Reversed Biased MOSFETS & BJT's?

Your model may not include the 'parasitic' diodes that normally exist in these parts. If in doubt, add them externally in the simulation.

Reverse battery protection at he output of a circuit is the same as for the input. Googling should provide various suggestions.

RL

Bipolar transistors will often "work" when connected with emitter and collector reversed (although their speed, gain, and current-carrying capacity will likely be quite a bit different than in their normal orientation).

In your situation, whether the BJT will conduct will depend on what the base voltage is, compared to the "emitter" (i.e. what should be the collector) - if the base is biased to a voltage which will allow current to flow through this junction, then the collector/emitter path will start to conduct as well.

Most power MOSFETs have a "substrate diode" between source and drain. This diode is reverse-biased during normal operation, but will be forward-biased (and will start to conduct) if you reverse the source and drain connections. If your charger/discharger puts the MOSFET across the battery, with a current limiting resistor, then it's likely that reversing this battery connection will (in effect) put the battery into "discharge" mode regardless of what the control logic is doing to the gate.

... until it turns into a NED (noise-emitting diode) for a brief moment.

You might need to use a Schottky diode to isolate the battery from the charge-discharge circuit, and run a second connection to the battery (bypassing the diode) to do the voltage sensing.

It may be worth avoiding the issue altogether - on most of the commercial battery chargers I have, a relay click can be heard which is the reverse polarity protection. Although most of the chargers contain a microcontroller, the protection could be implemented much more simply in your charger with a comparator driving a relay to connect the charger circuitry to the output leads.

This would have the same shortcoming as the commercial units that if the battery is discharged dead flat there is no voltage for the comparator to sense as "correct polarity" I keep a spare current limiting PSU on hand to get the battery started on these rare occasions.

For discharge I was planning to have a NPN transistor with a current sense resistor going to ground. Then apply some base current to get my programmed discharge current to flow. Perhaps I need the current sense resistor on the collector so I don't have to figure in base current. If I need to, I could add in a Schottky diode to prevent current flow in a reverse polarity situation. But if I don't need the diode, I would prefer not to have the drop in situations of discharging a single 1.2V cell.

For the charge side, if I have the battery grounded through a current sense resistor, and the switch on the high side, then a properly connected battery would have Vplus - Vbatt across the collector to emitter. A reversed battery would have Vplus + Vbatt across the transistor. I thinking a PNP common emitter or P channel transistor with base or gate pulled high and switched lower to turn ON. Seems like the charge side would be correctly biased and I would just need to use a transistor rated for supply voltage + battery voltage.

Does this sound right or is there a better way?

Thanks! RogerN

You could also put a hefty diode in series with an appropriate fuse (fast fuse) across the battery. If the battery polarity is reversed, diode conducts and the fuse pops.

Fred

----== Posted via Pronews.Com - Unlimited-Unrestricted-Secure Usenet News==----

---= - Total Privacy via Encryption =---

Sense the battery voltage with a comparator - if correct polarity, energize a relay that connects the remainder of the circuit to the battery.

Ed

It took you over a month to read my reply and pinch my idea!

Sorry - I didn't see your reply.

Ed