Reply-To: "Kevin Aylward"
From: "Kevin Aylward"
References: <3DF6537E.B0FC9DBE@worldonline.fr> <3DF90203.C9691374@worldonline.fr> <3DF94AE1.firstname.lastname@example.org>
Subject: Re: Determining bjt noise parameters for Spice models?
X-Newsreader: Microsoft Outlook Express 6.00.2800.1106
X-Mimeole: Produced By Microsoft MimeOLE V6.00.2800.1106
X-Inktomi-Trace: public1-pete2-5-cust33.pete.broadband.ntl.com 1039907820 17430 184.108.40.206 (14 Dec 2002 23:17:00 GMT)
Date: Sat, 14 Dec 2002 23:16:54 -0000
NNTP-Posting-Date: Sat, 14 Dec 2002 23:17:00 GMT
Robert Baer wrote:
> I do not quite understand what you meant by "..(using) the standard
> 2SC3329 model..could not reach..much lower levels obtained.."
> Sort of sounds like you got better noise levels than the reference.
> And that may be correct; were your conditions identical (circuit,
> operating current, temperature)?
> The Rb on most bipolars is so low that it can be ignored in most
> cases; Rbb' (base spreading resistance) is in a similar class.
I disagree that rbb should be ignored. BC109s might be 400 ohms, 2n4401s
might be 12 ohms, others might vary from 10 to 500 ohms. For low
resistance sources, rbb *is* the *dominate* noise source in an optimally
designed amplifier. e.g. as I noted, shot noise might be only 0.14nv at
10 ma, verses a 50 ohm resistor at 0.89nv.
> The Re resulting from the biasing current is the major En
Only at low collector currents, which are only used for high source
resistances (well excepting low power, not noise, as a strong
> and this is where one gets very good correlation
> measured VS calculated (take the simple single transistor case).
> Now when Ie is "large" to make for low Re, then Rb and Rbb' must be
> added; most data sheets will not give you any information about these
Re is a bit confusing, this usually refers to an external emitter
resistor. re, imo, is usually clearer for the dynamic transistor emitter
> So, measure noise with low Ic and with high Ib (low source R); in
> both cases, the noise is very close to the vector sum of Rs, Re, Rb
> and Rbb'. However, at the lower current, Re is higher and the base
> resistances can be ignored enough to approximate (closely) their
> combined (effective) value.
> Armed with that info, one can calculate NF at any reasonable
> frequency (say 1Khz to near Ft) and reasonable current; there is
> excellent correlation in AF and RF measured values.
> Sorry about the long-winded mess above; hope is of some help.
But there is not much point in doing tests at low currents at all. *All*
transistors have the *same* white noise at any current, except for rbb
generated noise and the assumtion that typically hfes are 100-400. All
you want to detrmine is what rbb is.
e.g. noise due to collector current accross re is:
V1noise=re.sqrt(2qIc) = sqrt(2q/40^2IC)
noise due to base current accross Rs is:
A rough optimum being to set re=Rs/sqrt(hfe)
Noting from this that the noise variation due to spreads of hfe is
rather low, so hfe can usually be ignored.
It is the fact that transistors have a large spread in rbb' from device
to device that actually differentiates low noise devices from any othe
device. Its inherent that if one tries to chose a low noise device, then
one is selecting principly for *low* rbb. Nothing else matters much.
A low noise transistor is essentailly a transistor with low rbb, and low
1/f noise. End of story.
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.