Need help

Discussion in 'Homework Help' started by thuyvd78, Aug 4, 2005.

  1. thuyvd78

    Thread Starter New Member

    Aug 4, 2005
    5
    0
    Hi
    This is my small exercise to build a game circuit, Game Roller Or Chase Circuit with multisim v7.
    my circuit includes:
    1 555 timer chips, I set frequency = 1kHz
    1 74LS93D, count to 16
    1 74LS154, demultiplexer, to select output lines
    16 LED
    When one output line is selected, the LED of that line will be turned ON.
    I have trouble with pulses. I dont understand why if one line is ON, another line still has an edge of pulse.
    Can anyone help me fix this problem.
    Thank in advance
     
  2. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Hi thuyvd78,

    I took a look at your circuit and I think I know what is causing the glitches.

    The glitches on the output lines of your 4 to 16 line decoder are due to the fact that the 4-bit counter is a ripple counter. That means that the outputs from counter are subject to some delay between the transitions whenever the counter changes count. Since several of the counter's output lines change during some of the counter values, there is a brief period of time when the counter's output is in between count values. Your decoder is fast enough to respond to these brief intervals when the counter is transitioning between counts and so you see the resulting glitches.

    One way of dealing with this is to take the output of your 555 and use the positive going edge to update your counter and then use the negative going edge to clock a 4 bit latch (for example a 74LS75) that you add in between the output of the counter and the input to your decoder. Be doing this you give the counter's output lines time to settle before you present them to the decoder via the 4 bit latch. If the counter and the 4-bit latch are clocked by the same edge-transition you can just use an inverter to invert the output of the 555 and feed it to the 4-bit latch.

    Hope my description of the technique is clear.

    Questions are welcomed.

    hgmjr
     
  3. thuyvd78

    Thread Starter New Member

    Aug 4, 2005
    5
    0
    Hi hgmjr

    I tried to add a chip 74LS75 between my counter and decoder chips, but I still got the same problem. I think it could be my frequency which is currently too high causes LEDS running too fast ?
    Thank you for your help
     
  4. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    OK.

    Let's try something a little different.

    Get rid of the 74ls75 and replace it with a couple of 74ls74 d-type latches. This particular latch requires a positive edge-triggered clock input. The more I thought about it the more I think the 74ls74 is a better choice since the 74ls75 is a level triggered device.

    Now connect the output from the 555 to the clock input of the 74ls74.

    Since the 74ls93 changes counts on the high-to-low transition of its input clock and the 74ls74 triggers on the low-to-high transition this should work nicely.

    hgmjr
     
  5. thuyvd78

    Thread Starter New Member

    Aug 4, 2005
    5
    0
    Sorry I tried to build the circuit with 74LS74 , but the same problem happened.
    :(
     
  6. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Something looks a bit fishy with the 74ls74 schematic symbol. The little circle at the clock input makes me wonder if the model simulation is correct or not. You can check this using the virtual scope probe. I expect the input to be latched on the output at the low to high transition of the clock input. The dot on the clock input usually indicate that the device is clocked by a high to low as in the case of the 74LS93 counter in your circuit.

    If after checking it with your virtual scope you find that the 74ls74 is latching on the negative clock edge then all you should need to do is invert the clock input to the 74LS93.

    Isn't this fun. I wish I had had circuit simulation software when I was starting my career in electronics. It is so cool to be able try one circuit configuration after another without ever having to order any parts.

    Good Luck,
    hgmjr
     
  7. thuyvd78

    Thread Starter New Member

    Aug 4, 2005
    5
    0
    Hi hgmjr

    I relized that 74LS47 doesn't work well on simulation. Maybe it happened to the decoder 74LS154?.
    Finally, I think I should build another circuit. This one is too hard for me :(

    Thank for your help Hgmjr!
     
  8. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    joejester,

    It is interesting to see that TINA-TI is exhibiting similar behaviour as thuyvd78 is observing in his multisim similation.

    thuyvd78,

    You could consider replacing the 74LS154 with two 74LS137 decoders. The advantage of the 74LS137 is that the state of its input lines can be latched. You can feed the three low order bits of your counter to both of the 74ls137 and then use the high-order bit of your 74LS93 counter to enable the appropriate decoder by feeding it into the G1 or G2 enable lines..

    The GL input to the two 74ls137's can be feed by the inverted version of the clock feeding the 74ls93.

    hgmjr
     
  9. hgmjr

    Moderator

    Jan 28, 2005
    9,030
    214
    Very impressive joejester!

    hgmjr
     
  10. thuyvd78

    Thread Starter New Member

    Aug 4, 2005
    5
    0
    Wonderful circuit, JoeJester!

    Thank you guys for help!
     
Loading...