Connect with us

Help with simple LED circuit please

Discussion in 'Electronic Design' started by [email protected], Nov 26, 2003.

Scroll to continue with content
  1. Guest

    Hello people,

    I have built a small device for my sight-impaired Mother and have hit
    the wall with my limited electronics knowledge. Ma is not blind just
    can't see too well, and is a fresh air freak and I do not like her
    going to bed or leaving the house and having windows open.

    Basically what I have is a 12v regulated wall wart and 4 very bright
    LED's to indicate windows that are closed using a reed switch and
    magnet setup. I need to show the LED's when the windows are closed as
    she may not notice if I did it the other way round and they were off
    when closed.

    Ma has complained that in the night the LED's are too bright <sigh> so
    what I need is a simple(?) circuit with a light sensor (CDS maybe??)
    to adjust the inline resistance for each of the LED's so at night the
    resistance to all four is higher and thus not so bright.

    Sometimes I can't win but Mothers are Mothers and you have to love
    them right? <g>

    Can anyone please point me at some online circuit or suggestions to do
    this job?

    Thank you.
     
  2. The simple solution seems to be a too bright LED that turns on when
    the window is open. Not much of a bother during the day, but after
    dusk it becomes very noticeable. After all the windows are closed,
    the brightness is not a problem. Unless she decides she wants to
    leave a window open, even if you think she should close them. And a
    single LED on her bedside table is the only one needed. Goes off when
    all windows are closed.
     
  3. Guest

    Perhaps I didn't make myself clear. She fails to notice the OFF LED's
    so having them on when closed is the better option, that's why I am
    doing it this way.

    The four LEDs are needed so she doesn't have to go check ALL the
    windows. I didn't ask for a different solution, merely some help in
    dimming the LEDs. They have to be bright in the day time but are too
    bright at night.
     
  4. Tony

    Tony Guest

    I didn't ask for a different solution, merely some help in
    dimming the LEDs. They have to be bright in the day time but are too
    bright at night.>

    You answered your own question in your first post. Put a CDS photcell in series
    with each LED. Depending on the photocell used, some of the super brite type
    leds will glow faintly when the cell is totally dark. If not or if its too dim
    put a resistor in parallel with the photocell to obtain the desired brightness
    when dark.

    Tony
     

  5. Hi, Frank. Folks on this group are regularly asked to design devices that
    aren't in fact the best approach to solving the poster's real problem.
    (Indeed, it's not just on Usenet - it's part of the daily life of an
    engineer.) Good engineering includes questioning the assumptions of the
    person asking for the feature, and considering that there might be a better
    solution than the one that is proposed.

    In the case of your question, John picked up on what seems like a
    discrepancy in your analysis of the best solution, so he probed into it. I
    still think it's interesting, so let me reframe it:

    - Your concern is that the windows should be closed when your mom goes to
    bed.
    - Bright LED's trouble her, and she doesn't notice LED's that are turned
    off.
    - So, why not (as John suggests) have the LED's be bright when the windows
    are OPEN? That way, if they're closed, it's fine if she doesn't notice,
    because there's nothing to do anyway. But if they're open, she's sure to
    notice and close them.

    Again, no offense intended, it's just good engineering practice that if you
    don't understand why a client thinks a particular solution is the best one,
    you should question it before implementing it. It may be that there's
    additional information that is obvious to the client (you, in this case) but
    unknown to the engineer.
     
  6. It sounds like what you want is a way to make the light output of the LEDs
    somehow proportional to the ambient light in the room. Is that right?

    Most of the photoresistors I've seen will increase the resistance quite a
    bit as the room darkens, however, so I don't think just adding one in series
    will work.

    One thing you can do is to pick a resistance that you want to use for the
    dark room, and another that you want to use for the light room. Then, use a
    transistor controlled by the photoresistor as a switch to cut off the lower
    resistance path:

    http://home.comcast.net/~rcmonsen/duallevel.jpg

    I built this one, and it seems to work pretty well with a photoresistor I
    had laying about. However, photoresistors seem to vary quite a bit in how
    sensitive they are, so you may have to experiment with the resistor values.

    Regards,
    Bob Monsen
     
  7. Guest

    On Thu, 27 Nov 2003 04:00:22 GMT, "Robert Monsen"

    Hi Bob,

    Thanks for the reply.
    Thank you very very nuch for that. It will get me started. My
    knowledge is a bit brute force when it comes to electronics. At the
    risk of being greedy, may I ask how I could use the one photoresistor
    to control the 4 LED's with their own switch?

    At a guess can I just parallel the four LED's and adjust R1 and R2 by
    trial and error and have the reed switches in the cathode side of the
    LED's? Brute force here remeber. <g>

    Thanks again.
     
  8. Guest

    It is NOT a lack of additional information, it is more a lack of you
    and John thinking about the problem.

    Try wearing two or three pairs of sunglasses around the house to
    simulate very diminished vision. Then tell me an LED off is as
    noticeable when missing in the enveloping grayness as an LED on.

    You should grant the requestor (me) of postings such as this, with
    some intelligence to have studied the issue and experimented some
    before posting.

    Despite your most fervent beliefs, you and John may not be the only
    two experts left on the planet. It may ALSO be possible that you are
    not always dealing with the great unwashed (thus inferior to you), but
    just someone who has a scant grasp of electronics. Are you and/or John
    in academia?? Either that or you are young and flushed with the
    elitism that fresh out of college tends to invoke.
     

  9. I assumed you had the intelligence to have studied the issue, and I granted
    you the courtesy to ask you for the additional information rather than
    assuming you were wrong. Your desired solution continues to make no sense
    to me: you are seemingly asking for a device that is maximally noticeable in
    the situation where it is least important that it be noticed (that is, when
    the windows are closed already), and least noticeable in the situation where
    it is most important that it be noticed (when they are open).

    My courtesy was evidently misplaced, as you've responded defensively (and
    now offensively) to polite, respectful posts. Your response is, frankly,
    bizarre.

    I wish you the best of luck in solving your problem.
     
  10. Fred Bloggs

    Fred Bloggs Guest

    There is one other situation you have not covered, and that is the case
    of room lights off but a window still open- do you still want to dim
    under these circumstances? The standard technique is to use a CD-cell,
    you can buy cheap little assortments from Radio Shack, a simple switch
    on ambient is easiest, and the trick is to use a single CD to control
    multiple LEDs. I can work an example with some specific assumptions:
    blue/green/red ultra- LEDs, full brightness 15mA @ ~3.5V and dimmed
    level ~2mA @3V; CD is <500 ohms in average room light, >1500 ohms dark;
    12VDC power supply available. In full room brightness, the LEDs require
    (12-3.5)/0.015=560 series resistance, and in darkness they require
    (12-3.0)/0.002=4.7K ohms series resistance. How to do this? One way
    would be something like this-all parts available from Radio Shack:

    Suggest you experiment with those 560/3.9K combination on a single LED
    for best effect before you build out full circuit. The same goes for the
    CS cell controller- if you measure the light/dark resistance, applicable
    component values can be adjusted- the IC comparator is the simplest
    option overall.

    Please view in a fixed-width font such as Courier.

    (1)
    560 sw
    +-----/\/\--o/o------------------------------------------+
    | |
    | 560 sw |
    +-----/\/\--o/o---------------------------------+ |
    | | |
    | 560 sw | |
    +-----/\/\--o/o------------------------| | |
    | | | |
    | 560 sw | | |
    +-----/\/\--o/o---------------+ | | |
    | | | | |
    | | | | |
    | --- --- --- ---
    | \ / ~~ \ / ~~ \ / ~~ \ / ~~
    | --- --- --- ---
    | 5.6K 1N4148 | | | |
    +------/\/\-------+---|<|-----+ | | |
    | | | | | |
    | | 1N4148 | | | |
    +-----+---+ +---|<|-----|--------+ | |
    | | | | | | | |
    12VDC | | | | | | |
    ----- === | | 1N4148 | | | |
    -+- 10U | +---|<|-----|--------|--------+ |
    | | | | | | | |
    +-----+ | | 1N4148 | | | |
    | | +---|<|-----|--------|--------|--------+
    | | | | | | |
    | (3) | | \ \ \ \
    | +-------+ c 3.9K 3.9K 3.9K 3.9K
    | | CS | |/ (2) \ \ \ \
    | |CONTROL|--| 2N3053 / / / /
    | +-------+ |\ | | | |
    | | e | | | |
    | | | | | | |
    | | | | | | |
    +---------+-------+-----------+--------+--------+--------+





    (1) Use 1/2 Watt rating for 560 ohm resistors, all others 1/4W

    (2) 2N3053 vailable from RS, Mouser, also NTE128 over the counter


    (3) Cadmium Sulfide Control Detail:

    1N4148
    12VDC +---------|<|--------------+
    2x 1N4148's| | |
    +-|<|-|<|--+--------+-----------------+------+------+ |
    | | | | | | | |
    | +-----+ | | | / | |
    | | / | | | | 10K | |
    | | \ | CD | | | / | |
    | | / | | | | \ e |
    | | \ | / / | | |/ |
    | +-----+ 4.7K 100K | +----| 2N3906 |
    | | / / | | |\ |
    | e \ \ | / c |
    | |/ | | | 10K | |
    +--------| 2N3906 | | | / +-------+
    | |\ | | | \| \ |
    | c | +----------|+ \ | |
    | | +---+ | | | \ | /
    | | | | | 1K | | >----+ 2.2K
    | +---------------/\/\--+-|- / /
    | | | | | | / 1/4 \
    / / | / | | /| LM339 |
    3.3K 1K === \<--+ | +--->to 3053
    / / 1U /1K POT | | base
    \ \ | \ | /
    | | | | | 1K
    | | | | | /
    | | | | | \
    | | | | | |
    +----------+----+---+-----------------+-------------+




    Pathetic!!!- I know...but not really as complicated as it looks.
     
  11. This might work:

    http://home.comcast.net/~rcmonsen/misc/dimmer.jpg

    I didn't build this one, only simulated it. However, it should work
    properly. You can get the parts at radio shack. Any NPN transistor (like a
    2N2222A or 2N4401) should suffice. Radio Shack also sells assortment sets of
    photoresistors, so you can pick one that works well.

    Regards,
    Bob Monsen
     
  12. Fred Bloggs

    Fred Bloggs Guest

    Yep- that will work . A variation on that is to use a light-level
    variable power supply driving the LEDs like so:

    Please view in a fixed-width font such as Courier.

    4.7K
    +--------------------/\/\-----------------+
    | 4.7K |
    +--------------------/\/\-------------+ |
    | 4.7K | |
    +--------------------/\/\----------+ | |
    | 4.7K | | |
    +--------------------/\/\-------+ | | |
    | | | | |
    | 120 | | | | ~
    | +---/\/\-|>|---|--|--|---+-|>|----------+
    | | 1N4148 | | | |
    | | 120 | | | ~ |
    | +---/\/\-|>|-- |--|--+-|>|---------+ |
    | | 1N4148 | | | |
    | | 120 | | ~ | |
    | +---/\/\-|>|---|--+-|>|-------+ | |
    | | 1N4148 | | | |
    | | 120 | ~ | | |
    | +---/\/\-|>|---+-|>|-----+ | | |
    | | 1N4148 | | | |
    | | | | | |
    | +----------+ | | | |
    | | | | | |
    | +---LM317---+ | | | | |
    | 47 | | | | | | |
    (+)-+-/\/\--+--| IN OUT |----+------+ o o o o
    1/2W | | | | | \ \ \ \
    | | | +-+CS | o o o o
    | | ADJ | | \ | | | | | |
    | +-----+-----+ | / | | | | | |
    | | | \ | | | | | |
    12VDC === | | / | | +----+----+----+
    0.22U e +---+ | |
    | \| | | |
    | 2N3906 |--------+ === |
    | /| | 10u |
    | c / | |
    | | 1K | |
    | | / | |
    | | \ | |
    | | | | |
    | | | | |
    (-)---------+--------+----------+------+------+
     
  13. Sam Lloyd

    Sam Lloyd Guest

    I hope you're just trolling here, because there is a serious lack of
    logic in your statements. What does it matter if she won't notice the
    OFF LEDs? The only time you need to illuminate the LEDs is WHEN A WINDOW
    IS OPEN.

    Having the LEDs on when the windows are closed is analogous to having a
    burglar alarm that only sounds its siren when nobody is breaking
    in...you MUST be trolling.

    Sam
     
  14. Guest

    Nope, just following the wishes of an older loved-one. I tried it both
    ways, she prefers to see when the windows are closed by the LED's
    being on. If it aint logical it doesn't matter squat.

    As I mentioned, I had tried it with the LED's on when the windows were
    open and off when shut but she cannot tell if she is looking in the
    right direction.

    I will say it again, you need to think about this some more.

    Being almost blind is not like being sighted and thinking what it
    might be like to be almost blind.

    She has a very broad frame of scope for where she may be looking and
    having to scan half the room to see if something is missing is about
    as stupid as your burglar alarm analogy. Notice "analogy" begins with
    "anal" perhaps as in "arse" Hmmmm. <g>

    Seems if I do not agree with **your** logic I must be a troll. How
    quaint. It's refreshing to see open-minded assesment of a problem is
    long gone.
     
  15. Guest

    Bizzare, eh?? Perhaps see my response to Sam. <sigh>
     
  16. Guest

    Fred, thanks you so much for taking the time. The ASCII art must have
    taken quite a while and that is much appreciated.

    I will mull over your question regarding the room lights off.

    I think that is covered but I need to think about the question some
    more in case I am not understanding the option fully.

    I have used 6000mcd LEDs in the lounge room, and in
    daylight/room-lights she can see them well. At night time when all the
    room lights are off she can see the bright red glow from her bedroom.
    She is then not sure if it is the heater left on or stove burner or
    something so I wanted to dim them so she could still see if she looked
    in that direction before going to bed, but they would not look like a
    corner of the lounge was on fire from her bedroom. <g>

    I think using the circuit you and/or Robert have supplied will fill
    the bill. As I said though, I will give it some more thought in case I
    am overlooking something in your option.

    Thank you again for taking the time with the design.
     
  17. Guest

    Thank you Bob.

    <g> After thinking about my last response to you I realized the
    ineptness of it unless all LEDs were on or off.

    Thanks for the schematics, I will scoot down to RS and get a handfull
    of parts and try yours and "Fred's" versions.

    Thanks again.
     
  18. Ian Stirling

    Ian Stirling Guest

    Simple solution.
    Pick a wacky colour of LED.
    White, blue, ...
     
  19. Guest

    Hi again Fred,

    I have studied the ASCII art a little more.

    Am I correct in assuming I could add a trim-pot with the CS and use
    that to preset the dimmed level?

    Don't want to get too fancy here. <g> but give someone an inch and
    they will probably ask for a yard.

    I am most grateful already so do not feel obliged to offer further
    suggestions but it just occurred that if this setup is still too
    bright I might be able to adjust the level. Since I am using 6000mcd
    LEDs even some dimming in daylight would be acceptable as they are
    very bright.

    As her eyesight continues to diminish, we will eventually arrive at a
    point where the whole plan will be of no further use so don't bust a
    gut with it. <vbg>

    Even as it is though, I am sure she will be very happy with the
    results.

    Thank you again.
     
  20. Fred Bloggs

    Fred Bloggs Guest

    If you take an ohmmeter and measure the light/dark resistance of your
    cadmium-sulfide cell, and also the forward voltage across the LED when
    fully/dimly lit to your satisfaction, then I can re-post a circuit with
    exact component values.
     
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Similar Threads
There are no similar threads yet.
Loading...
Electronics Point Logo
Continue to site
Quote of the day

-