# Using half wave rectifier for low power applications

Im trying to make a low power application feeding it from a 9vAC, source, i have been reading that i can use a half wave instead of a full wave rectifier, i already did the circuit and been doing tests

Using the osciloscope you can see the wave forms from the Vin, the output of the diode, and finally the output odf the diode using a Avg sampling, im not sure if this waveform is enough to power a microcontroler using 3.3v im thinking about increasing the Cap value, and see how it modifies the rize, however, my main concern here is, how can i justify to do this instead of a full wave rectifier?, and also how can i make sure i will get the steady, 3.3v with as little noise as possible?

• No problem as long as the circuit meets needs of the load with regards to voltage and current. Add an inductor in series with load to filter and you should get a smooth DC. May 16, 2015 at 16:21
• Simply put: with a bridge rectifier you need half the filtering capacity at constant ripple. From a different POV: with the same capacity you'll have a smoother output. Is it really worth saving the cost of 3 rectifier diodes? May 16, 2015 at 17:36
• Something is wrong with your pictures. First screen shows a sine with about 17 volts peak - this should produce a 16V peak after rectification but it's more like 15.2 volts peak. Last screen does not make sense at all. What rectifier diode are you using - it should NOT be a 1N4148! May 16, 2015 at 17:41

I'm not sure what the third screen represents. What do you mean by "Avg sampling"?

The ripple you are seeing in screen 2 is normal. The rise is from the fast charging of the capacitor, and the fall is caused by the drain current from resistor R. I suspect that the small rapid fall in the capacitor voltage is caused by non-ideal behaviour of the capacitor (perhaps it is an electrolytic capacity with significant internal serial inductance).

To power the microprocessor, you need to install a regulator between the filter capacitor and the microcprocessor circuit. Options include linear regulators or switch-mode regulators. The former are easier. Through-hole options include LP2950-33LPRE3 or LM3940IT-3.3. In either case, you will need a second capacitor to filter the 3.3V output. Which regulator you choose depends on the expected load current, and power dissipation (power = voltage drop x load current). Your voltage drop will be rather high (about 10V ~ 13V - 3.3V), so you may want to think about changing the transformer to produce a lower output voltage on the secondary coil.

Powering a microcontroller with a half wave rectifier seems a bit unusual to me, since a very constant voltage should be used. However, as you are using a 9V AC voltage source with a half wave rectifier the output voltage of the rectifier should be more than enough to be used as a supply (as I can see in your pictures, the output voltage is about 12V).

I think the common way to get a constant supply voltage is to use a linear or switching voltage stabiliser (for the desired voltage of 3.3V DC). This would have two main advantages:

• You don't have to worry about stabilising the rectified 9V AC to 3.3V DC since the stabiliser does that for you.
• You don't need to have a very high capacitance since the stabiliser usually only needs 1V or 2V more than its rated output voltage. Nevertheless the capacitor after the rectifier is necessary, because otherwise you will get points in time where there is no output voltage at all.

A few more things you will have to worry about are:

• the thermal characteristics of a stabiliser, because the difference between input and output voltage is about 9V DC and this voltage will produce a lot of heat together with your load current.

• maybe some capacitors with smaller capacitances to stabilise the output voltage of the stabiliser.

You can put a bigger capacitor in if you need to, if the voltage to the microcontroller drops too low, then you need a bigger cap.