Home     Contact     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science     RSS     Terms of services     Privacy policy  
 Home      Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science   

12 January 2010
Author: Giorgos Lazaridis
Miniature 2.4 Volts LED Flasher

The tiny LED flasher under heavy test. The circuit runs day and night with two 1.2volts rechargeable batteries, since the 11th of January 2010, 20:00.

First of all, something about the 3V LED flashers. As everybody knows, AA and AAA size batteries have 1.5 Volts per cell. Thus, two of them connected in series will deliver 3 volts. Using normal or alkaline batteries, all 3 volts LED flashers should operate normally. If you happen to have a rechargeable battery on your hand, it is most likely that it does NOT have 1.5 volts per cell. Instead, they usually come at 1.2 volts. This is rather annoying especially when a 3 volts flasher needs to be powered. Usually they do not work. I tried 3 different circuits. Either they did not oscillate at all, or the LED power was rather low. So, i decided to make another circuit that operates also at 2.4 volts.

The circuit is ideal for safety rear flashing light on a bicycle. This is the main idea after all. That's why i used 7 high brightness LEDs. You can connect of course as many LEDs as you want in parallel, or even use one 10mm ultra bright LED, but the 7 LEDs can make a nice long flashing line. It can be seen from 100 meters away very clearly! Of course, it could be also used for decorating purposes or for security light when you go for a walk at night.

The Circuit

8 components! The complete circuit is made of 2 transistors, 5 resistors and one capacitor. It can be made extremely portable. The batteries will be significantly larger than the PCB itself. For more portability, coin-type batteries can be used instead. The power that this circuit needs is extremely low, and even very small batteries can have quite a long life.

On the 11th of January 2010, i turned on this circuit with a pair of rechargeable batteries, 2500 mA fully charged. I run this circuit day and night to see how long the batteries would last. The circuit kept running until the 23rd of March! That is 71 days of continuous operation, more than 1700 hours! Of course the LEDs were very dim the last weeks and could not be sued for security light, but it still flashed.

To make this application legal for on-road use, i had to check the Vienna Convention on Road Traffic. For the rear signaling light it says:

A light without a steady mode is considered approved only if it flashes at a constant rate of between 60 to 240 flashes per minute and has a luminous intensity of at least 4 candela.

The second pard is covered by far, as each LED is ultra bright with luminosity 11000 mcd. As for the first part, it describes the pulsing frequency. It must be between 1 to 4 Hz. My first design was around 1.2 Hz pulsing frequency. I chose such a low frequency, because i wanted to extend the battery life as much as possible. The result was rather... boring... The LEDs were flashing once per second. And believe me, it really does not sound as slow as it is. So, i decided to increase the pulsing rate up to 2.6 Hz, a little bit faster than the car's flashing direction lights.

Another circuit parameter that i experimented was the pulse positive period. This is the time that the LEDs will remain ON. This was rather difficult to choose. I had to find a setting that would keep the LEDs ON long enough to be seen, but not too long because they would drain the batteries faster. I ran some experiments in a dark road about 100 meters long. After many tries, i decided to keep the positive period between 10 to 14 mSec.

Bill Of Materials
R1Resistor 220 Ohm 1/4 Watt 5% Carbon Film 
R2Resistor 220 Ohm 1/4 Watt 5% Carbon Film 
R3Resistor 2.2 KOhm 1/4 Watt 5% Carbon Film 
R4Resistor 4 MOhm 1/4 Watt 5% Carbon Film 
R5Resistor 1 Ohm 1/4 Watt 5% Carbon Film 
C147 nF Ceramic Capacitor
T1BC327 Switching and Amplifier Applications PNP Epitaxial Silicon Transistor 
T2BC547 Switching and Applications NPN Epitaxial Transistor 

Relative pages
  • Basic transistor circuits
  • The LED theory
  • The transistor theory of operation
  • How to make a light / dark activated switch - 3 different circuits under the microscope

  • Comments


      Email (shall not be published)


    Notify me of new posts via email

    Write your comments below:
    BEFORE you post a comment:You are welcome to comment for corrections and suggestions on this page. But if you have questions please use the forum instead to post it. Thank you.


  • At 1 May 2015, 10:10:32 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @hamid nasr drive them through a proper transistor

  • At 24 April 2015, 6:30:12 user hamid nasr wrote:   [reply @ hamid nasr]
    • hi !
      i have made a flashing circuit ,but based on 555 ic . when i add more then two smd led the light gets very dim. any idea if i can add 16 or seventeen smd led on this circuit , i run it on 12 volt. lipo.
      appreciate to reply

      hamid nasr

  • At 17 December 2014, 8:25:11 user Colin Mitchell wrote:   [reply @ Colin Mitchell]
    • A "47" or "470" by itself means 47pF.


      A "47" by itself means 47pF.

      A "470" means 470 pF

  • At 17 December 2014, 7:02:16 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Newbie here C547 is ok, do id the 5M resistor. The 47 aec is something i do not know. A "47" or "470" by itself means 47pF. For 47nF the capacitor should be marked "473"

  • At 17 December 2014, 2:55:59 user Newbie here wrote:   [reply @ Newbie here]
    • @Colin I really want the 2.4v.Is it okay that the store gave me a C547B transistor instead of BC547 and a 5 Mohm resistor instead of 4Mohm resistor,they told me its the same?I also want to know if 47/aec ceramic capacitor is same with 47nf?? Im really dont have a knowledge about electronics,but im interested doing this project..

  • At 16 December 2014, 13:24:28 user Colin wrote:   [reply @ Colin]
    • "mine only stays on,doesn't blink.I dont know whats wrong.any tips"
      The first tng to do is increase the voltage to 4v5.

  • At 16 December 2014, 4:36:37 user Newbie here wrote:   [reply @ Newbie here]
    • mine only stays on,doesn't blink.I dont know whats wrong.any tips

  • At 26 January 2014, 5:49:05 user LedPowered wrote:   [reply @ LedPowered]
    • Hello,

      I was wondering whether you have found the formula to change the frequency and duty cycle of the oscillator by altering the C and R values. It would be of great help. Thanks.

  • At 26 December 2013, 10:48:29 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Blackbird Its R3 that you want to change. There can be a formula extracted out of this but i have not gone into the trouble thus far :D

  • At 18 December 2013, 18:08:10 user Blackbird wrote:   [reply @ Blackbird]
    • I was considering building this great circuit, but before I buy it I have a question: which resistors do I have to exchange to increase the frequency of the blinking? I was thinking about something like 30-50Hz. Or is there a formula to calculate the resistance needed?


  • At 27 November 2013, 23:01:10 user Anonymous wrote:   [reply @ Anonymous]
    • Just Ask Yourself "Why Is It Have To Be between 1 Hz and 4Hz ?" Who Gives Them the right to make it law? Why must have more power than another? the same thing goes with drugs like "weed" any one who has smoked it has never killed any one yet but its illegal, but u can go buy penicillin (its in banana's... its stops the heart in high dosages if in blood stream or in stomach) yet you can go buy it at a pharmacy for cheap and a lot of it too

  • At 19 July 2013, 21:35:52 user yusuf ishaq wrote:   [reply @ yusuf ishaq]
    • Hi Giorgos,
      How exactly does this circuit oscillate? The interaction of both NPN and PNP BJT is kinda fuzzle to me.

  • At 29 September 2012, 11:19:30 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Colin Mitchell so the video is fake? It worked for me 70 days 24/7. Maybe you connected it wrong. Try again.

  • At 29 September 2012, 9:56:45 user Colin Mitchell wrote:   [reply @ Colin Mitchell]
    • The circuit DOES NOT WORK.
      White LEds need between 3.2v and 3.6v
      The circuit will only work on 3 x 1.2v cells = 3.6v supply.

  • At 28 September 2012, 18:39:57 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @Jess no you cannot use this circuit. You need either a boost power supply with constan current output,(i dont have something to show you though) or a simple constant current like these:


  • At 27 September 2012, 5:41:43 user Jess wrote:   [reply @ Jess]
    • Hi Giorgos,
      I wan't to create a steady LED Lights using 2-pcs. AAA batteries. Can I use the circuit "Miniature 2.4 Volts LED Flasher" by removing the transistors?
      Or do have other simple circuit that uses 2-AAA batteries for a 7-LED steady light source.


  • At 25 March 2012, 13:04:33 user Giorgos Lazaridis wrote:   [reply @ Giorgos Lazaridis]
    • @mohdosman post in the forum some pictures to see what is wrong. Maybe you've used other leds than white.

  • At 25 March 2012, 13:01:01 user mohdosman wrote:   [reply @ mohdosman]
    • The LED is not flashing up.
      Its is in on mode only.
      suggest me the solution for flashing the LED's.

  • At 25 January 2012, 19:47:57 user Kammenos wrote:   [reply @ Kammenos]
    • @DanD i do not have such a circuit ready, and i would reccomend making it with a low voltage microcontroller instead.
      As for the alkaline rechargeable, frankly speaking this is the first time that i hear. i will check them out though for i hate this 1.2V

  • At 25 January 2012, 17:15:14 user DanD wrote:   [reply @ DanD]
    • @Kammenos @Kammenos
      I will design it using 4 "aaa" batteries as a power source. Is there a circuit already designed to to the chase or sequencer where one lights up, stays lit then the next one etc. until all are lit?
      Are you familiar at all with the rechargeable alkaline batteries? ( PureEnergy in Canada makes them and they are supposed to output 1.5v)

  • At 25 January 2012, 5:55:50 user Kammenos wrote:   [reply @ Kammenos]
    • @DanD @DanD No this is not possible with this circuit. This is only a low power LED flasher.

  • At 25 January 2012, 1:07:08 user DanD wrote:   [reply @ DanD]
    • Just what I'm looking for, however I would like it to do two things - chase or sequence the led's so they light up one at a time (rapidly) and when all are lit go dark and start the sequence again and again for about 100 seconds then shut off, and the start cycle is activated by a key-chain remote controller. Is this possible?

  • At 6 January 2012, 19:40:58 user Kammenos wrote:   [reply @ Kammenos]
    • @Phillip M. Feldman The circuit runs at 3V, and each LED draws about 15mA (if i remember well). 7 LEDs means 105mA, which eventually means about 100mV voltage drop on the resistor. The resistor was used to further decrease the power used by the circuit. It was an experimental model.

  • At 5 January 2012, 1:58:30 user Phillip M. Feldman wrote:   [reply @ Phillip M. Feldman]
    • Is the 1 Ohm resistor really needed? I'm surprised that such a low resistance would have much effect.

  • At 5 June 2011, 13:40:34 user Kammenos wrote:   [reply @ Kammenos]
    • @Fung two LEDs in series would require double the voltage to light, so the would not light at all.

  • At 5 June 2011, 11:28:58 user Fung wrote:   [reply @ Fung]
    • Would the power delivered to the LEDs be not enough if connected in two by two (ie 2 LEDs in series, then connect them in parallel with other pairs)?

      Also, after viewing the image concerning R4, the actual value of that resistor is 4.7Mohm, not 4Mohm. I found that the 2nd band is violet but not black.

  • At 24 March 2011, 8:09:52 user Kammenos wrote:   [reply @ Kammenos]
    • @Fung theoretically, you can connect as many as the transistor can hold. But the brightness will fade because of the battery capacity.

  • At 24 March 2011, 7:38:29 user Fung wrote:   [reply @ Fung]
    • How many LEDs allowed to connect in parallel in this circuit?

      I changed the value of R5 to 2.2ohms (the smallest value of the resistors that I have), and number of LEDs changed from 4 to 8, can I load more LEDs?

  • At 4 January 2011, 20:41:41 user Kammenos wrote:   [reply @ Kammenos]
    • This is one part of the case. there is something else. LEDs have a huge voltage drop compared to the power supply (2.4). Actually, most of the voltage through R5 drops across the LEDs. Without R2, the capacitor would never exceed the remaining voltages.

  • At 4 January 2011, 20:26:48 user Steve wrote:   [reply @ Steve]
    • Thanks for the quick response but I am still confused. I understand the circuit to work like this: Power is applied, T1 and T2 are open, C1 begins to charge through R2 and R5(plus LEDs). While C1 is charging the LEDs are on since most of the current to charge C1 comes through that path since R2 is much higher resistance. Once C1 is adequately charged T1 closes and therefore T2 closes which discharges C1 through T2. When C1 is adequately discharged T1 opens therefore opening T2 and the cycle starts again. With that as my understanding I think C1 mostly charges through R5 and that makes R2 irrelevant. Now as I\'m writing this I may have answered my own question. As C1 nears fully charged it draws less current. Perhaps when the potential across the LEDs is not enough to light them up they act as open circuits. This would mean that R2 is needed simply to top off C1 after the potential across the LEDs becomes too low. Is this the case?

  • At 4 January 2011, 19:47:50 user Kammenos wrote:   [reply @ Kammenos]
    • R2 is used to charge the capacitor C1, when T2 is in cut-off

  • At 4 January 2011, 19:43:44 user Steve wrote:   [reply @ Steve]
    • Can you please explain why the R2 to V+ connection exists in this circuit? I am new to electronics and I can\'t figure out what purpose R2 to V+ serves. The best I can come up with is that it reduces the current through the LEDs, however, R2 is so high in comparison to R5 that the decrease in current seems irrelevant.

  • At 3 January 2011, 7:11:56 user Kammenos wrote:   [reply @ Kammenos]
    • The LEDs have probably 3 volts (or higher) forward voltage. They will not be full bright with no way. First of all, the 10 ohms resistor is too big. You need to have 1 ohm. To have full brightness, you can either increase the supply or use LEDs with less forward voltage.

  • At 2 January 2011, 16:40:24 user Fung wrote:   [reply @ Fung]
    • I use 2.6V to power the circuit (2xAA, but only 1.3V each), 10R for R5, and 4 white LEDs are used. But the LEDs are not bright enough, if R5 unchanges, should I increase the supply voltage?

  • At 11 November 2010, 10:31:07 user RICK wrote:   [reply @ RICK]

  • At 24 March 2010, 6:27:42 user Kammenos wrote:   [reply @ Kammenos]
    • Well, yesterday (March 23) the batteries died. The circuit ran continuously for 71 days, more than 1700 hours! The last weeks the LEDs were too dim for the circuit to be useful though, but it still was flashing.

  • At 13 March 2010, 9:34:47 user Kammenos wrote:   [reply @ Kammenos]
    • It still flashes. Really. The LEDs are significantly dimmed, but it still flashes.

  • At 13 March 2010, 2:28:56 user Tim wrote:   [reply @ Tim]
    • So what was the final battery life determination? Today is March 12th, 2010.

  • At 30 January 2010, 8:01:07 user Kammenos wrote:   [reply @ Kammenos]
    • I still runs. I will post the results when it stops running.

  • At 29 January 2010, 20:49:02 user Mihalis wrote:   [reply @ Mihalis]
    • How long the batterys hold?

    reddit this Reddit this

     HOT in heaven!

    NEW in heaven!

    New Theory: AC electric motor working principle

     Contact     Forum     Projects     Experiments     Circuits     Theory     BLOG     PIC Tutorials     Time for Science     RSS   

    Site design: Giorgos Lazaridis
    © Copyright 2008
    Please read the Terms of services and the Privacy policy