Zener noise / oscillations

On a sunny day (Mon, 23 Jul 2012 01:16:24 -0700) it happened Robert Baer wrote in :

I do not normally use zeners for reference, just for over voltage protection. For reference you probably want bandgap based chips. There are some very low power reference chips.

I'm not so sure about "quiet" but I suppose everything is relative. There's a lot of nice things about a 6.2V zener at the sweet spot current, that are not true for higher voltage zeners, lower voltage zeners, or zeners run away from their sweet spot.

At the low microamps, YOU ARE ON THE WRONG SIDE OF THE KNEE, you are on the high dynamic impedance side.

And remember that gas regulator tubes make excellent sawtooth oscillators blipping from one side of the knee to the other.

At the microamp range everyone else is just using bandgap refs.

Tim.

Like zeners, they also make excellent noise generators, usually with the help of a magnetic field.

Tim

Bandgap ones are more convenient. But the very best semiconductor references in terms of noise and stability are those based on zeners.

Buried zeners, specifically.

On a sunny day (Mon, 23 Jul 2012 14:53:16 +0100) it happened John Devereux wrote in :

At 2-50uA?

Some of my scribbles on the subject:

There was a nice - and tolerably long - thread on the subject "Zener Diode Oscillation" - here back in 1997. Winfield Hill and the late Tony Williams posted a lot of interesting measurements and some interesting references from the semiconductor physics literature.

Not as far as I know - usually about 5mA and ~6-7V. Like I said, not very convenient.

On a sunny day (Mon, 23 Jul 2012 18:12:40 +0100) it happened John Devereux wrote in :

protection.

Yes, I remember those (Intermetal IIRC) had a zener and Ge transistor that used together (zener in emitter) had exactly a zero Tc. Indeed about 5 or 6V, do not remember the exact value. For the rest the Tc is never really flat for zeners AFAIK...

current,

That is not noise, you have inadvertently built a negative resistance oscillator:

er

What sort of voltage range are you talking about. My limited experience would suggest that at lower voltage (8-16 V) the low current noise is very variable piece to piece.

At higher voltage the low current noise is larger and more consistent

And above 20 (something) Volts there maybe a point in the I-V curve that can lead to oscillations.

Though all of this depends on the power rating of the diode too.

George H.

to

gives

gher current,

es

illator:

Woah, a 4kV zener! You don't use those every day. (at least I don't)

George H.

ear

to 500uA)

subject to

ndumb

ude and

mp;quot; as that gives

oise. Then at higher current,

itude.

, one sees

do not or

e oscillator:

You might note they extended the observations to the low voltage diodes too= .

protection.

I've done bandgaps on CMOS processes... 100uA total operating current. My bag of tricks includes producing other voltages than the standard

near

A to 500uA)

d subject to

randumb

itude and

noise. Then at higher current,

plitude.

er, one sees

y do not or

nce oscillator:

oo.

Oh,I don't have access to the whole paper. I was looking at some data sheets (a while back) and zeners in the

(I guess the last thing the OP wants is an oscillator.)

George H.

You can heat treat zener diodes (by passing a lot of current through them) = for use in saw-tooth oscillator circuits. Obviously if you that you are cau= sing poorly controlled physical changes of the semiconductor properties. Yo= u know a TL431 shunt regulator costs so little these days that you should u= se it in your circuits in preference to a zener diode. It is so much better= and 1960's style voltage regulation is not such a great idea these days.

Reference? Said nothing about that; app is for voltage regulation but zener pass current must be low.

My app is only for voltage regulation, not reference purposes; zener current to range from 2-500uA. Gotta be cheeeep 1-3 cents.