voltage divider

Discussion in 'General Electronics Chat' started by matty204359, Aug 17, 2012.

  1. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    I need a stable 2.5volt reference voltage. I figured since I already have a stable 5 volt source, that two same value resistors would make a quick and cheap voltage divider. I have two 25.5kΩ 1% resistors wired on a bread board in series. A quick check with a calculator and ohms law indicated that the current draw of the two resistors is approximately 0.1mA. which is very good since I'm intending it to run on battery power. I intend to use the 2.5 volts in a voltage comparator circuit with a MOSFET Op Amp. So it was my thought that the load of the voltage divider output(2.5volts) be more or less negligible. however when checking the voltage with my digital multimeter, the voltage across each resistor is actually 2.46 volts, which is close enough for my application. what has me confused is it reads 5 volts across both.

    I'm having a hard time figuring out why the two voltages across the resistors don't add up to 5 volts. Is this normal behavior? are my resistor values too high?

    attached is a picture of my set up
    [​IMG][​IMG]
    [​IMG]
     
  2. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
    Hello,

    You have NOT created a voltage divider.
    The resistors are both shorted by the blue line.
    See the attached PDF how the internal connections of a breadboard are arranged.

    If you want a stable 2.5 Volts reference voltage, have a look at the TL431.
    See the attached datasheet.

    Bertus
     
  3. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    sorry I should have made more emphasis on the second part of the image. I expected that response which is why included the second image. I modified the breadboard's bus line by cutting the connections between each five pin group.
     
  4. absf

    Senior Member

    Dec 29, 2010
    1,493
    372
    Funny:rolleyes:, if you see the second picture, the bottom of the breadboard of the blue and red lines are not joined together in one long strip.
     
  5. bertus

    Administrator

    Apr 5, 2008
    15,649
    2,348
    Hello,

    Wich wires are from the power supply?
    Wich wires are going to the multimeter?
    If the white and the green on the front go to the DMM, you will measure the full voltage.
    If the green wire is placed between the resistors, you will measure 1/2 the voltage.
    Remember that this voltage will follow the changes in the power supply by half.

    Bertus
     
    matty204359 likes this.
  6. absf

    Senior Member

    Dec 29, 2010
    1,493
    372
    IMHO the reason you're not getting 2.5V with your voltage divider is because the voltmeter itself also has an internal resistance expressed as KΩ/V. Though digital meters has reasonably high resistance it would still affect your reading when the voltage measured is low.

    Think of it as 25.5K parallel with a 1M resistor on one branch while the other branch is 25.5K and recalculate the divider again. I hope you'll get what I meant.

    Allen
     
    matty204359 likes this.
  7. #12

    Expert

    Nov 30, 2010
    16,343
    6,828
    The internal resistance of the voltmeter can knock your voltage off by the 1.6% that is shown.
     
    matty204359 likes this.
  8. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    Yes the white and green are to the DMM which measures the full voltage, it shows up as -5volts because I had the polarity mixed up. I was using two DMM's so I could make sure that when reading the voltage across each resistor it wasn't changing the overall voltage. The yellow and white wires come from the power supply.
     
  9. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,394
    1,606
    It's expected the overall voltage remains the same. It is not the voltage across both resistors that is changing, it is the impedance (resistance) of your meter that is changing the circuit. Since you have 2 meters place one across the top resistor only, then the 2nd meter across the lower. See if when you add the 2nd meter the reading on the first meter remains the same. I would expect some small change on the 1st meter.

    Also, unless you bought absolutely dead nuts matched value resistors the tolerance will lead to getting a not quite 50% divider. The good news is as long as your 5V supply is stable your reference scheme should be acceptable.
     
  10. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    You guys were completely right. After thinking about the situation for a while I tried a crazy idea of using both my meters at once. I had the idea that being a similar make and model they must have similar internal resistances. when I measured the voltage across both resistors at once with the two different DMM's the value came up to exactly 2.50 volts on both DMM. so you guys were right that when I was measuring one at a time I was changing the ratio of the resistors thus giving me a incorrect reading. which is funny because I was so meticulous at choosing two resistors that were close to the same value, I was getting some pretty crazy values with the 5% resistors.

    that is why I went with the 1% resistors instead of the 5% resistors.
     
  11. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Why did you cut the bus rails?:eek: That is a naughty thing to do!:mad:
     
  12. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    It looks to me like it was made that way.
     
  13. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Op says in post #3 that he cut the connections.
     
  14. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    I cut them because I needed a little work light, I wired up some 1/2 watt (45 lumen) white leds in series. it seemed safer to wire it up in series at the time(ie. to have a consistent current through them). I didn't want to solder the LEDs to a veroboard because it was just a temporary thing. I have since retired the work light and moved the LEDs to another project. The little bus strips have come in handy for other small circuits I need to bread board(obviously not DIP compatible though).

    Indeed for observers of the circuit who are unaware I have cut the bus strips it can make circuits look confusing but I'm the one who cut them so it makes sense in my head.
     
  15. MrChips

    Moderator

    Oct 2, 2009
    12,449
    3,365
    Maybe one of these days you will learn that that was not such a good idea.
     
    matty204359 likes this.
  16. KJ6EAD

    Senior Member

    Apr 30, 2011
    1,425
    363
    It's alright to gap the buses like that if you have a good reason but you should make corresponding gaps in the red and blue marker strips to indicate the modification.
     
  17. ramancini8

    Member

    Jul 18, 2012
    442
    118
    The initial tolerance is 1%. The real life tolerance is 3% because of temp., life, etc.
     
  18. ramancini8

    Member

    Jul 18, 2012
    442
    118
    Any noise on the 5V supply shows up at half strength on the 2.5v reference. Place a 0.1 UF cap across the bottom resistor to filter out supply noise.
     
  19. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3
    that seems reasonable. so by a similar logic would a 5% tolerance resistor actually be something more like 15%?

    good idea I'll scrape the paint off with an xacto blade.
     
  20. matty204359

    Thread Starter Member

    Apr 6, 2011
    105
    3

    not sure what you mean by bottom resistor

    I drew an image of what I think your telling me to do

    [​IMG]
     
Loading...