[Amps] bipolar input and output impedance

[dennis12amplify at aol.com]

Steve and Dan,

?RE: MRF454 Bipolar Input Impedance

?The Input impedance of?MRF454 Bipolar RF device's?B-E junction?varies widely with the applied B-E voltage and needs to be?analyzed?as you would?a diode.

?Since it 'looks' like a diode, it has a 'dead zone' (when the forward voltage applied across the diode is less than it's threshold of conduction),?and therefore has an extremely high ohmic value, and because of that fact the input impedance of the circuit at that point in time is?almost completely determined? by the shunt impedance?the bias circuit and any other parallel 'swamping' resistances that may be present,?Since it 'looks' like a diode it has what?I would consider to be three?entirely different?regions of operation.?(1) It has a junction voltage 'threshold' which must be met before it starts to conduct, then (2) it has a narrow semi-linear region of operation where the input current is relatively proportional to the input voltage, and then (3) It has a region of 'saturation' where the voltage across the junction increases very little for a very large change in input current up to the point of the junction's self-destruction. 

?Unless the device is biased class A and the voltage excursions across the input junction are very small, (in the neighborhood of 1 volt Peak to peak MAX), the input impedance of the bipolar device all? by itself can vary from megohms in the cutoff region prior to reaching the forward bias 'threshold', to only fractions of an ohm at saturation.

?The most frequently used and cheapest solution to the problem is what is called a 'swamping resistor' which is placed in parallel with the B-E junction. The value of that resistor is usually less than 50 ohms with smaller values adding linearity and the larger values improving conversion efficiency. A value of 1 ohm or less would be preferred at the cost of conversion efficiency with a 'compromise value of 10 ohms being the most?frequently used value.

??When the bipolar device's input resistance is?'swamped' by the?one ohm resistor, those huge variations in the input impedance impedance of the bipolar device's junction?are 'swamped' and reduced from megohms of change to only fractions of an ohm of change.?

? That is the most commonly used method of providing a relatively constant bipolar device input impedance, (albeit a bit low in value),?which can in turn be?'matched' back to the output impedance of the driving device, (another stage or to an external driver such as a transceiver),?as a fairly constant 50 ohm impedance?with the use of broadband transformers.

?Bottom line is that, as always, there is?that tradeoff?where we?have to reduce efficiency to gain linearity.

?Hope that helped,

?Dennis Ostrowski,
Retired Bell Labs design engineer and Motorola contract design engineer.

---------------------------------------------------------------------------------------------------------------------------------------

-----Original Message-----
From: Steve Thompson <g8gsq at fs2.com>
To: amps at contesting.com
Sent: Fri, 5 Jun 2009 2:56 am
Subject: [Amps] bipolar input and output impedance



MRF454 is a 12V transistor, TH430 runs from 45-50V. Follow 
Manfred's calculations and the transformer ratios will drop out. 
Motorola App note AN758 has a comprehensive design description 
that should be relevant to TH430s.

Steve

> What are the bipolar HF power transistor specs that influence or 
> determine the input and output impedance? I have an ENI circuit using 
> TH430 transistors with 4:1 input and output transformers, the EB-63 
> circuit based on MRF 454s uses 16:1 transformers. What transistor 
> parameters influence transformer design?
> Thanks - Dan
_______________________________________________
Amps mailing list
Amps at contesting.com
http://lists.contesting.com/mailman/listinfo/amps