PDA

View Full Version : Help Needed On Feedback


geelude
Jan26-05, 08:22 AM
Hey guys,

I am currently solving a problem involving shunt series-series feedback. So far this is what I can say about the circuit:

- Series mixing is feedback combined with source
- Negative voltage feedback samples output voltage
- Series voltage feedback stabilises close loop gain and should be used to get precise voltage gain
- Negative feedback has disadvantage of reducing gain of amp but can be overcome by extra stages of amplifier
- Advantages are stabilises gain, reduction in nonlinear distortion, reduction in noise and also controls input and output impedance
- Unwanted noise coming from power supply, coupling of signals from other circuits, feedback can only reduce thermal and shot noises
- In voltage feedback input terminals of the feedback network are in parralel with the load and output voltage appears at the input terminal of feedback
- A way to test for voltage feedback is to see if signal vanishes as short circuit
- As series voltage feedback stabilises close loop voltage gain it is good for designing amplifiers with precise voltage gainse
- Negative feedback factor is larger than unity therefore negative series feedback increases input impedance
- Negative voltage feedback reduces output impedance


Can anyone else help me out, what else can I add and say about this feedback network?

The link to the problem is as follows:

http://www.angelfire.com/electronic2/helpneed/index.html

Please help!!!

Thanks!

rayjohn01
Jan28-05, 02:24 PM
In many circuit applications feedback is used to help fix input and output impedances . Typically shunt will reduce the impedance to low values , so that an external but well defined resistor then defines the (say ) input impedance.
Series does the opposite
. It is possible to use a combination to define an accurate matching impedance (say 50 ohms) without ever using a 50 ohm resistor and hence gain a noise advantage.
One of the most important uses is to linearise a circuit -- most transistor circuits are inherently non-linear ( not all) -- the transfer funtion io/vi
is partly diode characteristic dependant ( exponential ) but also beta dependant ( i.e. transfer efficiency ) , so that for linear applications ( Audio amplifiers , or instrumentation ) you must use feedback to overcome this.
The limitations of these techniques is that unless carefully designed they can oscillate , to avoid this the internal amplifier is usually rolled off in gain at a 6db/octave rate -- Opamps are a good example where a single capacitor dominates the roll off .
This has the effect of reducing the amount of feedback with frequency
so severly limits the 'useful' range where the feedback is effective.