mA at that voltage. Most of the time they'll need 10V or higher to get > 1A drain current.

A quick search for low saturation voltage transistors gets us a the OnSemi NSS40301, which has a maximum collector current of 3A continuous.

The graph shows that $V_{CE(SAT)}$ at $I_C$ = 2A is less than 200mV, so average power dissipation is less than 200mW, far less than the 2W maximum rating. $H_{FE}$ is minimum 200, then you'll need at least 10mA base current.

-

Even though the input voltage is only 6.2 V, there will be some kickback on the input side of the coil. The transistor needs to be able to handle significantly more than 6.2 V. A 60 V transistor is probably a good choice, especially if you clamp the input to a bit less than the transistor can handle.

You haven't given a reason for using a MOSFET other than you'd like to, which is no reason at all. A bipolar NPN will be easier to drive from this low voltage. There are plenty of medium power NPN transistors with respectable gain out there. At a gain of 40, for example, you need 50 mA base current.

-
Is this 60V necessary? I thought the kickback will go as high as twice the supply voltage: the supply voltage itself + the same voltage generated in the coil superposed. –  Federico Russo Apr 15 '12 at 8:36
@Federico: If there coil were just a single inductor, the kickback would go arbitrarily high. The coil is actually the primary of a flyback transformer, so the secondary is intended to catch most of the kickback. However, there will be some on the primary, and then any leakage inductance of the primary produces unbounded kickback, depending on how fast the transistor turns off. –  Olin Lathrop Apr 15 '12 at 13:33

Design Rule #1
Thou shalt not tell how to design a solution unless you define exactly what output you need.
"don't need full power output from the ignition coil"

Design Rule #2
Always specify inputs and outputs fully before suggesting any implementation.
We do need a full description of the output spec.

I gather you want some sort variable spark generator. Did you know an ignition coil has a turns ratio around 1k and you can drive it with a pulse switch with clamp diode to $V+$? Think $Vout= L \cfrac{di}{dt}$ and this is an autotransformer.

-