36 LED Chaser/Bounce Pattern

Discussion in 'General Electronics Chat' started by Lil_Fox, Feb 15, 2013.

  1. Lil_Fox

    Thread Starter New Member

    Feb 15, 2013
    3
    0
    Is it possible to expand the Cylon/Knight rider chaser seen here: http://forum.allaboutcircuits.com/blog.php?bt=684

    I'd like to have 36 LEDs that chase 1 thru 36 then back 35 to 1. Also can it be powered by a 12v source instead of 9?

    Thanks for any help!
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Yes, and yes. Basically look at the data sheet for the LM4017, it shows the basic concept I used in my drawing. The only thing I did different than theirs is use diode AND gates. In case you missed it I am the author of the blog an the article LEDs, 555s, Flashers, and Light Chasers.

    There are other ways though that would be worth exploring. I used diode to steer the current so one LED can share two states on the logic. A simple shift register could do this pretty well too.

    One of the problems you are dealing with is numbers. 36 LEDs means you have to have around 73 permutations. That is a lot of outputs, and a lot of diodes (but diodes are cheap). This means you are fighting a level of wiring complexity that most people find intimidating, especially beginners. I do not know your skill level, but I assume a beginner to intermediate hobbiest.

    Another approach (that I would be absolutely no help with) is to use a couple of µC chips, such as PIC or Arduino's. The fact these chips can be programed means you could eliminate a lot of side components, but the number of outputs required suggests you would need more than one just to handle the numbers.

    So it comes down to this, how serious are you about this project? It is going to be a bit more work than I suspect you originally thought, but is very doable.

    Occasionally I help folks out on the side with design work if I am convinced I am not wasting time, and right now time is in short supply. I do not respond to technical requests via PM, but if PMed pointing to a post I will do my best.
     
  3. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    One approach is to use 3 stacked LM3914s, with a triangle wave driving them.
     
  4. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    A fast ramp for the movement and a slower ramp for the bounce distance?

    I'm trying to think of an easy way for it to start at 100%, bounce back to 50%, bounce back to 25%, bounce back to 10%, then restart the sequence. I'm not sure if that is what the OP is looking for, however.
     
  5. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    A simple back and forth sweep of LEDs. I am intrigued by the LM3914s, since they are easy to get. Analog approach vs. digital, but it would work. I wonder if a sawtooth ramp from a 555 would be OK?
     
  6. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    Interesting enough, I can do 36 LEDs with a PIC 16f54.
    The oldest, most basic PIC.

    Wiring them in a matrix, 12 I/O, 6x6 = 36.

    Even the chaser program can run inside the 16f54.

    This technology is from the early 1980s, it's so old already that not many people are remembering it.

    A 16F54 is very easy to program. 52 pages documentation all together. Cost is about 50 cents.

    My work on the weekend. Putting these serial LED display PCBs to a condition I can sell them.

    Think- you can even use LED display driving chips. Who is saying these individual LED display segments have to exist on a 7seg display?

    Without to use a controller, OP will end up with a fairly complicated circuit anyway.

    Driving displays without resistors is so much oldfashion. I did not invent it, neither anyone on the internet.

    It was done in the early 1980s, having these displays in pocket calculators.

    The technology never really caught on, LCDs came up, and it all disappeared. The 16f54 still exists as super simple entry level microcontroller.

    On the one hand, people complain about this controller, and say it is so much out of date and can't do anything.

    On the other hand, effervescently complicate circuits with many transistors, and wires.

    Maybe one day you will accept the 16f5X as standard IC here to stay same as the 74XX and 40XX series.
     
  7. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    OP could eliminate ALL side components. Use a coil to clock, 100uH will do, and one 4.7k resistor.

    I have these PCBs (blank) on offer for 2.99, a K150 programmer also costs 10 dollars only.

    http://www.ebay.com/itm/1PCS-PIC-US...le-r-Programmer-K150-ICSP-cable-/251122310653

    What's the risk? 50 pages to read for the controller, 10 dollars for the programmer module.

    If people don't use this technology, it really can not be made easier or cheaper.

    All these questions on the internet about crystal loading capacitors are superfluous as well- use a coil.
     
  8. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I would be interested in seeing that. We should start a separate thread for it though, and show a link to it from this one.

    I have a PICKIT III I have not learned to use though, so the bare bones is all I need. It is a long learning curve for me.
     
  9. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    I made such chasers when I was a young boy. Including to wrap tinfoil around a 7805 (getting very hot). There was nothing beneficiary coming from dealing with these 7805.
     
  10. SPQR

    Member

    Nov 4, 2011
    379
    48

    Can you help a noob?

    By "use a coil" do you mean like an RC circuit, but the capacitor is replaced with an inductor?

    Pros and cons vs an RC circuit?
     
  11. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    No. The coil is connected between OSC1 and OSC2 with no extra components. It works fine, have used it many times.

    Some larger coils need a small capacitor but not all, most work. Depends on the making of the coil (material and turns).

    A 1000uH coil gives about 650 KHz.
    A 22uH coil gives 14 MHz but I could be wrong with that.

    The pro: Only one component. Wide frequency range, including between 600 KHz and 3.5 MHz, crystals are uncommon for that even if they do exist.

    The con: It is not mentioned in the official datasheet.

    I have seen it used in at least one microcontroller circuit in the early 1980s.
     
    SPQR likes this.
  12. SPQR

    Member

    Nov 4, 2011
    379
    48
    OK, I see now.
    It's the clock in and clock out pins on the 16F54 - and particular to that chip.

    I thought there was a new type of basic electronic oscillator circuit I needed to learn about!:D

    Thanks!
     
  13. takao21203

    Distinguished Member

    Apr 28, 2012
    3,577
    463
    It works probably with most PICs and most microcontrollers. I have used it with some other 16Fs as well.

    Not that I don't have crystals available here :)

    Crystals are only need for an accurate timebase.
     
    SPQR likes this.
  14. Lil_Fox

    Thread Starter New Member

    Feb 15, 2013
    3
    0

    Yes and no on the skill level. I'm quite familiar with delicate soldering and expansive wiring projects, it's the chips and resistors and how they work together that I'm rather clueless about. My thought on how to go about it was to cascade the 4017 chips like shown in Fig. 11.4 of your blog post and build in the bounce back like the other diagram. If that works then my next question is does changing to 12v affect the type of resistors needed? Also what wattage of resistor?
     
  15. thatoneguy

    AAC Fanatic!

    Feb 19, 2009
    6,357
    718
    How wide is the Control pin response?

    Generate a sawtooth wave, then modulate it with control? The amplitude change is the difficult bit so the "range" of bounce changes after going through the 3914.

    Whacky idea... 2 555s, one generating the sawtooth for movement, another clocking a digital potentiometer set up as voltage divider for amplitude feed to the 3914?

    I'd use a PICAXE, but the analog ways are interesting.
     
  16. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    I would not even mess with the control pin, the fundamental fact is a sawtooth is what you are going to get, and if I wanted to change the amplitude I would probably use a amplifier (as in Op Amp) followed by a simple attenuator.

    But even with a 555 there are ways to make a linear ramp function, I am just speculating how important it is. I suspect the answer is not very.

    I have never used a LM3915. Tried to get some yesterday from Tanner's (a local Mom/Pop electronics components shop), they were out. I may try doing this, it has been a while since I've done something new. I do love do nothing circuits, wid da blinky lights.
     
  17. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Don't get a 3915. Get a 3914. 3915s are nonlinear.
     
  18. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    A microcontroller with 12 output pins would be a lot simpler, IF you are good at programming.
    Yeah, I know, I'm the one who suggested the LM3914. That's because I'm not proficient at programming (yet).
     
    Wendy likes this.
  19. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
    Which also sums up where I am at. I am going to start drawing a preliminary schematic pretty quick using the LM3914.

    I will start a new thread so as to not hijack this one, and point to where it is.
     
    Last edited: Feb 17, 2013
  20. Wendy

    Moderator

    Mar 24, 2008
    20,766
    2,536
Loading...