The Cyber-Spy.Com Usenet Archive Feeds Directly
From The Open And Publicly Available Newsgroup
This Group And Thousands Of Others Are Available
On Most IS NNTP News Servers On Port 119.
Cyber-Spy.Com Is NOT Responsible For Any Topic,
Opinions Or Content Posted To This Or Any Other
Newsgroup. This Web Archive Of The Newsgroup And
Posts Are For Informational Purposes Only.
From: SNIPrf_man_frTHIS@yahoo.com (Frank Raffaeli)
Subject: Re: How does a mixer work?
Date: 28 Oct 2002 20:37:56 -0800
References: <firstname.lastname@example.org> <email@example.com> <firstname.lastname@example.org>
NNTP-Posting-Date: 29 Oct 2002 04:37:56 GMT
"Kevin Aylward" wrote in message news:...
> "Frank Raffaeli" wrote in message
> > Jim Thompson wrote in message
> > > On 17 Oct 2002 08:05:59 -0700,
> > > SNIPrf_man_frTHIS@yahoo.com (Frank Raffaeli),
> > > In Newsgroup: sci.electronics.design,
> > > Article: <email@example.com>,
> > > Entitled: "Re: How does a mixer work?",
> > > Wrote the following:
> > >
> > > |firstname.lastname@example.org (Asa Cannell) wrote in message
> > > |> How does a mixer work? I don't understand why it needs a
> > > |> element. I think examples of why a linear element (resistor) wont
> > > |> would help illustrate why a non-linear element (diode) will.
> > > |
> > > |It doesn't need a non-linear element. Non-linear elements, like
> > > |or over-driven transistors can provide a simple way to derive a
> > > |product component of two signals. For a nearly linear approach,
> > > |up "Gilbert cell mixer". This uses linear elements and current
> > > |steering / sharing. An AC current with a DC bias is split between a
> > > |differential pair and steered to one side or the other by altering
> > > |bias on the bases / gates of the pair.
> > >
> > > NAAAAAH! The "elements" ARE NOT LINEAR!
> > Hmm ... ok, that's true. No transistor is perfectly linear, even when
> > you try to operate in a linear range. I guess my point was that a
> > Gilbert cell multiplier works very well when operated in the linear
> > range of the devices, and the operation of the (Gilbert) multiplier is
> > not dependent upon the inherent non-linearity of the transistors.
> Yes it is!.
> A Gilbert multiplier fundamentally works *because* of the *exponential*
> behaviour of a transistor. Its re is determined by its collector
> current. If re was a constant, indepandant of collector current, it
> could not multiple at all. The multiplication is due to
> re is a function of it own current, therefore it is non-linear, by
> Kevin Aylward
> SuperSpice, a very affordable Mixed-Mode
> Windows Simulator with Schematic Capture,
> Waveform Display, FFT's and Filter Design.
I'm not sure why I forgot this. I was thinking about how a Gilbert
cell still works with emitter degeneration. Adding emitter
degeneration resistors allows the cell to handle larger amplitudes,
but the current steering still can't happen without the nonlinear
effects of the transistor, as you pointed out so clearly and
Frank W. Raffaeli
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup