Alternating LED strobe with no ICs

Discussion in 'The Projects Forum' started by jptrsn, May 3, 2008.

  1. jptrsn

    Thread Starter Member

    May 3, 2008
    10
    0
    Hi,

    I'm new to the boards, and this is my first post, so please be gentle!

    I'm trying to create an alternating LED strobe flasher, similar to many others that I've read about in other posts on this site. However, I wanted to try and do it using only parts that I had available for free from discarded electronics (specifically, one answering machine and two 1-channel remote-control trucks).

    I managed to piece together a few different circuits to generate my alternating signals using a two-transistor two-capacitor flip-flop, and a two-transistor LED flasher. I fiddled with it on the breadboard, and managed to get basic functionality, and it seems to work pretty well. It really is just a prototype to see if my concept was possible; there are still many things I want to implement before this project is complete.

    However, I'm pretty sure there are some flaws with the way I've done gotten where I am, based on how little I know about the actual goings-on of electrical circuits. This is a hobby I touched briefly on about 10 years ago, and I've only just begun again. I've been researching and reading quite a bit, but that just makes me feel more certain that there are problems to be solved.

    I have attached a schematic I put together in *ugh* Word, but it'll have to do.

    Would anyone be willing to make suggestions as to how I could improve it? I would like to be able to drive about 1A on each channel, and I'm pretty sure I've got the parts. I can post what I've got, if it would be helpful. I plan on using an LM7805 to regulate the 14V battery power down to 5V, and I'll adjust the resistors to suit. What's I've read suggests that sinking 1A is much more efficient, and that may be the way I go with it, but I'd love suggestions.

    However, I want to avoid using ICs, because I don't want to spend any money on the actual circuit. I have been toying with the idea of activating the circuit through a touch-switch (maybe with a darlington and a bistable flip-flop?) but I haven't really put any research into that idea yet. I'm really willing to scale the final project to whatever size my parts can handle.

    Sorry for the long post... Thanks for any help!

    P.S. Please ignore the isolated schematic symbols at the top: they're just my library of symbols.
     
  2. iONic

    AAC Fanatic!

    Nov 16, 2007
    1,420
    68
  3. jptrsn

    Thread Starter Member

    May 3, 2008
    10
    0
    That circuit forms one part of the overall circuit I have designed. It only provides part of the functionality that I am looking for, and I paired it with an LED strobe circuit to get the LEDs to strobe as they flash alternately.

    My circuit strobes the left-hand LEDs, then strobes the right-hand LEDs, and so on - the above circuit only illuminates alternate LEDs.

    Take a look at the schematic: it's the upper left-hand side.
     
  4. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    R6 and R7 should connect to the collectors of Q1 and Q2, not the bases.

    D3 doesn't do anything except maybe steal current from the base of Q5. D3 should be removed.
    Q3 and Q5 should have current-limiting resistors in series with their bases because the base current of Q5 is about 380mA now.
    The collector of Q3 should have a collector resistor to 0V.

    LEDs have a max allowed current rating. if the current is higher than their rating then they burn out.

    1A is a lot of current for this simple circuit unless it drives many more LEDs or if the LEDs are high current ones.

    If the battery is 14V and it feeds a 7805 regulator then the regulator has 9V across it. With a load of 1A then the heat in the regulator is 9W which makes it too hot without a pretty big heatsink.
     
  5. jptrsn

    Thread Starter Member

    May 3, 2008
    10
    0
    I will have to try out those changes tomorrow and see what happens. D3 was the original location of the flashing LED in that circuit, and for some reason that circuit doesn't working without D3 in place. Perhaps by adding current-limiting resistors to the bases of Q3 and Q5, something will change? Any suggestions on what value resistor I should use? My guess is that I would want to lower Q5 to about 20mA (that's a shot in the dark), so I would want to use a 15 ohm? I really don't know for sure...

    I'm familiar with the current limitations of LEDs, and I was proposing the 1A figure as a total to drive up to 50 LEDs for each side of the flasher (100 LEDs total, but only half would be flashing at any time). I wonder if I could add some heavier-duty NPNs to sink that type of current independently of the LM7805? I have a couple KEC 5610 NPN 80V 5A transistors: would they work in that fashion?

    Thanks for the suggestions - I'll let everyone know how it turns out.
     
  6. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    If D3 is an LED then it won't light because the base-emitter voltage of Q5 is only 0.7V.

    Q3 won't work without a collector resistor to 0V. Try 10k.

    Try 10k for the series base resistors for Q3 and Q5.

    The circuit is designed for low current in the LEDs. It would need to be redesigned to provide much more base current into Q8 and Q9. Maybe additional transistors would be needed to provide enough base current.
    I couldn't find a datasheet for the KEC5610 PNP transistors.
     
  7. jptrsn

    Thread Starter Member

    May 3, 2008
    10
    0
    I'm pretty sure the datasheet is here: http://www.digchip.com/datasheets/parts/datasheet/300/2N5610.php

    I've updated the schematic: how does it look now?

    Thanks for the suggestions. I had originally placed a 10K resistor in place of D3, but found that the circuit stopped working if I removed the LED (which I did not want in that location).

    How did you arrive at the figure 0.7V for the base of Q5? I'm trying to understand this stuff more; would you be willing to explain the logic for your suggestions?
     
  8. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    The transistor is a 2N5610 made by "KEC" (Korean Electronics Corporation.) They are NPN. Hfe is 200 max.
     
    Last edited: May 4, 2008
  9. Audioguru

    New Member

    Dec 20, 2007
    9,411
    896
    If you design a circuit so it works only if the transistors have their max hFE then it probably won't work.
    I design circuits that work when the transistors have minimum gain or higher. Then all my circuits work pefectly.

    Your datasheet link says the datasheet for the 2N5610 is not available.
    I an not in Korea and my American seach engines could not find the datasheet for the 2N5610, but a short description has its minimum hFE at 70 but its saturation characteristics (when it is used as a switch) were not mentioned. It probably needs a base current of 100mA to switch a load of 1A. Your circuit would need to be re-designed for such high currents.

    R6 and R7 are supposed to connect to the collectors of Q1 amd Q2 instead of being connected to their bases.

    A silicon transistor has a base-emitter forward voltage of about 0.7V.
    A red LED needs 1.8V so it won't light if base-emitter of a transistor reduces the voltage to only 0.7V.
    A blue and white LED need about 3.5V.
     
  10. jptrsn

    Thread Starter Member

    May 3, 2008
    10
    0
    Is this circuit is designed to operate only when the transistors have their max hFE? I'm really over my head here - how would I change the circuit to operate with the transistors having minimum gain? Is there any way to amplify the voltage, rather than current (like using a mosfet)? I'm still learning about what's possible and what isn't, so the finer points of how things are possible are still beyond me.

    My original concept was to have this circuit simply generate two signals - a "left" and "right" strobe signal. I was planning on using this signal to switch power on/off using some other hardware which is capable of providing enough current to run about 50 LEDs for each channel.

    I have made most of the changes you suggested, and the circuit seems to be working better. I changed R6 and R7 to 1M in order to slow down the alternating signal (when it was using 10K, the circuit was switching at a rate of about 4-5 Hz, and I wanted it at 1 Hz). However, I think that the 10K values for R9 and R10 may be too high: when I have them in place, the circuit doesn't function. I may try lower values of resistors, and see how that affects it.

    Thanks for the help! If this thing won't run more than 6 LEDs, then I'm not too worried, it was a great learning experience! I'm still trying to research and learn about semiconductors (as I have no experience in this area), and I'm part way through the semiconductor text on this site...
     
  11. thingmaker3

    Retired Moderator

    May 16, 2005
    5,072
    6
    "Max hFE" is a value listed on the datasheet. I'm not sure why anyone would want to design a circuit near any component's maximum values.

    Try driving your 2N5610s with a two-transistor multi vibrator: http://www.allaboutcircuits.com/vol_6/chpt_5/18.html

    Use the 2N5610s in the common emitter configuration, rather than the emitter follower shown in the experiment.
     
Loading...