Voltage regulating pulses!

Discussion in 'The Projects Forum' started by alistair, Mar 18, 2010.

  1. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    Hi

    I'm currently working on a project to monitor the flow of pesticide being sprayed on to a field. I have just about got everything working, and i guess the problem i have is very easily answered.

    The flow meters that i want to use run at 12 vdc, and give off pulses depending on the speed of flow of the pesticide. I'm monitoring them (3 flow meters) with an arduino which runs at 5 vdc, so I need to drop the voltage.

    Has anyone any suggestions on, the type of voltage regulator to use, any potential problems, do the regulators alter the pulses in any way?

    Cheers
    Alistair
     
  2. Bychon

    Member

    Mar 12, 2010
    469
    41
    How about just putting a resistor in series with the signal pulses and a zener diode to ground or a 1n4004 diode to the +5 supply?
     
  3. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    Is this an advantage to using an off the shelf voltage regulator?
     
  4. retched

    AAC Fanatic!

    Dec 5, 2009
    5,201
    312
    A decent regulator with protection will act as a resistor with a diode. So its up to you. I guess one component over 2 wins out.
     
  5. someonesdad

    Senior Member

    Jul 7, 2009
    1,585
    141
    Or just a voltage divider with two resistors?
     
  6. retched

    AAC Fanatic!

    Dec 5, 2009
    5,201
    312
    Yup. You can take you pick. Then its just using the uC as a frequency counter and display the readings of each sensor.
     
  7. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    Sorry for being a bit slow in all this, but as usual i've jumped in the deep end and am hardley able to doggy paddle!
    Are you saying the 12vdc supply should be fused to 1 amp or so? Or is there another way i don't know about:confused:

    This sounds simple and i have the resistors. Any reasons for not doing it this way?
     
  8. retched

    AAC Fanatic!

    Dec 5, 2009
    5,201
    312
    No I was talking about a voltage regulator IC that has built-in protection diodes.

    If the 12v is regulated or stable, from the sensor. IE always 12.0vDC. not 13.6 then 11.2 etc...

    Then you can use a voltage divider that will split the 12v to 6v then you can use a in line resistor to drop it to 5v.

    If the voltage jumps around in the signal, you will want to use a regulator.

    If it jumps to 13v, then split to 6.5vDC then the resistor drops it to 5.2, then you will be sending higher voltage to the uC than it wants.
     
  9. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    So that sounds to me that I'm better off with a voltage regulator (eg 7805) to be on the safe side, and its just one component.
     
  10. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    Thanks for the quick and helpful responses. Most appreciated.
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Ahhh, wait - you should not use a 7805 regulator without the 0.33uF input and 0.1uF output caps, or you risk high-frequency oscillation. See the datasheet. The caps are required for stability.

    The caps will likely prevent sensing the dips in voltage.

    Bychon's suggestion was much better.
     
  12. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Here is a viable circuit to solve your problem:

    [​IMG]

    Bychron proposed using just a Zener or a 1N4004, but including both would be better. The 1N4004 will help to prevent damage should the Zener diode fail, and will help in the event of unforeseen circumstances. Better yet would be to replace the 1N4004 with a Schottky diode like a 1N5817; as they have a much lower Vf (forward voltage) than the 1N400x series.

    The additional cost is minimal considering the cost of replacing the uC.

    A single 7805 with no caps is not a viable option; neither is a 7805 WITH caps.
     
  13. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    All you need is a two resistor divider to reduce the voltage to around 5V, then a series resistor to the microcontroller input to limit the current in case or spikes or surges.

    6k8 from the input and 4k7 to ground make a suitable divider, then a 10K from the junction or those two, to the uC input.
     
  14. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    someonesdad suggested the voltage divider, then other people said that it wasn't what i needed, so I dont know who to believe on that ?

    Thanxs SgtWookie, makes a lot more sense drawn out. Finding it difficult to source the 1N750 4k7 zenner (can import from Canada on ebay), would something like a BZV85C 4v7 or a BZX79C 4v7 do instead coz i can get them for 17pence each.
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Yes, the BZX79C, 4V7, QH06G is basically equivalent to the 1N750; a better match than the BZV85C.

    For the 1N4004, you can use a 1N4003S Code: QL75S, £0.16

    The exact parts aren't terribly critical.

    How far away from your arduino are the pumps?
     
  16. rjenkins

    AAC Fanatic!

    Nov 6, 2005
    1,015
    69
    Sorry, but the one resistor with zener and/or diode is not a good idea.

    With an input less than 5V it has effectively infinite impedance and would be extremely sensitive to noise and stray triggering.

    A resistive divider puts a permanent load on the source and also divides any input noise by the same ratio.
     
  17. atferrari

    AAC Fanatic!

    Jan 6, 2004
    2,650
    765
    May I know why not? Are we trying to regulate the voltage for the Arduino, isn't it?

    Intrigued.
     
  18. alistair

    Thread Starter New Member

    Mar 18, 2010
    10
    0
    atferrari, I'm trying to convert a 12 vdc pulse to a 5vdc pulse for an arduino to read.

    SgtWookie, the flowmeters (not pumps, if that makes a difference) will be around 6ft of wire away from the arduino.

    rjenkins, the input voltage is 12vdc from the flow meters, if that makes a difference.
    wouldn't be good! Your option is the easiest because i have all the pieces.

    Would like to hear your opinions plz on what rjenkins is describing.
     
  19. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    If the input is a brief 12v pulse, the input and output caps will absorb the pulse.

    Our OP has merely stated that the input from the pumps is a 12v pulse.

    We don't know the duration of the pulse; that has not been explained.
     
  20. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Have a look at the attached.

    "Rwire" is somewhat misleading; it is an arbitrary number picked from a random hat, so to speak, to represent the resistance of the wire (which should actually be <1 Ohm) and the output impedance of your 12v pulse signal. I have no idea what the impedance of your 12v signal is, but it should be accounted for. It is somewhere between zero and infinity, non-inclusive. The higher the impedance of your 12v signal source, the worse that any of the proposed solutions will perform.
     
Loading...