It should be based on a simple Open-loop configuration using the following components (no ICs)
a) any NPN transistor
b) Several zener diodes
c) Any passive components
d) A heat sink for the power transistor
Specifications:
Nominal output voltage = 12V @350mA load current
Maximum load current capability = 700mA
Variation allowed at the output = +/-200mA
Input voltage range = 16 to 20V
Maximum temperature =65deg Celcius
Use of E12 range of Resistors with +/-10%
Calculate approximate range of output voltage, under worst conditions with temperature range of 20 deg C to 50deg C, and output impedance.
What happens if we use another Zener diode to get the effects of input voltage variations buffered from another diode?
How do we select the value of the second zener diode?
How do we select the transistor?
How do we deal with the temperature specification?
If you're alowed to use PNP's as well, you could have maked an poor man's opamp circuit. From there you could have used it to stabilised voltage, ref voltage folower.
Then use a Darlington!
I don't see how to use multiple Zeners, one to per-regulate the other, when your input voltage can get down to 16 volts and the output needs to be 12 volts. A transistor needs to have 12.6 volts on the base or a Darlington needs 13.2 volts. That does not leave much across the resistor, and if you have two resistors even less voltage per resistor. This looks like home work so you can use 10 watt Zeners and power resistors. No one said this has to be efficient.
The simplest transistor voltage regulator is made with an NPN transistor with a zener diode cathode on the base and anode to ground. The zener is biased with a resistor from the positive supply voltage to the cathode of the zener. The output load is connected from the transistor emitter to ground. You can improve on the basic design in various ways.
That gives your basic regulation subject to temperature changes because of the response of the zener and the base emitter diode of the transistor. It's not that bad though. If you use a zener above 5.6v the zener temp co partly cancels the transistor base emitter temp co.
I do not agree with MrAl.
The temp co of a Zener is best at about 5 volts.YES But it does not cancel out the B-E of the transistor.
Attached is part of a diode data sheet. At 3 volt the temp co is (-) and goes positive with higher voltage. Usually it crosses zero around 5 volts. If you want good temp then use two 6 volt Zeners and add a 1n4001 diode to compensate for the transistor.
I do not agree with MrAl.
The temp co of a Zener is best at about 5 volts.YES But it does not cancel out the B-E of the transistor.
Attached is part of a diode data sheet. At 3 volt the temp co is (-) and goes positive with higher voltage. Usually it crosses zero around 5 volts. If you want good temp then use two 6 volt Zeners and add a 1n4001 diode to compensate for the transistor.