LED flasher

Discussion in 'The Projects Forum' started by popnbrown, Aug 26, 2008.

  1. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    I was making a LED flasher and have a circuit built.

    It works off a 555 Timer, and connects to a bunch of LEDs (which form the number I want flashing) and those LEDs all connect to a resistor. all of the LEDs are in parallel.

    I have attached the bitmap below.

    I was just wondering if this circuit would work and what problems would I have?

    I know I do not have a source right now, the flasher was going to run off a +5V source which would connect using a 3pin cable. I am going to have to work that in later.

    Thanks,
    PNB
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
    Hello,

    Never connect leds directly parallel, each led (or leds in series) should have a current limiting resistor.
    Otherwise you will have a chainreaction of leds blowing up.
    The NE 555 can deliver upto 200 mA, so an amplfier circuit like a power mosfet for switching would bring enough current.

    Greetings,
    Bertus
     
  3. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    The 555 can't deliver enough current to directly drive all those LEDs. It would be better to divide the LEDs into small, series connected groups, with each group having its own resistor.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    With a Vcc of 5v, a 555 timer can only source 100mA, and the output voltage drops to around 3.3v.

    With that many LEDs, you'd really be better off with a higher voltage supply so that you can make LED 'strings'. A standard 555 can work with as high as 16V DC.
     
  5. kammenos

    Active Member

    Aug 3, 2008
    127
    0
    On the output of the 555 you will put a relay driver, and of course the relay itself. Then, you can connect as much load as you want. Now your load is nearly 1 Ampere.

    A relay driver can be seen here http://www.uoguelph.ca/%7Eantoon/circ/actrelay.htm

    It is the transistor and the diode and a resistor on the base of the transistor.
     
  6. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    What is the advantage of using a relay rather than driving the LEDs directly off a transistor?
     
  7. kammenos

    Active Member

    Aug 3, 2008
    127
    0
    Ok he can also drive them directly from the transistor.

    The advantage is that for a new commer to electronics, it would be easier to get a relay and test with different voltages and different currents rather than finding the proper transistor.
    A transistor relay can be found almos everywhere and needs no calculation.
     
  8. MLCC

    New Member

    Aug 15, 2008
    8
    0
    I noticed on the schematic that you had a NE566-A timer hooked up, if so, the pin diagram is wrong. But I’m going to assume that it is a typo on the schematic, and you are using a 555 timer. First, 5V is a little weak for the timer, it can handle up to 15V max (I recommend 12V). Also, when powering the 555 timer, it has a max of 200mA. Another note is that there are too many LEDs to be powered directly, so like everyone else said, you will need either a transistor or relay ( I think the transistor is better than the semi-noisy relay). A recommendation also is to hook pin 4 to V+ so that it is not accidentally triggered.

    I have a revised circuit below. 600 Ohm resistors for R3-5 reduce the current to 20mA. You may need to change the resistor value due to the 5 LEDs in one strand, and you can do this with Ohm’s Law.
     
  9. kammenos

    Active Member

    Aug 3, 2008
    127
    0
    Wouldnt this be easier?

    http://pcbheaven.com/handbook/trans_circuits/

    And you can connect as many leds as you want, with whatever power supply and as you may, parallel, in series, mixed...

    You may choose a 2n2222 transistor or something like BC182 with 1K base resistor. But also can calculate your transistor
     
  10. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    Just for clarification, is my circuit wrong because there is not enough current flowing through all of the LEDs?
    In which case how much current should I assume the LEDs need?
    I was looking at the specs of a couple LEDs, I saw a couple for 50mA and others around 20mA. I have no idea which LEDs I'm going to use, because I'm very new at this and don't know any good places to get cheap but good LEDs (RadioShack seems like a rip-off $1.5 for 2 LEDs).

    So if any has any good ideas for where to buy Yellow and Green LEDs?
     
    Last edited: Aug 27, 2008
  11. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
  12. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    MLCC, one transistor could easily drive all the LEDs. Also a common-emitter arrangement would avoid the 0.6Vbe drop of the emitter-follower.

    popnbrown, your circuit cant deliver enough current to drive all those LEDs but all you need to add is a transistor which can deliver the current. As has been mentioned, a higher supply voltage would be an advantage and the LEDs should be connected in series groups each group having its own resistor.

    The number of LEDs and resistor value per group depends on the type of LEDs you choose and the supply voltage.

    The cheapest place to buy LEDs is ebay. Take a look at some of the Chinese sellers.
     
  13. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    One transistor would be able to amplify the current enough to drive all of these LEDs correct? In which case, I could hook up the LEDs in different groups and hook them up in parallel.

    I was thinking of using a 10V source instead of a 5V. Since the board will be running off of a robot control system, it will have to rely on the power provided by the control system and the control system provides +5V for each seperate output, I only have the ability to work with 2 of these outputs.
     
  14. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    I've been doing some calculations, and if I were to use the 10V source and group the LEDs (44 of them) into groups of 4, making 11 groups. Then the transistor would power 11 groups of 4 LEDs.

    Since the 555 supplies 200mA the transistor (assuming Beta is 100) would supply 20200mA to all the groups combined. Divide that by 11 for each group and thats 1836 mA to each individual group. Since the majority of LEDs have a forward current of 20mA, theres 1756mA left for the resistor.

    Now the LEDs (the same with 20mA current) have Von at 2.2V. Which means a set of 4 LEDs needs 8.8V to be powered. Since the 10V runs through a transistor, which is not going to be in cutoff, you have Vbe=0.7V leaving 9.3V for each group. Thus leaving .5V for the resistor, since the LEDs need 8.8V.

    Now 0.5V and 1.756A for the resistor leads to a resistance of .283 ohms. I'm still new to this but I'm pretty sure there are no resistors in existance that have decimal values.

    This whole design was based off of MLCC's proposed design.

    blocco, if I were to use a common-emitter, that means hooking up the groups to the collector and attaching the Vcc to the other end of the group correct?
     
  15. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    If you have 11 groups of 4 LEDs and each led has a forward voltage of 2.2V and requires 20mA then:

    4 LEDs in series have a forward voltage of 4 x 2.2V = 8.8V

    The supply is 10V so the resistor has to drop 10V - 8.8V = 1.2V

    The value of the resistor for a current of 20mA is R=V/I
    or R = 1.2V/20mA = 60Ω

    So you have 11 strings of LEDs each requiring 20mA so that 11 x 20mA = 220mA in total which is only a little higher than the 555 can deliver.

    If I remember correctly a BC337 NPN transistor has an hFE of around 100 and is good for 300mA.
     
  16. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    Golly...I always forget that current stays the same when it flows through...but the thing is Vbe for the BC337 NPN transistor is 1.2V, and since you have 4 LEDs attached you have 8.8V. Now 8.8V + 1.2V = 10V, which means there won't be a voltage drop across your resistor.

    Specs for BC337: http://www.biltek.tubitak.gov.tr/gelisim/elektronik/dosyalar/4/BC337.pdf
     
    Last edited by a moderator: Aug 27, 2008
  17. blocco a spirale

    AAC Fanatic!

    Jun 18, 2008
    1,440
    368
    Max Vbe is 1.2V at a collector current of 300mA according to the datasheet but in practice I would expect it to be lower. There is also the issue of the 555 not being able to swing its output all the way to Vcc.

    However, if you connect the transistor as a common emitter i.e emitter to ground and collector to led+resistor string to +V then things look better. The datasheet says the collector will get within 0.7V to ground Max at Ic = 500mA and Ib = 50mA.

    Actually, 3 LEDs in series would be better than 4 as a larger value resistor of 170Ω could be used. LEDs should be driven from a current source and a larger value resistor makes a better current source which means the current to the LEDs will vary less with changes in supply voltage and also with differing LED forward voltages.
     
    Last edited: Aug 27, 2008
  18. popnbrown

    Thread Starter Member

    Aug 22, 2008
    21
    0
    I was just looking at MLCC's design. It seems like the common emitter is a good idea. I'll design a circuit calculate etc and see if I understand it.

    Just for curiousity, since I have a lot of it, would MLCC's design work. Base connected to output of the 555, collector connected to Vcc and LED string w/ resistor to emitter.

    And one other thing, for the entire project there are about 44 Leds. I'm planning on using Yellow and Green LEDs, unfortunately these two colors dont seem to have the same forward voltage. Green LEDs seem to have around 3.2V forward voltage and Yello seems to have 2.2V. Since you suggested using 3 LEDs instead of 4, I was thinking have a 3 Yellow LED w/ resistor string and a 2 Green LED w/ resistor string.

    Good idea?
     
  19. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    Yellow and green LEDs have disfferent voltages.
    Since you don't have any LEDs then buy some then design your circuit for their spec's.
     
  20. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Vf for LEDs can vary widely between manufacturers and models. In a typical batch, Vf may vary as much as 10%. You can't order them by Vf even from the manufacturers.

    I have red LEDs of various ages from various manufacturers. Vf @20mA ranges from 1.7V to 2.3V

    I also have green LEDs of various ages from various manufacturers. Vf @20mA ranges from 2.0v to 3.2v.

    I was helping a new poster not too long ago with a project; to my surprise, his red LEDs measured at a higher Vf (2.2v) than his green LEDs (2.0v) at the same current! :eek:

    A very serious consideration is the type of LEDs you will be using vs the viewing distance. Super-bright water-clear LEDs from China can be purchased very cheaply on Ebay. However, these LEDs typically have a very narrow viewing angle. This boosts their brightness rating, but makes them nearly invisible from more than 20° off center. More importantly, viewing super-bright LEDs from a short distance can lead to blindness.

    I'm not an eye doctor, and have no qualifications/certifications to give advice as to how much light from LEDs may or may not be "safe" for viewing. However, you could (and should) ask one, or do searches on the Web. Your milage may vary. Caveat emptor. Err on the side of caution.
     
Loading...