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: "John Jardine"
Subject: Re: Limitations of the BC337
Date: Thu, 9 Jan 2003 21:19:39 -0000
NNTP-Posting-Date: 9 Jan 2003 21:05:47 GMT
X-Newsreader: Microsoft Outlook Express 5.00.2615.200
Paul Burridge wrote in message
> I can't seem to get a BC337 to provide any worthwhile voltage gain
> with any signal over about 200Khz. However, the quoted transition
> frequency for this device is 200Mhz. Anyone have any ideas as to what
> the problem might be? Surely I'm not expecting too much of the BC337
> at such a low frequency?
> "What is now proved was once only imagin'd"
> - William Blake, 1793
That 200kHz will be due to, too high a gain, capacitive loading effects,
high source resistance.
At a gain of say 10x, then the BC337 trannie is only capable of a flat
20MHz. A gain of 100 and the trannie can only do a flat 2MHz.
This assumes *no* capacitive loading across the collector resistor and that
the source or driving impedance is *very* low.
Assume your circuit has (say) a gain of x20.
This instantly sets an upper frequency limit (-3dB or 70% gain) of 10MHz.
If your design also has say a 1k collector load with 5pF of stray
capacitance across it, (quite normal) and another 5pf of 10:1 scope probe
capacitance hooked onto the collector, then this will cause a simple RC
low-pass filter, with yet another low-pass break frequency of 1/(2*pi*C*R)
this one is 16MHz.
That's 2 things we have (effects are both running together) that are now
killing the HF gain off.
Add a third! called 'Miller effect'.
This is yet another RC low-pass filter. The 'R' in this case is the
source-impedance driving the trannie. The 'C' is the trannie 'base-collector
capacitance' (plus strays) who's combined effect is stepped up by the
trannie gain (the "Miller effect" bit).
Say the actual Cbc+strays is 3pF and that your circuit is being driven from
a 50ohm sine generator. As the trannie gain is 20x then the 3pF of
collector-base capacitance has an effective value of 60pF. Yet again ...
1/(2*pi*R*C) gives the third roll-off at 53MHz.
So ... There are 3 low-pass roll-offs all at work together, 10Mhz, 16MHz and
53MHz. If you're looking at the waveform on a scope, then expect to notice
the response start dropping off in the 3-5MHz area.
In your case the figure is a very low 200kHz!, which implies a designed gain
of (say) x100 with a signal source of 2k7ohms or a gain of x100 with a low
source R but a collector resistor of (say) 4k7 with 170pF of loading, (or
any mixture of these two effects).
This stuff is the magical part of analogue design and to get a decent H.F.
response you need to try a number of tricks/techniques. The easiest is to
keep the gain low and design the collector resistance to be as low as
possible. Keep the source impedance as low as possible. Use 'R.F'
transistors with high Ft's and low collector-base capacitances. Keep the
collector strays and loading capacitances as low as possible.
As an example ... a single 2N2369 trannie with a 100ohm load can run out to
hundreds of MHz at low gains and small strays.
Other (more esoteric) techniques are also available, such as tuning out
stray C's by means of added small inductors and the use of 'shaped'
A HF, wideband, hi-gain ampifier can be a work of art so don't be dismayed
to have to use two transistors at HF when one would be OK at LF.
Also bear in mind that the two transistor 'Cascode' circuit is the best
technique available for wideband HF work. (other than putting the bits on
A Spice circuit can be very useful for this kind of HF stuff, as it saves
the wodge of tedious RC 'what-if' knock-on response calculations that need
to be done, if one small change is made to a single transistor stage.
Go Back To The Cyber-Spy.Com
Usenet Web Archive Index Of
The sci.electronics.design Newsgroup