Help Creating Basic 10 Second RC Alarm Circut

Discussion in 'Homework Help' started by Mawangs1, Apr 14, 2012.

  1. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    I could've just put one resistor value, but I wanted to show you guys exactly what I have hooked up. Now, I'm basically an ABSOLUTE beginner to circuit design, so I'm not 100% why this doesn't work.

    In my mind, using the supplied resistor and capacitor values, this gives me 10s until the alarm sounds. Instead, it sounds within 1 second of power being supplied. What am I missing here?
     
  2. panic mode

    Senior Member

    Oct 10, 2011
    1,320
    304
    you are missing sensitivity level that triggers alarm. apparently it is lower than you think
     
  3. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    Your schematic just says "to speaker". If this is truly going to a speaker, I wouldn't expect you to hear anything. Is it going to a trigger input of an alarm circuit? If so, what matters is when the voltage crosses the threshold of the trigger circuit. You'll either have to tell us much more about what the signal is hooked up to, or read through the documentation for anything indicating what input voltage conditions result in the alarm sounding.

    A couple of other points for consideration:

    Assuming your components are ideal, you have an RC time constant of 9.1s. So I'll agree that this is, ideally, in the realm of the 10s you are shooting for. It also means that you are expecting your alarm to turn on when the voltage at your "speaker" terminals reaches 6V. Why are you expecting it to happen at this voltage? But let's just go with that threshold, for now. Now consider the following in light of the fact that your components are not ideal:

    What is the tolerance on the resistors? Let's say that the 39kohm resistor is has a 5% tolerance, meaning that it's resistance could be up to 1.95kohm in either direction. Thus the impact of your other three resistors, particular the two 100ohm resistors, is pretty much lost. In fact, there is a fair chance that they are making you further away from your desired total resistance than if they weren't there at all! Even if your 39kohm has a 1% tolerance, it could still be off by 390ohms in either direction, again making your 100ohm resistors meaningless.

    The same is true of your capacitor, and capacitors in this size range typically have 10% tolerances or worse.

    To top that off, larger capacitors, especially if they are electrolytic, have leakage currents (modeled by an effective shunt resistance) and if your timing current is not large compared to the leakage current, then you can see longer than expected time delays or can have the circuit stall out completely when the voltage across the resistors drops to the point where it can just support the leakage current. I doubt this is the case here, because while 10s is getting to be a fairly large time delay, it is not too far out in left field.

    How critical is it that the delay be ten seconds? If it has to be within less than a couple of seconds of that, you will probably want to either test it with different resistors and find the one that works best, or put a trim pot in there so that you can easily calibrate it. Even then, you will probably find quite a bit of variation as the temperature changes since the capacitor will probably change by quite a bit. On top of that, you will have different time delays as the battery voltage varies. But if you only need something that goes off at somewhere between, say, 6s and 15s, then you are probably okay with this approach.
     
  4. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    Sorry about that. I meant, "to buzzer." I just bought a radioshack piezo buzzer (number 273-065 to be exact) and connected it into my circuit.



    It was a completely random choice. Should I have chosen something different?



    My overall goal with this project is to create a RC alarm circuit that can time from 0 to 3 minutes in one minute increments. I chose 10s because my thinking was that if I could get a circuit to work with a really low time, than I could see my way to the major goal. Now that you mention 10s is so high, would I have to use something outside a capacitor and resistor to attain higher time delays?

    I have to figure out how to do this without a 555 timer or any other digital components.
     
  5. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    Yeah... like perhaps the voltage at which the buzzer turns on.

    I went to the Radio Shack website and could not find 273-065. I did find a legacy part 273-065A, but it apparently hasn't been sold since at least 2006 and there was no documentation for it. Where did you get the buzzer? Did it have ANY documentation or description at all?

    If you can't find anything, then just slowly increase the voltage on the trigger input until it just turns on and measure that voltage and use that as your starting point.

    Why is that?

    And, again, how accurate does it need to be? For instance, if you set if for 2 minutes, what is the shortest and the longest actual times that are acceptable?

    There are MANY ways to do this, but which ones are acceptable depend on your constraints. Don't piecemeal them to us. Tell us what you NEED the circuit to do and what the restrictions are.
     
  6. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    My apologies, the number on the speaker is 273-059.


    Okay, you asked. The project description is as follows:

    The only components we've been "introduced" to are resistors, capacitors, inductors, and op amps. I hesitated to layout the whole project at first since I didn't want to come across as lazy, but to be honest, I really have no idea what to do here....
     
  7. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    Thanks. Okay, the answer to why no digital is because it isn't allowed under the rules of this project. If it had been something else, like there not being a digital supply voltage, then we could have pushed back on that by generating the necessary voltage from the battery. But, in this case, the rules are absolute and they must simply be lived with.

    Don't worry about laying out the whole project making you look lazy. You are only providing information regarding what you are being asked to do. You look lazy if it becomes evident that you want someone else to actually do it all. The fact that you came to the table with an initial attempt that has been designed, tested, and found to have problems tells us that you really are asking for assistance and understanding.

    The fact that the problem doesn't specify accuracy requirements probably means that anything that is reasonably close (and the grader gets to define 'reasonable') will be accepted. In the real world, customers will come to you all the time with undefined specs. Mostly it is because they are only thinking in an ideal world or because they don't have any idea what the specs should be. As an engineer, it is NOT your job to make up your own specs. Instead, it is to identify what specs are missing and get the customer to provide them. Now, this frequently involves educating the customer on the issues involved and also working with the customer to find out what the real ultimate needs of the project are and translate those needs into performance specs and get the customer to adopt them.

    In this case, I recommend taking the same approach. Go to your instructor and tell them that it would appear that component tolerances might cause problems in getting super close to the desired time delays and that you need to know what the accuracy requirements for the design are. Even if they do nothing more than tell you to do the best you can (which is a copout, in my opinion), the fact that you asked the right question will bias them in your favor.

    Now that we know the context of what you are trying to do, I would say that pushing the practical limits of what you can achieve with an RC delay is not a big concern.

    I don't know what kind of opamp circuits you've been exposed to. There are several simple ones that could be quite useful here, including an integrator and a comparator (preferably a Schmitt trigger comparator). There are also a number of approaches for how to get the different options. You could even make the circuit so that you use two switches and the delay is given by the binary value represented by the switch positions.

    To give you a hint on one possible way to proceed:

    1) What if you had a voltage source that ramped linearly from, say, 2V to 7V at a steady rate of 1V/s?

    2) What if you also had a comparator with thresholds that could be set to 3V, 4V, 5V, and 9V depending on the positions of two switches?

    How might you combine those two things in order to achieve what you want?
     
  8. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    Okay, I don't really have anything solid, but here's what I've been able to think out so far.

    1.
    After researching how to create an analog to decimal converter, I came up with the circuit in 2bit.jpg. V1 would be my MSB and and V2 would be part of my LSB. I decided to only use a maximum of two bits since I'm only interested in times from 0-3 minutes. Thus, with two bits I have a maximum of 4 switch positions. Through this approach I get a little confused though since I think I could only supply a maximum of 1.5V which wouldn't be enough since the buzzer appears to need a minimum of 3V. Am I limited to using only 0 and 1V for V1/V2 or can I use higher voltages??

    2.
    I also did a little research on the comparator as well and came up with the circuit in comp1.jpg. The resistor values aren't exact, but my thinking is I could create a voltage divider circuit with the three resistors shown and have it so the necessary voltage of 3, 4, 5, or 9V could lie across one of each. At that point, I would wire up my circuit as such that my analog to digital converter switch would choose the resistor value with the voltage across it I need.

    I figure what I want is for Vo to be the voltage across my resistor of choice. Would I want to connect the voltage divider across the Vcc+ and Vcc- rails of the amplifier instead or would I somehow connect it to both?

    I don't know if all this would go to Vcc or Vee instead of the inverting/noninverting input though.

    3.
    As for the voltage source that ramps linearly from 2-7V at 1V/s I'm not sure. All I can figure is that if you use a 1F capacitor alone with a 7V voltage source, it might charge at a rate of 1V/s. That can't be the right way though since it would be simply to costly to do..


    In all, I'm still a little fuzzy on how to combine both your (1) and (2)
     
  9. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Did your instructor specifically exclude a 555? I ask because of the way you phrased this statement, as the 555 is not a digital chip.

    For the record... The single imperative attribute that an RC timing circuit needs to function properly is an extremely high parallel resistance across 'C'. The basic RxC formula assumes no load across 'C'. If your circuit loads 'C' you won't accomplish much. So, since your time delay is going to be very long I suggest that you start thinking FETs. ;) You can cascade two or more RC FET stages to get longer delays. This will allow for smaller values of 'C'.
     
  10. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Since this is your homework assignment I've excluded most of component values in this schematic. I've also excluded your Piezo buzzer but you should be able to determine what component to replace with it. Two caps were used for the 'C' element in t=RC because large value Tantalum caps are not as common and their voltage ratings are usually lower.
     
  11. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    Thanks for the info, but I can't use utilize it. I asked my professor and he says the only allowed components outside of a switch are resistors, inductors, capacitors, and opamps.

    What is confusing me the most is where to connect my components to each other. In particular, how would I connect them so that my led won't turn on until my capacitor reaches the minimum required voltage? Or really, how could I isolate the voltage and current from my LED until my chosen voltage or amount of current is reached?



    The included picture is what I'm thinking of.
     
  12. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    Go back and look at my earlier reply:

    http://forum.allaboutcircuits.com/showpost.php?p=476890&postcount=7

    Do you see at the very bottom the part about a comparator. Look at what a comparator does and how to make one with an opamp.
     
    Mawangs1 likes this.
  13. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Is the RC time constant dictated by your instructor still 3 minutes?
     
  14. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    I don't think the RC time constant was ever specified. In fact, I don't think it is specified how the delays are achieved, just the bag of components that can be used. He's not looking for a time constant to be 3 minutes, but for the delay between two things to be 3 minutes. Subtle, but critical, difference.
     
  15. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    This is your home work, so you're going to have to read the notes and take the ball from here.
     
    Mawangs1 likes this.
  16. panic mode

    Senior Member

    Oct 10, 2011
    1,320
    304
    ONLY: switch, resistors, inductors, capacitors, and opamps?
    what is then indication of timer complete supposed to be?
    smoking resistor? is the timer expected to be one time use? :p;)
     
  17. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    Thanks for the idea. Taking a cue from what you made, I tried to create a circuit by filling in the blanks in the attached "untitled.jpg".

    For some reason, the light immediately comes on when I hook up my circuit, but if I'm analyzing everything right,it should take about 38 seconds for it to come on.

    Have I just connected my components incorrectly?
     
  18. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Yes, look how you have R & C connected. Does that look correct to you? Take another look at my OpAmp circuit again.
     
  19. Mawangs1

    Thread Starter New Member

    Mar 26, 2012
    23
    0
    Oh wow! I did have my RC part of the circuit wired wrong. I can't believe I got somewhere with this circuit! Thank you guys so much!

    I believe that my biggest issue with the design of this circuit was assembly. After Wbahn provided me with the comparator tip, I had a good idea of what I needed to do to make the circuit. However, since I'm so uncertain of myself when it comes to assembly, I wasn't sure how my components needed to be connected. This is a good reason for the mistake I made on the RC part of my circuit..
     
  20. WBahn

    Moderator

    Mar 31, 2012
    17,737
    4,789
    What I recommend you do now, it write up a little report (even if you are the only one that ever reads it, so do it in a lab notebook so you can refer to it later easily) that walks through how the circuit works and what role is played by every component in it and why each component is the value that it is.
     
Loading...