Problem with Diode reverse current

[rakeshm55]

Hi,
In one of the project I am working I am having two source of power one from a wall wart 5V and other a single Li ion Battery. I am using power ORing using PMEG3020EP schottky Diodes for automatic switch over of source.

The Li ion battery is also connected to a Single cell Battery charger a
XC6802 a Torex chip with input from the wall wart 5V.
When the battery is unconnected the chip periodically Pulses its output at an interval of 2.5ms as an attempt to charge the Li battery which is not present. In order to avoid this pulsing I have implemented N channel switch (Q4) which will disable charging when battery not present.
Please refer to the schematics attached.

Now The switch (Q4) is always on whether the battery is connected or not . This is mainly due to the reverse current of PMEG3020EP which is of the order of 0.2mA @ 5V reverse. Due to which I can observe the pulsing even when battery is disconnected. How can i solve this problem in the current setup. Is there a work around ,What if I replace R33 with a diode.

PLease help.

[TerminalJack505]

If your diode allows 0.2 mA when reversed biased from 5V to GND then it basically looks like a 5V source in series with a 25K resistor.  Which would explain why it turns Q4 on in your circuit.

This is a pretty high impedance source.  When the battery is present it will look like a voltage source with a much lower impedance.  You might be able to exploit this. 

One way is to replace R33 with a short and change R35 to something like 2.2K.  That way, when the battery is absent, the bulk of the voltage drops across the diode instead of R35.  This will prevent Q4 from turning on.  When the battery is in the circuit then the diode isn't really in play since it will be forward biased so the battery's voltage will be dropped across R35--turning Q4 on.

I'd be a little worried about the BAT pin latching Q4 on, however.  Once it turns on, this scheme wouldn't be able to determine whether or not a battery is present.  I don't know enough about the charging IC to know if this would be a problem or not.

[amyk]

Presumably you chose a Schottky because of the low forward voltage, but the main drawback is the huge reverse current. You can try a different diode; I know there are specific low-leakage current types available (down to pA range) although they tend to have high forward voltage and low forward current ability, but even regular rectifier diodes will have lower reverse currents in the <10uA range.

One way is to replace R33 with a short and change R35 to something like 2.2K.

That will put a constant 2.3mA draw on the battery when it's connected, not sure if that would be significant, as opposed to the <5uA it would be in the original schematic.


...or how about an old-fashioned latching relay? Virtually no leakage with those

[TerminalJack505]

Yeah, I don't know if the 2.3 mA will be an issue or not.  The diode is rated for 2A so I kind of figured this is going to be used for a moderately powered application.

Another option would be to ditch the diode and use an OR-ing controller IC.  Something like the LTC4412.  This would avoid the reverse leakage problem completely and instead of a 500 to 600 mV loss across the diode you would have a ~20mV loss across a MOSFET.

[mikeselectricstuff]

Something to be aware of is that schottky reverse current increases significantly at higher temperatures, so before deciding whether you can live with it, make sure you check you can live with however high it gets at your maximum temp.

[rakeshm55]

My load current from the battery is about 150mA . A 2.3mA is a 2% increase I think I can afford it there is another load connected to directly to the battery which i have not mentiond in the schematics. its an LED. A good amount of reverse current will be absorbed by LED so I can even put higher values of resistance.
The data sheet specifies a reverse current of 30uA @ 5V but when I shot VBAT to ground I am getting a current of 0.2mA(May be due to charger). If I keep the circuit as it is the potential across R35 is 1.3V approximately. With LED disconnected it is 2.1V.
Let be try soldering 2.2Kohms instead of 1M....

[rakeshm55]

Thankx,
Connecting a 2.7K across R35 and shorting R33 has done  the trick. The Q4 is OFF with a gate voltage of 0.7V.
My device operating temperature range is 0 -40^C. This limitation is from the side of the sensor so I suppose I may cross the hurdle