Bipolar transistor bias question

[Richard Mackay]

The schematic is a dc power supply (13.8V) feeding a load directly if power is available or switches to the backup battery if the supply fails.
The TIP30A transistor is for current limiting to prevent the supply being overloaded when the battery is charging. (This works fine)
What happens however when the dc power supply fails is the transistor continues to conduct feeding the battery voltage from the collector to the emitter.
This holds the input to the comparator high (pin2) preventing the output of the comparator switching off.
The reason I used this transistor configuration was to minimize volt drops from the supply to the battery.
Apologies for no component identification. The comparator is a LM393

 

[MikeMl]

I see no means for measuring the current flowing into the battery. How can the circuit limit current?

 

[tunedwolf]

Looks like the transistor conducts all the time, there's no current limiting in there, as mike already pointed out. After a quick look at the schematic, one comparator sets the charging LED status whilst the other monitors the rail voltage and controls the relay. Unless I'm reading it wrong

 

[Richard Mackay]

The bias resistor connected from the base to 0V can be selected for the gain of the transistor for a maximum charging current (with a fully discharged battery)
The current drops off as the batt charges and is almost zero at the float voltage.
It's not the most elegant current limiting circuit I agree but it works better than a series resistor.

 

[MikeMl]

...
It's not the most elegant current limiting circuit I agree but it works better than a series resistor.

It is very dependent on temperature of the transistor, the phase of the moon, the value of the stock market, ... you get the idea...

 

[Nigel Goodwin]

It's not the most elegant current limiting circuit I agree but it works better than a series resistor.

No, it's not a current limiting circuit at all - you have no current limiting.

 

[Richard Mackay]

Current limiting aside see the transistor as a switch if that makes you happy.
The question posed is that when the the base-collector is forward biased the transistor will reverse conduct from the collector to the emitter.
This is the question I would like comments on. (e.g. Is this common knowledge & how well does it do this)

 

[crutschow]

............................
The question posed is that when the the base-collector is forward biased the transistor will reverse conduct from the collector to the emitter.
This is the question I would like comments on. (e.g. Is this common knowledge & how well does it do this)

Yes the transistor will conduct in the reverse direction, although its current gain will be much less than the forward direction.
One solution is to add another set of relay contacts (DPDT relay) in series with the transistor emitter that cuts off the power from the external source to the circuit (except for the two sense resistors to the comparator of course).

 

[Richard Mackay]

Thanks. This is what I figured but it's great to have it confirmed.
When the transistor isn't driven hard one can use a load resistor on the input that sinks this current. This isn't so easy when the current is higher since the resistor needs to be higher wattage. This is permanently connected to the supply so is wasteful.
Yes, relay and a diode diode are also options, each with their own issues.
In this case I looked elsewhere for an independent voltage reference indicating power was on. Fortunately one is available.