# RF amplifier stability question

Started by April 28, 2020
```The standard treatment of amplifier stability is based
on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
there an analytical way to estimate this ?
```
```On 2020-04-28 00:43, amal banerjee wrote:
> The standard treatment of amplifier stability is based
> on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
> there an analytical way to estimate this ?

S parameters work for amplifiers too.

C oscillation?

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

```
```On Tuesday, April 28, 2020 at 12:43:20 AM UTC-4, amal banerjee wrote:
> The standard treatment of amplifier stability is based
> on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
> there an analytical way to estimate this ?

If you just need an amplifier go buy a stable amp from minicircuits and don't worry about it.  If you are designing an amplifier where you need to know this stuff you are going to have to dive pretty deep.

What is negative resistance....go ponder that for a month or two
```
```On Tuesday, April 28, 2020 at 9:21:53 PM UTC+5:30, blo...@columbus.rr.com wrote:
> On Tuesday, April 28, 2020 at 12:43:20 AM UTC-4, amal banerjee wrote:
> > The standard treatment of amplifier stability is based
> > on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
> > there an analytical way to estimate this ?
>
> If you just need an amplifier go buy a stable amp from minicircuits and don't worry about it.  If you are designing an amplifier where you need to know this stuff you are going to have to dive pretty deep.
>
> What is negative resistance....go ponder that for a month or two

There is nothing profound about negative resistance. It
occurs for some semiconductor devices(e.g., Gunn diode)
when the forward current decreases with increasing forward
voltage, and physically means that the device is pumping
current into the circuit to which it is attached.

I am afraid you have not answered my question-does steady state behavior of an RF amplifier need large signal S
parameters ?
```
```On Tuesday, April 28, 2020 at 8:28:34 PM UTC+5:30, Phil Hobbs wrote:
> On 2020-04-28 00:43, amal banerjee wrote:
> > The standard treatment of amplifier stability is based
> > on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
> > there an analytical way to estimate this ?
>
> S parameters work for amplifiers too.
>
> Or are you asking about large-signal instability, e.g. snivets or Class
> C oscillation?
>
> Cheers
>
> Phil Hobbs
>
> --
> Dr Philip C D Hobbs
> Principal Consultant
> ElectroOptical Innovations LLC / Hobbs ElectroOptics
> Optics, Electro-optics, Photonics, Analog Electronics
> Briarcliff Manor NY 10510
>
> http://electrooptical.net
> http://hobbs-eo.com

behavior of an RF amplifier needs large signal S parameters. The S parameters supplied by transistor
manufacturers are small signal S parameters.
```
```On Wednesday, April 29, 2020 at 6:54:40 AM UTC-4, amal banerjee wrote:
> On Tuesday, April 28, 2020 at 8:28:34 PM UTC+5:30, Phil Hobbs wrote:
> > On 2020-04-28 00:43, amal banerjee wrote:
> > > The standard treatment of amplifier stability is based
> > > on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
> > > there an analytical way to estimate this ?
> >
> > S parameters work for amplifiers too.
> >
> > Or are you asking about large-signal instability, e.g. snivets or Class
> > C oscillation?
> >
> > Cheers
> >
> > Phil Hobbs
> >
> > --
> > Dr Philip C D Hobbs
> > Principal Consultant
> > ElectroOptical Innovations LLC / Hobbs ElectroOptics
> > Optics, Electro-optics, Photonics, Analog Electronics
> > Briarcliff Manor NY 10510
> >
> > http://electrooptical.net
> > http://hobbs-eo.com
>
> behavior of an RF amplifier needs large signal S parameters. The S parameters supplied by transistor
> manufacturers are small signal S parameters.

I think you are likely answering your own question.  You probably are not being provided the parameters you want.  What frequency? What power? How much into compression?  Class A or B or C or AB?

Not to be a pain, but how can anyone possibly answer your question when you are so vague?  My experience though is to go play with the part and figure it out on the bench unless this is some super laser kilo buck device.
```
```On 2020-04-29 06:54, amal banerjee wrote:
> On Tuesday, April 28, 2020 at 8:28:34 PM UTC+5:30, Phil Hobbs wrote:
>> On 2020-04-28 00:43, amal banerjee wrote:
>>> The standard treatment of amplifier stability is based
>>> on small signal S parameters, and involves conditional stability, unconditional stability, Rollet's criterion,
>>> there an analytical way to estimate this ?
>>
>> S parameters work for amplifiers too.
>>
>> Or are you asking about large-signal instability, e.g. snivets or Class
>> C oscillation?

> behavior of an RF amplifier needs large signal S parameters. The S parameters supplied by transistor
> manufacturers are small signal S parameters.

In general neither is sufficient.  The bias conditions change
continuously in large-signal operation, so that it's quite possible for
an amplifier to break into spontaneous oscillations someplace on the
large-signal waveform.  (That's called a snivet.)

Real RF guys like Gerhard have much fancier software for that stuff.
(Joerg's a real RF guy too, but AFAICT generally takes a lower-tech
approach.)

Cheers

Phil Hobbs

--
Dr Philip C D Hobbs
Principal Consultant
ElectroOptical Innovations LLC / Hobbs ElectroOptics
Optics, Electro-optics, Photonics, Analog Electronics
Briarcliff Manor NY 10510

http://electrooptical.net
http://hobbs-eo.com

```