Accidentally connecting a battery reverse can be a real killer in electronic circuits. There are a number of ways to provide protection and all have the expected advantages and disadvantages. Any protection device connected in series with the battery reduces its voltage, and when the battery voltage is only in the order of 1.5 or 3V, every lost millivolt is significant. The highlight of this article is the use of a very low loss active or synchronous rectifier in low voltage applications.
Schematic of different types of protection
What types of circuits are affected?
Generally speaking, circuits with bipolar transistors do not exhibit excessive fault currents when the battery is reverse connected due to the presence of strategically placed resistors. However, integrated circuits are VERY SUSCEPTABLE to reverse battery connection – most simply die instantly and silently – others may smoke and/or explode. For this reason, NEVER design or make a circuit that does not have reverse polarity protection. All battery operated consumer equipment has this protection for good reason.
Circuit protection matrix
A battery load of 50mA is suggested in all cases for the purpose of comparison – actual requirements may be lower or higher.
Series rectifier: OK for 9 or 12V applications, but the typical 1V drop is a big chunk of a 3V supply.
Schottky rectifier: better than a standard rectifier, but the typical 0.3V drop is marginal for 3V.
Synchronous rectifier: best for 1.5 & 3V applications, but marginal at 4.5V.
Shunt rectifier: I have seen this done in a low power application. While it does prevent reverse voltage from exceeding about 1V, the battery is in danger of exploding should it become overheated. What happens is that the rectifier shorts the battery thus preventing full reverse voltage – very crude! Not recommended.
One exception for the shunt rectifier is the solar panel application – the transient voltage protection diode can double as a reverse polarity shunt device. This works safely due to the current limiting properties of the solar panel.
A standard PNP transistor base-to-emitter junction can act as a low voltage rectifier (e.g. the 2N4403 is rated for 5V in this mode). When voltage of the proper polarity is applied, base current flows and the collector saturates thus supplying power to the remainder of the circuit. Saturation for the 2N4403 is 0.07V @ 50mA (suggested battery load current). Note that this is far lower than even the schottky rectifier.
Note also that an NPN transistor may also be used if connected properly, but I prefer to use a high side PNP transistor even if the saturation voltage is marginally higher.
There is one drawback: the base current that turns the transistor on is “wasted.” With an applied hFE of 10 and an Ic of 50mA, the base current is 5mA. While this adds load to the battery, I believe that less power is wasted here than in a series rectifier. If a more optimum bipolar device is used such as the Fairchild KSA928A, an applied hFE of 50 may be used so the wasted current is a factor of 5 lower or only 1mA – in addition, this device saturates to only 35mV @ 50mA.
If the synchronous rectifier is applied at a battery voltage of 4.5V or higher, the transistor Vbe breakdown voltage must be individually tested and evaluated at maximum battery voltage to assure proper operation. Many devices will support up to 7V or so. The voltage may be easily checked by connecting the base-to-emitter junction to a 10K series resistor, connecting it across a 9V battery and measuring the actual zener breakdown voltage. The 2N4403 that I was using indicated an actual breakdown voltage of 7.94V—much higher than expected. As a result, this could be possibly used in 6V applications.
Testing an actual circuit
I wired up a simple test circuit using a 1.5V battery, 30Ω load, 2N4403 transistor, and 180Ω bias resistor. The measured voltage drop (Vce sat) was 72mV. This compares very favorably with the published graph. Leakage current with the battery connected reverse was immeasurable.
Problems caused via dual power source
When a battery operated circuit also has an AC adapter, the voltage across the diode or synchronous rectifier may be more than doubled. Some adapters plug in via a jack that has an integral switch that physically disconnects the battery (including the reverse polarity protection rectifier). If this is the case, there is no issue. However, such contacts are often cheap and unreliable – I have seen this in numerous cases and it can be a real headache. As a result, an improved means of accomplishing this function is to avoid the unreliable contact and connect the external power adapter via a diode. In this case, the voltage across the reverse polarity protection device is more than doubled and probably precludes the application of a synchronous rectifier – keep this in mind.
Selected device specifications
For the future
Undocumented words and idioms for our ESL friends
applied hFE – my own electronic idiom –while each and every bipolar transistor has its own actual hFE prarmeter (Ic /Ib), the applied hFE is how it is applied in a circuit. e.g. the actual hFE may be 100, but in a circuit it may be applied with much higher base drive to assure full saturation – the actual collector current /actual base current is the applied hFE – a commonly used design value is a factor of 10.