JFET protection

[tempus]

Hey all;

Well, I'm doing yet another revision of my pedalboard. I use JFETs as switching elements to bring different effects in and out of my effects chain (a looper, I suppose), but I've done a little reorganizing, and I don't really need all the loops anymore, so I'm disconnecting some of the JFETs. Note that I could actually remove them, but in the (probably likely) event that I ever want some of the loops back in, I'd rather just leave them there. I don need to disconnect them from the switching and routing circuitry though to accommodate the changes. Do I need to ground any part of the JFET (i.e., the gate, drain, or source) to protect it from damage, or can they just be left floating?

Thanks

[R.G.]

JFETs are in general very - VERY - robust to static. I've never seen or heard of one killed by static. So no, they need no special protection IMHO.

[amptramp]

A JFET has a diode junction along a length of silicon.  There is no oxide layer that has to be protected from damage, so anything a silicon diode will survive, a JFET will survive.

[tempus]

Thanks for your replies. I must say I'm a little surprised - anytime I order a JFET from Mouser, they pack it in a static sensitive bag, which is what led me to ask the question. Come to think of it, diodes are as well.

Anyway, I'm glad I asked, because that just saved me a mess of work, not to mention having to remember that I did it months or years down the road when I reinvent my wheel.

So is it the oxide layer in a MOSFET or CMOS device that makes them so sensitive to static?

[R.G.]

Thanks for your replies. I must say I'm a little surprised - anytime I order a JFET from Mouser, they pack it in a static sensitive bag, which is what led me to ask the question. Come to think of it, diodes are as well.

Anyway, I'm glad I asked, because that just saved me a mess of work, not to mention having to remember that I did it months or years down the road when I reinvent my wheel.

Actually, being reverse-broken by voltage isn't all that good for any semiconductor junction. It's just that devices without MOS insulating layers tolerate it pretty well and the damage may well be infinitesimal. If the energy in the spark is small, it would be hard to detect, even with a barn full of special equipment, that it had ever been reverse broken. If the breakover energy is large, the breakover spot is heated, and may be damaged.  Low noise bipolar transistors do get noisier as their base-emitter is reverse broken over time. JFETs, I've never heard of damage from, but I'm sure it's possible.

So the industry took note that static wasn't really good for semiconductors and started packaging all of them in protective wrapping.

So is it the oxide layer in a MOSFET or CMOS device that makes them so sensitive to static?

The oxide layer is a **GOOD** insulator, but it's thin - usually a few NANO-meters. This is so thin that voltages of less than 100V - sometimes a lot less - can punch right through it, destroying its integrity and insulation, and effectively killing the device. One thing I read on gate oxide mentioned in passing something about a handfull of silicon atoms being the insulating layer in high performance logic MOSFET devices. So it's easy to damage it by punching through with even quite small energies, stuff that's too small to damage bipolar junctions.

Yes, it's the oxide layer. That's the smallest thing inside a MOSFET, the thickness of the gate oxide. The thinner the oxide, in general, the better it works - but the more fragile it gets.