 ### Network # Question about Ft (cut-off frequency)

J

#### Joe

Jan 1, 1970
0
Hi there, I am trying to understand more about High Frequency RF
measurements of Bipolar Transistors. I have a question about
measuring Ft.

Ft is defined as the frequency for which the magnitude of the AC
current gain (h21) drops to unity for a transistor in the
common-emitter configuration with an AC-shorted collector.

Now most of the readings I have done discuss calculating Ft through a
method of extrapolation. They discuss that there is a -8dB/decade
slope in the curve of (h21) vs Frequency, and that a measurement
should be taken in the region displaying this behaviour, then a linear
extrapolation (in the semi-log plot) should be done to determine the
cut-off frequency.

Now my question is why do we even bother with the extrapolation? Why
not attempt to measure Ft at the cut-off frequency? I understand that
Ft is measured using S-parameter data which is converted to
h-parameter then the calculation takes place, but why not devise a
system to systematically calculate h21 at each frequency and increment
the frequency till h21 is equal to unity (i.e. stop when h21 = 1).

Is there a stability reason, or is it something else?

Thanks in advance for the help!!! C

#### CWatters

Jan 1, 1970
0
Now my question is why do we even bother with the extrapolation? Why
not attempt to measure Ft at the cut-off frequency?

It's 20+ years since I did this but..... I think it's because in practice
the Ft you measure won't be at the same frequency as that predicted by
extrapolation - at such high frequencies secondary and parasitic effects
take over. So which Ft is more useful? Let's say you are trying to predict
the gain at a particular frequency. If you use the Ft extrapolated from the
data and -8dB/decade you the right answer (eg a point somewhere near the
actual measured data). If you use the measured Ft and -8dB/decade then you
are you will be out.
I understand that
Ft is measured using S-parameter data which is converted to
h-parameter then the calculation takes place, but why not devise a
system to systematically calculate h21 at each frequency and increment
the frequency till h21 is equal to unity (i.e. stop when h21 = 1).

.... and I've long since forgotten that stuff!

Colin

Replies
4
Views
1K
J
Replies
7
Views
4K
Active8
A
T
Replies
12
Views
2K
Guest
G
Replies
6
Views
270
Replies
2
Views
307