Simple reverse-polarity-protection circuit has no voltage drop
A relay does what semiconductors can’t do.
Aruna Prabath Rubasinghe, University of Moratuwa, Moratuwa, Sri Lanka; Edited by Paul Rako and Fran Granville -- EDN, October 20, 2011
Common methods of reverse-voltage protection employ diodes to prevent damage to a circuit. In one approach, a series diode allows current to flow only if the correct polarity is applied (Figure 1). You can also use a diode bridge to rectify the input so that your circuit always receives the correct polarity (Figure 2). The drawback of these approaches is that they waste power in the voltage drop across the diodes. With an input current of 1A, the circuit in Figure 1 wastes 0.7W, and the circuit in Figure 2 wastes 1.4W. This Design Idea suggests a simple method that has no voltage drop or wasted power (Figure 3).
Select a relay to operate with the
reverse-polarity voltage. For example,
use a 12V relay for a 12V supply system.
When you apply correct polarity to the
circuit, D1 becomes reverse-biased, and
the S1 relay remains off. Then connect
the input- and output-power lines to the
normally connected pins of the relay, so
current flows to the end circuit. Diode
D1 blocks power to the relay, and the
protection circuit dissipates no power.
When you apply incorrect reversed
polarity, diode D1 becomes forward-biased,
turning on the relay (Figure 4).
Turning on the relay cuts the power
supply to the end circuit, and red LED
D3 turns on, indicating a reverse voltage.
The circuit consumes power only if
reverse polarity is applied. Unlike FETs
or semiconductor switches, relay contact
switches have low on-resistance,
meaning that they cause no voltage
drop between the input supply and the
circuit requiring protection. Thus, the
design is suitable for systems with tight
voltage margins.
Talkback
-
This is an interesting idea, but it does have at least one flaw.
The design assumes that the power connected backwards will supply enough voltage and current to power the relay coil. Suppose the device is battery powered and the replacement battery, plugged in backwards, is used and doesn't have enough power to operate the coil. (Many people will 'try' a used battery to limp along until a new replacement can be acquired.)
It is quite possible that the power supply may not be able to activate the relay, yet still has enough power to damage the circuit it is intended to protect.
Chris Grams - 2012-3-1 21:30:50 PST -
One other option that would be simple and less expensive would be to use a polymer PTC PolySwitch device in series, and then place the diode in the parallel leg. If you accidentally plug in the battery backwards, the diode would be forward biased, and you would have a high short circuit current that would quickly trip the PolySwitch device. Since the diode is in the parallel leg, it would add no voltage drop. Depending on the PolySwitch device used, it would add very little resistance to the circuit. The additional cost of these 2 components will be much less than a relay circuit, and the PolySwitch device would help provide resettable overcurrent protection with no warranty returns or service calls.
Barry Brents - 2011-20-12 10:56:47 PST -
1.There some mistakes with battery electrodes images on Fig.3 and Fig.4.
2.Fig.3 the battery’s electrode positive terminal should be shown as a long and the negative terminal should be shown as a short on schematics. It means the positive battery terminal attached to ground. So the relay should be ON and contact S1 should be connected to NO and LED is ON.
3.Fig.4 the negative battery terminal is attached to ground, relay is OFF.
4.The relay contact resistance is not equal to zero and is dependent of relay design and could be in range from 5-10mOhms for high power relays and up to 0.25 Ohm or higher for reed relays, so it could be comparable with MOSFET Ron resistance.
5.Use a relay for switching power for a load with a capacitor in parallel will be significantly worse than use MOSFET.
Vladimir Doubovis - 2011-1-11 15:39:59 PDT -
It would be safer to have the relay open by default, as damage may occur before the contacts disengage. Old, old idea, but there are entire generations of EEs who need the common sense...
Stephanus van Schalkwyk - 2011-28-10 17:11:04 PDT -
I dont know how many times this circuit has been "re-heated" since the "ideas" columns in the 1970s, but it seems to crop up at regular intervals. It is hardly original.
Alan Melia - 2011-28-10 12:35:54 PDT
