Dick Cappels' project pages http://www.projects.cappels.org
Return to HOME (more projects)

White LED Battery Powered Power Failure Light
When the AC power goes out, this light comes on so you won't be left searching in the dark for a flashlight.

The use of the plastic case from a wall-mounted transformer results in very compact package.

Find updates at www.projects.cappels.org


Where I live most of the time, the AC power drops out upon occasion. This used to leave me groping around in the dark for a flashlight. I could have bought a battery-backed up "emergency light" for about US$35, but being basically metal boxes with floodlights attached, they would make my living room look more like a prison. And besides, I make things like this for fun, so thinking about the problem eventually lead to the solution on this web page.

The power failure light that I eventually designed and built is housed in the plastic case from a wall-mounted power transformer. I had accidentally destroyed the transformer by shorting its output, and had kept the plastic case for years, just waiting for a project like this one. Since its mounted in a hand-sized enclosure, when the power fails, it is a simple matter to walk over to the light, unplug it from the wall, and use it like a flashlight to walk into other rooms to retrieve a flashlight. This particular lamp dims after a few minutes, and is down to a dull glow after about 15 minutes. The glow can last for hours after that. I suspect that this is at least partly the result of using rechargeable batteries from a surplus store.

As of this writing,  I have had this power failure light in my living room for a little more than a year. It has come on many times when the AC line failed, and it has worked as intended. I am very glad that I made it.

The two white LEDs are the light source when the AC power goes out.
The green LED is on when the AC power is on to show that the battery is being
trickle charged.  The light
would look a lot nicer if I were to have painted
the case to cover up the markings from the original use of the plastic case.


This is not a do-it-yourself construction project. This is a project that I built for my
own use, and I have chosen to publish this as an example for the benefit of those
with similar interests and who have the technical skills to design their own.

The Circuit

Many of the components in this circuit are dedicated to safety-
keeping transient voltages from destroying the lamp,
and keeping the lamp from catching fire. Click on the schematic
for large version.

The circuit can be viewed as four blocks. The AC  line input and trickle charge ring circuit, the power failure switch, and the battery operated white LED driver. We will look at the blocks one at a time.

AC Input and Trickle Charging Circuit

The first thing you should notice about the AC Line Input and Trickle Charging Circuit is the fuse. Not only one fuse, but two, because R1 is a  fusible resistor.  I'm pretty concerned about personal electrical and fire safety, especially when it comes to AC line operated devices in my home. I'll come back to the fuse later, but first, a little bit about how the circuit works.

If we take the circuit node that is represented in the schematic by the lower AC line input to be our reference, I can talk about the upper connection shown in the schematic -the one that F1 is connected to, as having the full AC power line voltage on it with respect to our reference, and it will be a lot easier to describe the circuits operation.

Under normal, operation, a little while after power is first applied to the circuit,  the AC voltage from the power line sees R1, the 20 Ohm resistor in series with C,1 R3, and the combination of D1, D2, and D3. The 24 volt Metal Oxide Resistor (MOV) does not conduct in normal operation.

The RMS current through the input circuit is approximately equal to the input voltage divided by the series impedance. Impedance is equal to the square root of resistance squared plus the reactance of C1 squared.  The reactance of C1 is
1/(6.28 x 50 x .000000068) = 47,000.

This is very large compared to the resistance, which is dominated by the 20 Ohm resistor, so as a practical matter, we can say that current is equal to the input voltage divided by the reactance (47k) of C1.

The RMS current, in amps, is equal to 240 VRMS / 47k =  5 milliamps RMS. Since average current is equal to 0.9 times RMS current, the total average current available to trickle charge the battery and keep C2 charged up is 4.5 milliamps.

Since most of the voltage is dropped across C1, the current is practically independent of the state of charge on the battery, B1.

A few critical words on the election of components for this part of the circuit. The key safety components are C1, R1, and F1. C1 is an X type capacitor. X type capacitors meet certain safety specifications and are designed to be safely used across the AC power line. Other types of capacitors may fail from the constant AC power line voltage across it, and when they fail, the results could be deadly. Don't use other types of capacitors. The reason that C1 is a .068 uf capacitor is that this is the largest value  X capacitor that I could get my hands on.

A review of safety capacitors can be found at the URL below (you should be able to click on URL -just remember to come back here!):

F1 is there just in case C1 does short out. I feel that, although the likelihood of C1 failing as a short are extremely remote, that to me, its worth the small cost of the fuse to provide a little additional peace of mind.

Relevant to C1, R2 and R3 are across C1 to bleed off the charge on C1 so that I won't get shocked if I touch the power failure light's AC line contacts. The time constant is only 1.4 Meg Ohms x .068 uf = 0.1 second, so the voltage would bleed down to less than 5 volts in about half a second.  The reason I used two resistors in series instead of just using a single 1.5 Meg Ohm resistor, was to assure that there was sufficient margin in the resistor's voltage rating. The resistors I used were nonflammable 1/4 carbon film type.

R1 is there to limit the maximum current when the power failure light is first plugged into the AC line. Since the RMS voltage is 240 volts, the maximum peak voltage across the input is 240 volts x 1.414 = 340 volts peak. If the power plug makes contact with the AC line at the instant the power line is at 340 volts.  Since C1 is discharged when the power failure light is first plugged into the AC line, and the MOV clamps voltage at about 24 volts, the maximum current through R1 and the MOV (Z1) is approximately
(340V - 24V) / 20 Ohms = 16 amps. Without R1 the current would be very high, and a small fraction of that current could damage one of the semiconductors in the circuit if, by chance of a ground loop, some of that current were to get into other parts of the circuit. It also keeps the sparks down when you plug the power failure light into the wall.

R4 serves the important purpose of limiting the current into D1 and D2 or into D3, depending upon which half of the power line cycle is present when the power failure light is plugged in. Since the maximum voltage across V1 is around 24 volts peak,  the maximum current  into D1 and D2 or into C3 is 24 volts/330 Ohms = 7 milliamps. This is just a little bit more than the peak current current though C1 during normal operation, so V1 does not conduct much during normal operation.

On the positive half cycles of the power line, current through R1, C1, and R4 charge B1 through D1, while D2 charges C2 to the same voltage as B1. B1 is the nickle-cadmium battery that power the light in the case of power failure and the voltage across C2 is used by the circuit to sense when the AC line voltage is present.

During the negative half cycle of the power line, the green LED (Light Emitting Diode) conducts.  This diode is necessary for the for the circuit's operation, and it seemed to be a good idea to use the current for something useful - like showing that power is connected and that the charger is working. If you want, you should be able to use a silicon diode such as an 1N916 instead of an LED. I have not tried this substitution, so I can not vouch for it; I only recommend it as an experiment.

In my power failure light, B1 is made up of two Nickel Cadmium AA cells in series. I tried two Lithium Ion cells, but the internal resistance was too high. Both the Lithium Ion and Nickel Cadmium cells were both surplus, so I would not be surprised if a good Lithium Ion cell would work well in this circuit.

Power Failure Switch

Q1 is the power failure switch. When it is on, it connects B1 to the White LED Driver part of the circuit. C2 is kept charged to the battery voltage- the same voltage as Q1's emitter, and since the base of Q1 is connected to C2 through R5, Q1 is "off." When AC voltage is removed from the circuit, C2 is discharged through R6. The voltage at the base of Q1 is about 0.6 volts lower than that on the emitter, and when C1 is discharged, the base current is (2.5V - 0.6V)/(330 Ohms + 1k Ohm) = 1.4 milliamps. This should be sufficient to provide over milliamps of collector current, given that  the transistor is specified to have a minimum gain of 100 in the range of 10 to 100 milliamps.

C3, The 100 uf capacitor on the collector of Q1 is there to maintain a low output impedance to the White LED Driver at the White LED Driver's power supply input.

Battery Operated White LED Driver

The White LED Driver is an adaptation of the simple flyback/blocking oscillator white LED power supply shown elsewhere on this web site.
White LED Drive Circuit       http://www.cappels.org/dproj/ledpage/leddrv.htm

Please see that page for the general theory of operation, the specifics of component selection, and making the inductor.

The major difference between the circuit described in the White LED Drive Circuit web page and this one, is that this circuit has a secondary on the transformer. The flyback pulse on the secondary is in series with the flyback pulse on the collector of Q2, and this available voltage higher. A good idea in this case, since there are two white LEDs in series.

Another difference to take note of, is that the base resistor for Q2 is only 330 Ohms. This increased the output drive to the LEDs of what I would have had with a 1k base resistor.

I briefly considered using the small 3 mm ferrite cores that were featured on the White LED Drive Circuit web page, but quickly realized that I would obtain better performance with one of the larger high permeability cores that I had on hand.


Your goal should be to test this circuit without any lethal voltages connected.

The testing was done in stages. First, I built and tested the Battery Powered White LED Driver, first using a bench top power supply, then using actual batteries. After the White LED Driver was running nicely, I built and tested the AC Line Input and Power Failure Switch circuits. In testing these circuits, I used a low voltage DC bench top power supply, set to about 12 volts, with a resistor between the bench top supply and the anodes of D1 and D2 to simulate the current from the AC Power Line.

By using a low voltage power supply instead of the AC power line, I could freely troubleshoot the circuit without fear of getting a nasty shock. 

After testing on the DC bench top power supply, I rechecked my wiring of the AC Input and Trickle Charging Circuit, mounted everything in the box from the wall-mounted power supply, carefully closed up the box and plugged it in. I did not troubleshoot any part of the circuitry with the circuit connected to the AC power line. Maybe I was lucky to not have made a wiring error.

If I had needed to check anything with AC power connected, I would have used a safety setup similar to the one shown on the "An Experimental Triac Dimmer with Simulated Diac web page on this web site.
An Experimental Triac Dimmer with Simulated Diac            http://www.cappels.org/dproj/dimmer/Triac%20Lamp%20Dimmer%20With%20Simulated%20Diac.html

Remember, this circuit uses lethal voltages. Try not to have the power line connected to the circuit while testing it. If you must have power line voltage connected to the circuit, use a Ground Fault Interrupter AND an isolation transformer setup as shown on the An Experimental Triac Dimmer with Simulated Diac page. And keep one hand in your back pocket so as to avoid a path through your heart to ground. Redundant precautions are prudent when working with something that can permanently kill you.


I built the circuit on a small piece of prepunched phenolic circuit board. I recommend using a board with no copper cladding (pads or traces) because it is necessary to maintain adequate creapage distances on the parts of the circuit that have high voltages on them. The torroid core and batteries were fastened to the circuit board, if I recall correctly, with nylon cable ties. Make sure none of the circuit can be touched from the outside. One part of keeping the circuitry insulated by a barrier, was to carefully file the holes for the three LEDs (two white LEDs and one green LED) so that the LEDs fit into them snugly. This also helped keep the LEDs in place.

If you are going to build something like this, give the whole project a lot of thought first. Think about safety, both in building the building and testing of the project, and also in the daily use of it. The safe design, assembly, testing, and use of anything you might build that is similar to the project I have described on this page is YOUR RESPONSIBILITY, NOT MINE. Do your homework and make sure that your design is safe. I have provided you with the best information that I can. The rest is iup to you. Its your life.



Eye Safety

HOME (More Projects)
Contents 2007 Richard Cappels All Rights Reserved. Find updates at www.projects.cappels.org

First posted in October, 2007

You can send  email to me at projects(at)cappels.org. Replace "(at)" with "@" before mailing.

Use of information presented on this page is for personal, nonprofit educational and noncommercial use only. This material (including object files) is copyrighted by Richard Cappels and may not be republished or used directly for commercial purposes. For commercial license, click here.


  Liability Disclaimer and intellectual property notice
(Summary: No warranties, use these pages at your own risk. You may use the information provided here for personal and educational purposes but you may not republish or use this information for any commercial purpose without explicit permission.) I neither express nor imply any warranty for the quality, fitness for any particular purpose or  user, or freedom from patents or other restrictions on the rights of use of any software, firmware, hardware, design, service,information, or advice provided, mentioned,or made reference to in these pages. By utilizing or relying on software, firmware, hardware, design, service,information, or advice provided, mentioned, or made reference to in these pages, the user takes responsibility to assume all risk and associated with said activity and hold Richard Cappels harmless in the event of any loss or expense associated with said activity. The contents of this web site, unless otherwise noted, is copyrighted by Richard Cappels. Use of information presented on this site for personal, nonprofit educational and noncommercial use is encouraged, but unless explicitly stated with respect to particular material, the material itself may not be republished or used directly for commercial purposes. For the purposes of this notice, copying binary data resulting from program files, including assembly source code and object (hex) files into semiconductor memories for personal, nonprofit educational or other noncommercial use is not considered republishing. Entities desiring to use any material published in this pages for commercial purposes should contact the respective copyright holder(s).