The Diode Rectifier

A Note on behavior with input sinusoids with time-varying envelopes

        This note is intended to help you learn more about the behavior of the diode rectifier, especially when the diode rectifier is used to produce a DC signal from an AC transducer.  In those situations, the AC signal being rectified can - and will - change.  (That's the purpose of using the sensor - to measure a changing physical quantity.)  When the input to the diode rectifier changes, that produces a situation different from the one we have considered so far where the input is a constant.

        We have two simulators available to help you understand what happens when you have a sinusoidal signal with a time-varying amplitude.  (You should have gone through the interactive problem available here.)  In the first simulator, the input signal is a sinusoidal signal that has an amplitude that increases in a linear fashion.  In the second simulator, the input signal is a sinusoidal signal that has an amplitude that decreases in a linear fashion.  Both simulators use an ideal diode model in the simulations, i.e. there is no forward voltage across the diode when it is conducting, and no current leakage what it is back-biasesd.

        Using these simulators you can learn about the behavior of the diode rectifier circuit when you have a sinusoidal signal with a varying amplitude - and after all the information is often contained in the amplitude.  Here are some things to do first.

• Run the up-ramp simulator using the pre-loaded paramters.  Note the following:

• The signal that you recover does not increase linearly.  At first it looks like a sequence of steps.  That's not quite what's there.  You can see that better if you change the time constant to 10 sec. or 5 sec.  You can see that there is more droop at the higher levels.  Actually, the percentage droop is the same for all cycles, but it produces a larger slope at the higher levels.

• Run the down-ramp simulator using the pre-loaded paramters.  Note the following:

• The signal that you recover does not decrease linearly.  There is droop, and at first the droop slope is larger than the slope of the down-ramp on the envelope of the sinusoidal signal.  Later, however, the droop slope cannot keep up with the decrease in the sinusoidal signal and the recovered signal just decays exponentially.

        Now, when you want to use these circuits you need to be aware of the two phenomena that you can see in the simulators.  In particular there are a few points to note.

• For increasing signals, the rectifier circuits does not interpolate between successive peaks.  You will get plateaus or decreasing signals between peaks.  There's not much you can do about that unless you go to a more complex circuit.

• For decreasing signals, the rectifier circuit does not follow successive peaks.

• For peaks that do not decrease by a large percentage, the rectifier output voltage may fall below the next peak.

• For peaks that decrease by a large percentage, the rectifier output voltage will not fall fast enough to keep up with the decrease in the input voltage.

• Examples of the above two statements can be seen in the simulator for a decreasing ramp envelope.

• Use the default parameters, and you can see that the initial part of the response is a situation where the output peaks are not decreasing as fast percentage wise as they are later.  In this part of the response the rectifier output falls below the signal peaks.

• In the last portion of the response, the output peaks are decreasing by a large percentage each cycle, and the rectifier output cannot fall quickly enough.  At that point, the signal decreases to zero, but the rectifier output is decaying exponentially to zero.

        There is a dividing point between the two cases for the output signal.

• The output voltage will match the next peak exactly when:

• e^-t/t

  = Next Peak/Current Peak