Thread: 3.3V Biasing
View Single Post
Old December 5th, 2006   #4
Nigel Worsley
Junior Member
Join Date: Jun 2006
Location: London
Posts: 13

Th equations are:

Vbias = Vcc * 1 / ( 1 / Rt1 + 1 / Rt2 ) / ( 1 / ( 1 / Rt1 + 1 / Rt2 ) + R1 + R2 )

Rterm = 1 / ( 1 / Rt1 + 1 / ( R1 + R2 ) )

The specification requires Vbias to be greater than 245mV and Rterm to be 120R +/- 10%

Calculation isn't all that simple, successive approximation with the aid of a apreadsheet is probably the easiest way.

If you settle on 250mV bias voltage, then the values work out as:

R1 and R2 = 396R
Rt1 = 141.4R
Vbias = 249.98mV
Rterm = 119.98

Neither of these are standard resistor values, the nearest are 397R and 142R. However, these are part of the E192 series, which means that they are often impossible to obtain in small ( less than ten thousand ) quantities. The nearest readily available values are 390R and 150R, the values I originally suggested.

This gives 259.8 mV and 125.8R which is close enough. If you want a better impedance match, then reduce R1 and R2 to 300R, which will give an exact match but increases the bias voltage to 330mV. This isn't likely to be high enough to cause a problem, but neither will the 125.8R impedance.

Note that the power dissipation in Rt1 is likely to be about 170mW, this will be too much for most surface mount resistors.
Nigel Worsley is offline   Reply With Quote