Is the cap charging/discharging too fast?

Discussion in 'The Projects Forum' started by Riado, Sep 28, 2010.

  1. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    Hello everyone, I'm new here and also very new to electronics :)

    I've build the following circuit:
    http://www.youtube.com/watch?v=Klengns3oQg
    And I've achieved exactly the same results! So far so good.

    But If we look at this video:
    http://www.youtube.com/watch?v=o9f3BoPnGu0
    That uses the following circuit:
    [​IMG]

    The second circuit I posted is the same as the one I constructed with the exception of different res and cap values.

    Why in the second circuit the LED glows and dims smoother and slower than the one I constructed? Is it to do with cap quality and how fast/slow it charges/discharges? The first one seems to have a delay when it's off before it glows again, but the second one doesn't.

    I tried to use the different res and cap values like the second circuit but it seems that my 555 doesn't like anything more that 12k or so. Do 555 timers have different specs in regards what value res we can connect to them?

    Thank you for your time in advance.
    Riado
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Actually this one is a lot simpler, a nine volt battery matches the voltages better, creating a smoother transistion.

    [​IMG]

    Looking at it it matches the PCB you are showing. It works because Pins 2 and 6 have a triangle wave 1/3 the amplitude of the power supply voltage and dead center of the power supply, and this wave is transfered to the LED. Like I said, the voltages match the LED max and min better.

    Part of the reason it works better is it is self adjusting, the oscillator matches the RC, instead of trying to match an RC circuit to an oscillator.
     
    Last edited: Sep 28, 2010
  3. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    Hi Bill, thank you for the quick response!

    Your circuit requires PNP transistors and I only purchased NPN to build the other circuit. I will purchase PNP and try your circuit and let you know. Thanks.

    Riado
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Sorry, grabbed the wrong picture...

    [​IMG]

    As you can see, transistor polarity isn't critical. Their are prepackaged Darlington's for sale, do not use them, as they have a resistor between base and emitter. It must be two discrete transistors to work properly.
     
  5. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    Thanks again, I constructed the circuit exactly like the right hand side one, but there is still a delay before it starts glowing again.

    With your circuit:
    time LED is OFF = time LED is ON (from dim to bright, then bright to dim)

    Basically I need the time off to be near nill, so as soon as it goes OFF, it starts to go bright again immediately.

    Thinking about it again, I think what I'm after is not 50% duty cycle, I think I need an adjustable duty cycle with two variable resistors. What do you think Bill?

    Thanks again.
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    What color LED?

    Go through the voltages for a moment.

    The triangle wave is 1/3 to 2/3 of Vcc (9V).

    Therefore the bottom voltage is 3V.

    The transistors are dropping 1.2V, leaving 1.8V of the 3V. If it is a single transistor, it will drop 0.6V, leaving 2.4V. This might suite your needs, but it could also kill the circuit.

    If the power supply voltage is 15V (max for a conventional 555, a CMOS can use up 18V) you might like the results more. 5V is the bottom 1/3 point, minus the 1.2V transistor drop that will leave 3.8V, which will give you exactly the results you want.

    A better approach is PWM, you can use very low power supply voltage and still achieve the same results.

    You can read up on it here...

    LEDs, 555s, Flashers, and Light Chasers (forum edition). This is my working copy, I have a released version on my blog.

    Bill's Index

    If you are interested I can draw something up.
     
  7. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    I'm really sorry for not supplying the adequate information earlier.

    I'm using an old amber LED, I don't know the spec of it :(

    The voltage might be the problem, because I'm only using 5v and don't have the option to increase it.

    Is it possible if you draw a circuit that controls the ON and OFF times?

    Sorry again for the confusion caused.
     
  8. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    No problem guy, compared to some you are verbose.

    It will take a day or so, but yes, we can draw something up (someone might beat me to it). You'll note the old hands have thier part of the world on display. This aids in aquiring parts.
     
  9. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    Hi Bill, I'm waiting excitedly for the new drawing :D
     
  10. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    Sorry, I lost track of this thread because it was so similar to another one. Wanna guess who got the drawings? :rolleyes:

    Take a look at this thread, there are some other ideas you may want to tap into.

    PWM with Hex Schmitt Trigger: I'm confused

    [​IMG]

    Wookie's post has a worthwhile schematic too,
    http://forum.allaboutcircuits.com/showpost.php?p=282227&postcount=20

    I went all the way and made a 3 way, something that could be used for a RGB diode to constantly shift colors to cover the entire spectrum.

    [​IMG]

    Basically these drawings will be incorporated in my LEDs, 555s, Flashers, and Light Chasers (blog, current edition).

    You can see my updates, which have not been released yet in my LEDs, 555s, Flashers, and Light Chasers (forum edition).

    You probably want to read the forum edition, but it has not been prettied up.

    The whole advantage of PWM is the power supply can be just above where the LED voltage drop (Vf) is. If the LED is a 3.6V unit (such as most blue or white are) then you can use a 4.5-5VDC power supply, and it will cover the full range.
     
  11. Riado

    Thread Starter New Member

    Sep 27, 2010
    6
    0
    I found out where the problem is, if I use 5v then I get a long period of low voltage (LED off), but if I use 9v I get a very nice fade in and out with nearly no gaps in between by using this circuit:
    [​IMG]

    Is there a way to achieve the 9v results, but using 5v only?
     
  12. marshallf3

    Well-Known Member

    Jul 26, 2010
    2,358
    201
    Might take the ratio of C1/R1/R2 down a bit?
     
  13. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    The reason these designs work (shown in post #11) is the signal is 1/3 to 2/3 feeding the transistor. For a 9V battery this means 3V to 6V. Using the second schematic as a reference the emitter of the transistors are going from 1.8V to 4.8V. This is close enough to a red LED Vf of 2.2V to look pretty close.

    Yes, you can use 5V, but for technical reasons it will have to be a CMOS 555 such as a 7555 or TLC555 (available from Radio Shack) The design would have to be changed dramatically to handle the completely different voltages, and the complexity goes up.

    The PWM scheme shown on post #10 using a 556 (due to the low voltage you would have to use a CMOS version or a 2 resistor conventional 555 version) will work much, much better than anything analog, with a much brighter and more controllable range of brightness.

    I've been down with an upper respiratory infection for the past week, and am working besides. I'll have to get back with you catch as catch can, but I'll show examples of both. One will be that 1st drawing in post 10 modified to use 555s for the same effect. I think I can avoid confusing your thread this time. :D
     
  14. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    OK, I started a schematic and mostly drew it before I realized I was wasting my time.

    Here are the problems. 5VDC doesn't offer much head room. A red LED takes around 2.5V. A Darlington transistor (which is needed for high resistances to minimize loading) uses another 1.2V. This leaves us with 1.3V to swing the voltages, not much room at all, especially in the world of analog. With this little room even minor variations in components cause major variations in performance.

    Worse, a standard 555 top voltage output is less than 1.2V of the power supply. CMOS doesn't do this, but it has weak drive.

    All of this has forced me to throw out this design as too cumbersome.

    [​IMG]

    Figuring out the resistances for R2 and R3 is complex, it has too many parts, and will only work for a red LED. So back to the drawing board, especially since I think I have a different approach. I suspect for an analog approach the red LED limit is here to stay, since other colors use even more voltage.
     
  15. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    This circuit will work with 5.0V, and do what you want with a red LED. The LED will go completely out for a very short time.

    I showed both CMOS (7555) and regular (555) versions. The slight differences between the two types do make a difference at low voltages.

    [​IMG]

    Next I'll post a schematic using the PWM circuit and 555/556/CMOS versions.
     
  16. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    OK, this is going to be in my 555 article, but I'll post it here first. OP hasn't been back in a while, so there is no telling if he'll ever see it.

    [​IMG]

    It will work at 5V, or even 4.5V. It will power any color LED, and is basically my preferred design.
     
Loading...