zenering a jfet gate

Depends on the current and repetitiveness. Ultimately causes electromigration. I don't know what degrades with a JFET (maybe the gate gets leaky), but a similarly-structured BJT loses beta. ...Jim Thompson

-- | James E.Thompson, CTO | mens | | Analog Innovations, Inc. | et | | Analog/Mixed-Signal ASIC's and Discrete Systems | manus | | Phoenix, Arizona 85048 Skype: Contacts Only | | | Voice:(480)460-2350 Fax: Available upon request | Brass Rat | | E-mail Icon at

Shouldn't it be a lot less damaging than to a BJT? There's no surface-of-the-silicon high field involved in the JFET, and the field direction is the same as normal bias (so none of the precautions against surface contamination are voided).

The special-purpose two-terminal JFETs used for current regulators are completely specified for normal operation in Zener (actually, avalanche) breakdown. As long as you don't overheat them, they just go back and forth across the knee indefinitely. The JFETs in an IC amplifier, of course, might have other issues.

In the old Tek vertical amplifier inputs, Tek used a protection diode to prevent zenering for negative input overloads.

I've never checked to see how/whether the device conducts or not in this mode.

Is this necessary? I'm sure most of us just copied that part, assuming that the gates couldn't take as much current as could be delivered by the input current limiting resistor. With today's presumably smaller structures, it would probably be worse unless the zener voltages were proportionately reduced.

HTH...

I have a design where it's barely possible that under some transient or startup condition I could zener a BF862 gate by a couple of hundred uA for a short period. It's a super-sensitive node so I'd prefer not to hang unnecessary diodes on it, or add the Johnson noise of a resistor. The BF862 is rated for -20 and seems to zener about -24. There's probably other knobs I can turn to keep the negative gate voltage from getting over 20, but I was curious if anyone had any experience, or if there was anything comparable to bipolar damage from zenering the base.

John

Isn't there some low-capacitance over-voltage zeners/transzorps somewhere? ST?

LM a écrit :

Yup. I've found some real nice ones. John, if you're interested email me. (can't go public on this)