Electronics > Beginners

Looking for an alternative to 7812/7912 voltage regulators for low ripple PSU



I've gotten a very smooth output using a AC/AC transformer with a TIP 125 based capacitance multiplier. Less than 4.5 mVpp noise/ripple under a 650mA load.
The problem is, if I use a 7912 regulator, I either need a 16 volt transformer to get enough voltage (14-15 volts) after the capacitance multiplier. But this results in excess heat, either on the TIP125 or the 7912, based on which resistor value is use for the capacitor multiplier's base RC filter. If I used a 12 volt AC transformer, I get only 13.5 volts at the input to the 7912, which is not enough because it need at least 2 to 2.5 volts overhead (drop out voltage) to function.

So I'm looking for an alternative regulator, that has a lower dropout and can work with less than 1 volt overhead. This would keep the power supply much cooler. I need to sustain about 1 amp, so I'm looking for something rated higher than that.

I found the MIC29301 (spec sheet below) and it seems like a good choice, but I have some questions.

1. Is this a good choice?

2. Is there a negative voltage version, like the 7912/7812 or TIP 120/125 pairs? Because I need a bi-polar supply with a -12 volt rail too.

3. How would I use the fifth pin for on/off, if I wanted to have the output of the power supply wait for the smoothing caps to fully charge?

This is for an audio application, so it needs a very clean supply, but it also uses micro-controllers that will not initialize when they are connected while the power supply is "warming up". I really need a soft start on the input, to slow down the caps on initial power up, while at the same time, keeping the final output before the load "off" until it is stable at 12 volts, then turning the output on all at once.

I'm thinking I could use a Schottky diode backwards into the "on" pin of the voltage regulator, with a reverse breakdown voltage that is sufficient to drive the output, like 13 volts. This way, when the input to the regulator is less than 13 volts, no current flows out off it, but when the voltage at the regulator input is at least 13, it flows backwards through the Schottky diode to the "on" pin and opens the output. If the supply drops down below 13 volts (at the input side of the regulator), the Schottky stops flowing current to the "on" pin and the output shuts off. Is this a good way to make an output that's off until 12 volts is guaranteed, and shuts off before the output voltage drops below 12 (when the power supply is turned off)?

MIC29301 spec sheet:



need bigger main filter capacitor, capacitance multiplier wont help here

Look at the LM2490/LM2990 series, you should be able to make those work.  0.6V typical dropout at 1A, they don't mind output capacitance.  If you can give them ~1.5V VIN - VOUT then their output impedance and ripple rejection should be quite good.

David Hess:
There are alternative regulators like the LT1764 and LT1185 which are lower dropout and lower noise.

My preferred solution is to use the common regulators of your choice for current limiting and protection, but within the feedback loop of an operational amplifier and separate reference as shown below.  This can produce a reference quality high current supply. 

It should be implemented with modern parts.  A bipolar transistor can replace the JFET if the output range of the operational amplifier is sufficient.

Is there any reason to use a matched set for the positive and negative rails?

LT3015ET-12#PBF seems like a good choice, but I can't find a positive counterpart.


And is there a reason why the MIC29301-12WT in the original post is a bad choice? Other than not having a negative version?

I like the idea of having an on/off control on a separate pin, so I can use logic to have a low-voltage lockout during the initial charging of the caps, and in the event of any power dips.


[0] Message Index

There was an error while thanking
Go to full version