Problem with diode at output of voltage regulator

[isaac12345]

Hi,

I am using an Onsemi MC33269TG voltage regulator as a constant current source to charge a pack of NiMh batteries (12V). I found that on switching off the circuit's input the battery pack powered the circuit back by providing current through the voltage regulator to the relay, then to the 7805 and to the pic,etc. To prevent this back flow of current, I added a 1N4008 diode right before the battery pack. The problem is that the diode has now altered the Vdrop of the regulator and increased it by approximately 1V. Is there any way of fixing this apart from changing the diode to one with a smaller forward voltage drop?

 

[tom66]

You could try a NMOS transistor connected to the low side of the batteries with the gate wired to your power source and a resistor (~10k) pulling the gate to the ground. This should effectively isolate the batteries when the system is off. However, I haven't tried this, so YMMV. A NMOS transistor with low Rds(on) should be selected, around 10 to 100mohm, and it must be capable of handling the charging current.

You may want to consider a relay in a similar configuration, if your application is not too cost or size sensitive, because those are pretty guaranteed to turn the batteries on or off.

And of course a Schottky diode if you can use it will only drop 0.5 volts or so.

 

[isaac12345]

could you please provide a schematic? I did not quite understand the resistor connection.