Dick Cappels' project pages http://www.projects.cappels.org
Return to HOME (more projects)
A 1.5 Volt, 1970's Style LED Flashing Red Caboose Marker Light

A circuit that drives a red LED from a 1.5 volt battery and simulates an incandescent light.
Duty cycle can be changed by selecting resistor values.

Photo 1. This printed circuit board is 3 cm long. The first prototype was made with through hole parts.
This is a single sided surface mount board with one jumper.

Find updates at www.projects.cappels.org


It started with email from mrpiggs. Mrpiggs is a model railroad artesian and is a direct contributor to  this web page. Mrpiggs needed a circuit that would flash an LED, but nothing he was able to find quite fit his needs. The LED was going to be mounted inside a very small model caboose, and had to be battery powered. The lower the voltage, the easier it would be to fit the batteries into the caboose along with the flasher circuit. The venerable LM3909 LED flasher came to mind, but when it flashed, the LED only emitted a tiny blink, nothing the sustained flash on a real 1970's style caboose marker light. He needed a circuit that would provide a longer flash. And while we are at it, how about making the luminance of the LED quickly fade up and then quickly fade down similar to the appearance of the real thing. Incandescent lamps take many milliseconds to reach full luminance and even longer to die down, while LEDs switch on and off virtually instantaneously, giving a distinctly different visual experience.

Movie 1. Click on the image above to see an MPG video of the LED blinking.
Notice that the duty cycle is about 35% and that the light
fades up and down, rather than abruptly switching on and off.

Remember to use your browser's back button to come back here.

Or, Even Better, see this demonstration on Youtube by clicking on the image or link below.

Image above may be subject to copyright

The resulting circuit can be customized to flash the LED at a wide range of rates, keeping the LED over a wide range of duty cycles. In addition, it can be made to allow the luminance to increase and decrease faster or more slowly as needed, and the drive current to the LED can be decreased to obtain the desired luminance. Click on Movie 1 to see what the blinking of the LED looks like with the circuit values  shown in the schematic in figure 1.

The circuit

Figure 1. Capacitvely coupled bistable multivibrators set
the LED Power Supply switching rate and the blink rate.

The circuit can be thought of as being made from three major blocks: an LED Power Supply Oscillator/Driver, a Blink Rate Oscillator, and a Power Modulator to modulate the drive power to the LED, thus blinking it on and off. The discussion of the LED Power Supply Oscillator / Driver assumes that the LED is "on" -that is the signal from the Blink Rate Oscillator is always high, and this is modeled in the circuits in the description of the LED Power Supply Oscillator / Driver  by showing the emitter of Q2 grounded.

Figure 2. The basic 1.5 volt LED Power Supply provides
pulses up to 3 volts peak with which to drive the LED.

The Led Power Supply Oscillator / Driver shown in Figure 2 is a pretty good little 1.5 volt LED power supply. An oscillator, a bistable multivibrator made of Q1 and Q2 oscillates at about 15 kHz with approximately a 50% duty cycle. There are many explanations of multivibrators on the web, so I will not discuss it here, except to say that the main weakness of this kind of oscillator is that if it ever stops, it needs to be "kicked"  back into oscillation. A suitable "kick" can be made by disconnecting and reconnecting the battery. Don't worry, its unlikely that the oscillators will stop spontaneously. If your oscillator ever stops while power is applies, it will most likely be the result of an intermittent connection of a temporary short circuit. These oscillators are very reliable.

The collector of Q2 drives a low value resistor so this stage it can supply enough current to drive the LED. The only problem here, is that the pulses coming from the collector of Q2 are about 1.4 volts peak-to-peak, and a red LED needs close to 2 volts to operate. The trick that gets around the problem is to  alternately charge a capacitor to 1.5 volts, then place it in series with the 1.5 volt battery and and LED, so the LED can get up to 3 volts.

Figure 3. C3 is alternatively charged then discharged
to provide a high enough voltage to drive the LED.

Figure 3 shows the part of the circuit that converts the 1.5 volt battery voltage into a drive signal of up to 3 volts for the LED.  The collector of Q2 switches between ground and +1.5 volts through the 100 ohm resistor at about 15 kHz because it is part of an oscillator.

Figure 4. The 100 Ohm resistor limits the current available to charge C3.

During one half cycle of the 15 kHz oscillation, which lasts about 30 microseconds,  Q2 turns off. The collector of Q2 is connected to the end of resistor R4 that is not connected to the battery. This end of R4  rises toward + 1.5 volts and provides base current to Q3 through base resistor R5, causing Q3 to saturate and to hold the negative end of C3 at ground. The positive end of C3 charges to up toward 1.5 volts through R4 during this period. The actual voltage that C3 charges to is dependent on how much C3 had discharged during the discharge cycle.

During the charging up half cycle, the LED in figure 3 does not conduct because the voltage across it is only about 1.5 volts.

At this point, C3 is charged to up to 1.5 volts.

Figure 5. R10 limits the discharge current through the LED.

To get the LED to light up, C3 is placed in series with the LED. This happens when Q2 turns on. Assuming that C3 is charged all the way to +1.5 volts. When Q2 turns on, the positive end of C2 is connected to ground, forcing the negative end of Q2 to go to negative 1.5 volts (see photo 2).  The LED and R10 would then have 3 volts across them. The current during this half cycle is limited mostly by R10. If R10 is made 0 ohms, then a higher peak current, and a brighter LED can be obtained.

This circuit is meant to drive a red LED for direct viewing - in other words, in a case in which the viewer looks directly at the LED rather than looking at a scene illuminated by the LED. As such, in most cases, the LED should not be very bright. The luminance of the LED can be reduced by increasing either R4 or R10.  The benefit from increasing R4 is that doing so will result in a lower battery current than would be obtained by increasing R10 to obtain the same luminance. R4 can only be increased until the voltage when Q2 is off does not go high enough to turn on Q3. Once that point is found, further dimming can be obtained by changing R10. If you are not particularly miserly about battery current, you can just simple adjust R10 for the luminance you desire.

Before we leave R10 - you don't really need it. If you leave it out completely, the peak current in the LED will be limited by the equivalent series resistance of the battery, capacitor C3, and the characteristics of Q2.

Photo 2. The trace shows the voltage envelope on the cathode of the LED during one flash.
Looking at the center of the flash, while Q3 is on and C3 is charging, the anode is at about
+0.2 volts -the saturation voltage of Q3. While Q3 is off and C3 is discharging through the LED,
the anode voltage is at about -0.6 volts. The anode is held at the battery voltage, 1.4 volts
at the time this picture was taken, meaning that the voltage across the LED was about 2.0 volts.

With the voltage across the LED switching between 1.5 volts and 2 volts, how can the LED stay on all the time? It can't. The LED switches on and off about 15,000 times per second.

Won't that be hard on the eyes? No. The human visual system isn't fast enough. The overall visual effect is that the LED is on all of the time.

Since our intention is to have the LED blink on and off once or twice a second, we need to add a second oscillator to generate the timing for the blinking and flashing. That is the purpose of the Blink Rate Oscillator.

Figure 6. This bistable multivibrator runs at about 0.7 flashes per
second, which means that it completes one LED on/off cycle in 1.4 seconds.
You can change this by using different values for R8 and R9.

The blink rate oscillator determines the flashing frequency and duration of the flash. It a bistable multivibrator just like the oscillator in the LED Power Supply / Driver. In this case, the oscillation frequency is much lower -about 0.7 Hz with the component values shown in figure 1 and figure 6 for R7, R8, R9, R13, C5, and C6 in figure 1.

The time constant of R8 and C5 controls how long the LED is on during a flashing cycle, and the time constant of R9 and C6 determines how long the LED is dark during the flashing cycle. Add these two times together to get the flash cycle period.

With the component values shown in figure 1, in which R8 = R9 and C5 = C6, you would expect the LED on time and LED off time to be equal, making the duty cycle 50%. But it doesn't work out that way, because the collector of Q6 is loaded down by R6 and the base of Q4, which increases the time that Q5 is off. There is no similar effect for the other half cycle, so the output is not symmetric.

You can change the LED on time, the LED off time, or both by changing the values of R8 and/or R9. I think a reasonable range for these resistors is about 2.2k to 150k. If you need much longer or shorter periods, then you should change C5 and/or C6. Large resistor and capacitor values mean longer times.

Now that we have a signal to blink the LED one or two times a second, we need to connect this signal to the LED Power Supply Oscillator/Driver circuit. This is done with the Power Modulator. Cool name, huh?

Figure 7. Q4 is the power modulator.

It would be a simple matter to use the Blink Signal to control the output of the LED Power Supply Oscillator/Driver circuit. One way would be to connect R2 to the Blink Signal. When the Blink Signal is low, the LED Power Supply Oscillator/Driver would not oscillate, and the LED would not light up. But one of the goals of this project is to get the LED to fade up then fade down in luminance. Therefore, just stopping and starting the power supply won't quite cut it.

The power to the driver, of which Q2 is part, is controlled by Q4. If C4 were not present, the Blink Signal from the Blink Rate Oscillator would simply switch Q4 on and off. When Q4 is off, the emitter of Q2 floats and the LED Power Supply Oscillator/Driver circuit does not oscillate, so there is no light from the LED. When Q4 is saturated (on), the emitter of Q2 is connected to ground, and the LED Power Supply Oscillator/Driver circuit both oscillates and supplies power. In some parts of the range of operation of Q4 when Q4 is between being off and being saturated, the LED Power Supply Oscillator/Driver circuit operates but does not supply full power to the LED. It is the controlled transition of Q4 between being off and being saturated that allows us to fade the LED from off to full luminance and then back down.

Here is how it works:

When the Blink Signal into R6 is low, Q4 is off. When the Blink Signal goes high, Q4 begins to draw current, and its collector voltage begins to drop. As the voltage starts to drop, it causes current though C4 which opposes the current through R6 from the blink signal. This reduces the base current into Q4, and slows the transition of Q4's collector voltage between the off and saturated states. During the transition, the more current Q4 draws through its collector, the more power is available to the driver stage to drive the LED, and the higher the luminance of the LED.

This effect also occurs when the Blink Signal goes low, causing Q4 to make the transition from saturated to off.

Operation in the region between on and off occurs over about 200 milliseconds. The rate of change is controlled by the values of C4 and R6. Since R6 needs to be this low (1k) in order to keep Q4 saturated when Q2 is on, C4 needs to be this large to get a 200 millisecond transition. A longer transition would require a larger value for C4. A faster transition would require a smaller value for C4.

When operating, the average current drain is about 8 milliamps when powered from a 1.5 volt battery.

That pretty much sums up operation of the circuit.


First of all, you can make substitutions in this circuit. In fact, I hope you do so and pass what you learn on to me.

For starters, the schematics call for 2N4401 transistors, but I built a second one using 2N2222 transistors. All this circuit really needs is a good saturation voltage at the collector current - its 15 ma max. with the values shown - and a current gain of 100 or more at the relevant collector current. Lots of transistors fit the bill.  You could even use PNP transistors like the 2N2907 if you just reverse the polarity of the LED, battery, and capacitors.

C4 does not need to be 0.47 uf. It should be at least 0.47 uf, but you can use any handy value up to a couple hundred uf. The second one I built used a 1 uf ceramic chip capacitor because that was what was  available. and could fit the pads on the PCB.

C1 and C2 can be nearly any kind of capacitor, as long as they are somewhat close to .01 uf.

I built two of these. The first one was built "ugly bug" style. The parts were soldered to a copper clad board with their leads up in the air. Its really fast, but not rugged at all. The second one was on a printed circuit board. The ugly bug version was much easier to measure signals on, but the printed circuit version will probably give many years of pleasurable blinking.

The printed circuit artwork is reproduced in the GIF image below. Use it entirely at your own risk!

Figure 8. Use this layout as a guideline only. Both 1026 and 0805 resistor sizes are used.
All of the capacitors except C3, C4, C5, and C6 are 0805. C4, C5, and C6 are intended
to be used for surface mounted leaded capacitors.

If you have any questions or comments about this project, please send email to me at the address below.

By the way, you might want to check out some of the other LED projects on my web site. Check them out on the home page.

Find updates at www.projects.cappels.org

(More Projects)
Contents ©2007 Richard Cappels All Rights Reserved. Find updates at http://www.projects.cappels.org

First posted in May, 2007. Updated 22 January 2011.

Comments are welcome.  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).