Connect with us

Detecting voltage drop?

Discussion in 'General Electronics Discussion' started by cristian9509, Dec 27, 2012.

Scroll to continue with content
  1. cristian9509

    cristian9509

    2
    0
    Dec 27, 2012
    Hi,

    I am trying to light up some LEDs on my Rustler VXL RC car only when braking/reversing.
    I found out that on my receiver I have three wires: black(-), red(+), white(+). The one that interests me is the white (signal wire). This wire tells the speed controller to either turn the motor clockwise or anticlockwise.

    White wire voltage is as follows.
    - no throttle: 0.55V
    - full throttle forward: 0.71V
    - full throttle reverse: 0.39V

    So, when I brake/reverse the voltage drops from 0.55V to some value but no less than 0.39V.

    How can I design a system to catch that voltage drop and activate a switch to light up the LED?
    What parts would I need to buy?
    And, what's the difficulty level for this?

    Thanks
     
  2. gorgon

    gorgon

    603
    24
    Jun 6, 2011
    And you are sure this is not a standard servo pwm signal?
    Normally this would be 1-2ms pulse every 12-20ms. From your number I would presume that is the case. center is 1.5ms and this works out to (1.5/12)*5V = 0.625V in pulsed average. Depending on your meter this should not be too wrong. Check the signal out with a oscilloscope and see if this is the case.

    If you find that it is a pulsed pwm signal, you only need to compare the length of the pulse and activate the LED if it is shorter than the threshold you set as braking. You only need a oneshot and a Dlatch to do this, and maybe a transistor to drive the LED.
     
  3. cristian9509

    cristian9509

    2
    0
    Dec 27, 2012
    You are perfectly right, it is pwm signal. I've learned this a few hours ago. Anyway I don't have an oscilloscope to measure the signal so I'm stuck a bit.

    But I am curious, how would I design any system with the oneshot and Dlatch? (and what is a oneshot?)

    Thanks!
     
  4. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,839
    Jan 21, 2010
    You need a device that detects signals longer (or shorter) than a certain time.

    It's actually not that hard to do. I'll draw something up for you in a while...
     
  5. (*steve*)

    (*steve*) ¡sǝpodᴉʇuɐ ǝɥʇ ɹɐǝɥd Moderator

    25,497
    2,839
    Jan 21, 2010
    [​IMG]

    This circuit produces a low output (lighting the LED) whenever the input remains low for too long.

    I presume your pulse train is a fixed frequency, and the mark/space ratio changes.

    Whenever the input goes high, C1 is discharged and the output of the first schmitt trigger inverter goes low.

    When the input goes low, C1 starts to charge through R1. After a delay of approx 1 x RC, the input is low enough for the output of the first schmitt trigger to go high.

    If the input goes high before this occurs, the output will remain low.

    So the output of the first schmitt trigger is a pulse which occurs whenever the input signal remains low for more than R1C1 seconds.

    The second half looks almost identical to the first, but it is operating on inverted logic, so the actual function is completely the opposite! The function of the second half is to extend the length of any pulse from the first schmitt trigger to a period longer than the period of the input signal. The effect of this is that the output from this stage will be a constant value as long as the first half is generating pulses.

    So while the output of the first schmitt trigger remains low, the capacitor remains charged, and the output of the second schmitt trigger remains high.

    When the output of the first schmitt trigger goes high (briefly -- when the input has remained low for too long) the capacitor is discharged and the output of the second schmitt trigger goes low. It remains low for R2C2 seconds or until another pulse arrives from the first schmitt trigger.

    The value of R1 and C1 need to be chosen to detect the signal remaining low for more than a certain time.

    R2 and C2 need to be chosen to make this time constant longer than the period of the input signal.

    I'm not sure of the timings or periods, so I can't immendiatly advise on values. It would be wise to make R1 variable so you can get the delay exactly right. Just don't let the value get too low (so have a fixed resistor in series with a trimpot)

    OK, reading through this again, it appears you're trying to detect the length of a high pulse -- 1.5ms or longer in every 20ms period.

    In that case you need to invert the input signal (that's easy -- there are 6 of these schmitt triggers in a package, so you've got plenty) Oh, and by the way, use a 40106 as it will work from a 12V rail.

    so C1 and R1 should be perhaps in the order of 0.1uF and 10k

    Because the period is 20ms, C2 and R2 should be around 0.22 uF and 100k

    The R1 will need to be adjusted to get the right point to turn on and off. To do this, use a 5k6 resistor in series with a 10k pot.
     

    Attached Files:

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.
Electronics Point Logo
Continue to site
Quote of the day

-