Shop OBEX P1 Docs P2 Docs Learn Events
Analog to Digital >5V — Parallax Forums

Analog to Digital >5V

Bobb FwedBobb Fwed Posts: 1,119
edited 2008-06-16 21:48 in General Discussion
I am using a 5V Analog to Digital converter (ADC0831CCN). I am wanting to check voltages above 5V. The way I have done this is by placing a 1K-ohm resistor between V+in and V-in and a 3.3K-ohm resistor between the actual tested voltage and V+in (view attachment for diagram). This allows me to read up to 21.4V (or so - determined just by testing). I am wondering if there is a better way to test higher voltage and/or what is the math to determine exactly what the testable voltage is.
I understand that all I am doing is converting the 21.4V to 5V and anything between 0 and 21.4 is changed to something between 0 and 5. Any inputs or comments would be appreciated.
1600 x 1280 - 105K

Comments

  • MSDTechMSDTech Posts: 342
    edited 2008-06-16 18:30
    The idea of using a voltage divider is fairly standard.
  • stamptrolstamptrol Posts: 1,731
    edited 2008-06-16 19:36
    While the principle of a voltage divider is very simple, the actual implementation can have some unwanted effects on the circuit being measured.

    For instance, a 10 ohm in series with 90 ohms will generate a 1volt signal across the 10 ohm resistor when the pair of them are energized by a 10 v source. So will a 10K ohm and 90K ohm resistor pair. Which one loads the circuit under test the most?
    Sometimes, the loading doesn't matter ( say reading a battery voltage or power supply voltage), sometimes it will REALLY affect the reading of a sensor which has a high internal impedance.

    Similarly, depending on your choice of a/d chip, it may load the voltage divider and affect your converted number.

    Cheers,

    ▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
    Tom Sisk

    http://www.siskconsult.com
    ·
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2008-06-16 20:20
    I am just testing a battery supply. How does the voltage divider work? I looked up a few from a supplier I use...it doesn't look like it does much, is the is the right thing? www.alliedelec.com/Search/ProductDetail.aspx?SKU=8951115&MPN=HL25-06Z-10%205%&R=8951115&SEARCH=8951115&DESC=HL25-06Z-10%205%

    Post Edited (Bobb Fwed) : 6/16/2008 8:32:21 PM GMT
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2008-06-16 20:37
    oh...or is voltage dividing what I am already doing...
  • DufferDuffer Posts: 374
    edited 2008-06-16 21:36
    Yes you are. Here are a couple of sites that·show you/help you·calculate the values of a 2-resistor voltage divider. The first has the formulae as well as a handy calculator. The second is just a quick calculator.

    http://hyperphysics.phy-astr.gsu.edu/hbase/electric/voldiv.html


    http://www.raltron.com/cust/tools/voltage_divider.asp

    Steve
  • MSDTechMSDTech Posts: 342
    edited 2008-06-16 21:48
    The attached drawing shows a simple voltage divider. The voltage applied to the ADC will be

    VADC = (R2 / (R1+R2)) Vto measure

    As stamptrol·mentioned, you need to be carefull selecting the values of R1 and R2 so that the current drawn by the voltage divider does not significantly change the circuit being measured and does not feed too much power into the resistors.
  • Bobb FwedBobb Fwed Posts: 1,119
    edited 2008-06-16 21:48
    Yeah, I figure it out before that, but thanks. It's just a ratio between the two resistors, thus, in my case it wasn't 21.4V (from testing) it was a 21.5V max --pretty close testing though.

    So what do multimeter's do to allow for for inter-circuit testing? Really high resistance? like instead of 3.3K-ohm and 1K-ohm they would use 3.3M-ohm and 1M-ohm? There is still a little draw (4.884µA instead of 4884µA @ 21V), but I would still imagine it would mess up precise measurements.

    Post Edited (Bobb Fwed) : 6/16/2008 9:54:14 PM GMT
Sign In or Register to comment.