Switch with Timer Circuit

Discussion in 'The Projects Forum' started by dmanoukis, May 29, 2009.

  1. dmanoukis

    Thread Starter New Member

    May 29, 2009
    8
    0
    Morning All, I will try and keep this short. I am new to the board so I appreciate any help I can get. I am looking for help to design a circuit that runs in a 12V dc environment, and is triggered by a constant + signal that once below a specified voltage (4v) it creates a output (+12) for a relay. The catch is that it has to be able to filter out short period drops to 4v. What I mean is that if the voltage on the input is below 4v for say 3-5sec, then the output is given, if not the circuit stays open (no output). Any help would be great. Thanks again.
     
  2. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,534
    Several ideas suggest themselves, but it would help if we knew the application.

    My current thought is a RC network feeding a 555 circuit. If the application is voltage sensitive as you suggest then maybe a comparitor.
     
  3. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,534
    Posted by OP as a PM. Sorry guy, thought you were someone else (I just woke up) when I answered. I'll see what I can draw up, you'll need to prototype it (I strongly suggest a protoboard). It's going to take a bit of a while, since I'm going out in a bit.

    One more question though, You want this circuit to go back low after 3-4 seconds with light again, or latch in the high state?

    If you're not familiar with protoboarding checkout my blog and the several 555 projects I have in it.

    Bill's Index
     
  4. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,534
    OK, don't know if this will do what you want it to, but it is a starting point...

    [​IMG]

    You said you wanted an adjustable voltage. The pot R1 provides for this, measure the voltage on the wiper.
     
  5. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Bill, I think you have the inputs swapped on the first comparator. That'll cause the headlamps to turn off about 3 seconds after it gets dark. :eek:

    See the attached simulation. The green trace represents power in the headlamp, which I arbitrarily set to 13.8w (actually, just to keep it in scale with the rest of the traces). Other than that minor correction, it looks like it'll do what our OP wants to do.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    After posting the simulation, I realized there was a bit of a problem with the hysteresis on the 2nd comparator, due to the base of Q1 limiting the voltage swing on U1B's output.

    With this version, you can change the hysteresis by changing (increasing) the value of R7.
     
  7. Wendy

    Moderator

    Mar 24, 2008
    20,764
    2,534
    The only reason I have a Schmitt Trigger there is to prevent the output from chattering. I basically agree with your analysis, but it may be good enough. When the output is low you have 4V on the + pin, when the output is high you have around 4.5V, because the transistor is holding the voltage at 0.6V on the output. Not much, but probably enough. Another good fix would be to move the relay to the emitter of Q1. So if it does need fixing we would approach it differently (no problem there).

    I think you're right about the input, or the OP can use the other set of contacts on the relay (joking).

    We're still waiting for the OP to get back with us, I'm waiting to hear the verdict about the latching issue. I drew the schematic to start the ideas flowing. I also realized I have to add the pinout of the LM339 to my pinout sheet on PaintCAD. I would have bet money it was there.

    Think he'll be back?
     
  8. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Acknowledged. It certainly is necessary.

    It might be; we wouldn't know unless it was tested under actual conditions. I just simulated it. With the slight modification, the circuit still works the same way, but there's an added "tweak-ability" factor.

    Well, yes - that COULD be done. However, it's generally preferable in automotive environments to assert grounds than to assert Vcc. That's how most autos are wired; I'm not certain quite WHY it's done that way, but the engineers must've had a reason. Being consistent with the existing design does have it's advantages.

    ror!
    Did it work? ;)
    I don't worry too much about pinouts on my diagrams. Better that people download the datasheets, which provide a lot of in-depth information. Might be "too much information", particularly for a n00b - but getting exposure to such things is part of the indoctrination.

    Will our OP be back? That's up to them... they may have been hoping for a super-simple quick circuit - and this one may have intimidated them. One never knows.
     
  9. dmanoukis

    Thread Starter New Member

    May 29, 2009
    8
    0
    Guys, Thanks for all the ideas. I apologize for not replying sooner, was out of internet range for the weekend. I will prototype over the next two days and let you know how I make out. THANKS AGAIN!!!!
     
  10. dmanoukis

    Thread Starter New Member

    May 29, 2009
    8
    0
    Morning Guys... Problems. I built the circuit on a Beadboard. First thing is I used a TIP120 for the transistor, is that ok?? Its all RS had. Also, I could not find a 7.5k res, so I used a 4.7 and a 2.2 in series, is that too far off. I tested teh circuit, set the input to 4v and tried adjusting the variable res from end to end to see if I could get any cation out of the relay, I got nothing. Any help would be great guys. Thanks
    ps- I am using the 1 & 2 units in the 339, and also connecting the + and 1 on 3 & 12.
    Just making sure I still need to do this.
     
Loading...