III. Gm-C Filtering - Epublications - Université de Limoges

Idiff (A)
levels reached with this technique are still poor, as it may be observed in Table 8, and very
high currents are needed to reach high gm values.
0,0015
0,001
0,0005
0
-0,0005
-0,001
-0,0015
-0,5 -0,4 -0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 0,5
Vin (V)
0,5mA
1mA
1,5mA
gm_diff (A/V)
- 97 -
0,004
0,003
0,002
0,001
0
0,5mA
1mA
1,5mA
-0,5 -0,4 -0,3 -0,2 -0,1 0 0,1 0,2 0,3 0,4 0,5
Vin (V)
Figure 111. Linearization of the transconductance by current increase
III.2.c Active and Passive Source Degeneration Technique
The principle of active and passive source degeneration (or emitter when dealing with
BiCMOS technologies) is to use an active or a passive device, connected to the source of the
MOS transistor to linearize the transconductance gain. In this configuration, illustrated in
Figure 112, the impedance Zdeg acts as a local feedback.
Vin
Iout
gm
Zdeg
Figure 112. Source degeneration
Using feedback theory, the equivalent transconductance of this stage can be computed.
It gives:
g m
g m _ deg = . (III.33)
1+ g Z
m
deg
Hence, for gm.Zdeg>>1, it can be approximated to
1
g m _ deg ≈
Z
, (III.34)
deg
which is very interesting since the transconductance can then be set by a single impedance.
Resistors are usually used but they increase noise (as demonstrated in APPENDIX D). Indeed,
the noise power is proportional to Zdeg. That is why inductances are preferred. Feedbacks
making use of transistors are used as well.