Checking voltage levels

Discussion in 'General Electronics Chat' started by Johndon2000, Jun 14, 2013.

  1. Johndon2000

    Thread Starter New Member

    Jun 12, 2013
    25
    0
    When I'm testing a board and I measure the voltages I simply check the the 5V, +/-12V, 3.3V, 2.5V rails are present & correct... is that sufficient, or when people say "check clocks, resets, voltages..." do you actually do a more in-depth look into each voltages at different points on the board, check each IC has it's correct Vcc etc?!
     
  2. Shagas

    Active Member

    May 13, 2013
    802
    74
    Where are you probing for the presence of those voltages ?
    If you want to thoroughly test a board you will have to test more than just a few voltages
     
  3. Johndon2000

    Thread Starter New Member

    Jun 12, 2013
    25
    0
    I'm checking voltages at a few test points, on capacitors etc.


    I am simply checking that the cct has all the required voltages... at the other extreme you would check all voltages are correct on all devices etc... I guess my question is how do you decide how many points to check each voltage?!
     
  4. WBahn

    Moderator

    Mar 31, 2012
    17,748
    4,796
    That depends entirely on the context and purpose of the test.
     
  5. Johndon2000

    Thread Starter New Member

    Jun 12, 2013
    25
    0
    The purpose of the test is to see if the voltages are ok!
     
  6. WBahn

    Moderator

    Mar 31, 2012
    17,748
    4,796
    If that's all you are interested in, then just check those voltages that interest you.

    Usually, tests are aimed at discerning more than that. For instance, we would get ICs back from the fab house and needed to quickly determine which ones were likely to function (or, more accurately, identify the ones that didn't stand a chance in hell). So we powered them up and checked the bias voltages that where brought out and the current draw on each of the supplies. If all those were good, we had a pretty good indication that the chip was probably alive. That was one test. We then took the chips that passed that test and did additional tests to see if, for instance, their internally generated clocks were alive and correct (assuming they were accessible externally, which wasn't always the case). If they passed that test then we did yet additional tests to see if the outputs were qualitatively reasonable. If they passed that test then we might characterize them quantitatively.

    Each successive test required more time and a more involved setup, so we didn't want to spend the time setting up to characterize a chip if a simple test could tell us that the chip was dead to begin with.
     
Loading...