applied calculations to schematics and troubleshooting

Discussion in 'General Electronics Chat' started by relicmarks, Nov 11, 2008.

  1. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    Which formulas do u guys use when trying to troubleshoot ? or break down a schematic ?

    What formulas would u guys use to "Verify" when troubleshooting?
     
  2. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    these are the ones i have used

    1.) AC power output Vout*2/speaker ohms =
    2.) Gain for transitor or OP-amp stages Vout/Vin =
    3.) RC= time constants
    4.) To convert frequency into Time formula


    I can't really use ohms law because there is to many resistors and capacitors in series/parallel networks so it would be hard to verify that
     
  3. beenthere

    Retired Moderator

    Apr 20, 2004
    15,815
    282
    What do you mean by formula? That one does not resonate.

    If the device does not work the way it should, you look for the problem. Totally dead is great, as that is usually a power problem. The rest depends on how the device is failing to do what it's supposed to.

    Obscure case in point. I worked for a couple of hours on a old Univac 1206 computer (CP-642A in the Navy's terminology). It failed memory tests. I could make more errors come and go by adjusting X and Y driveline currents, but could not get rid of a failure at one address.

    Diagnosis - a ferrite core had cracked and failed. So we chucked about a zillion dollars of memory chassis and got a replacement.

    Had another computer - a 1208, or CP-642B. The multiply/divide timing chain would start to run erratically. We could relate the onset of the problem to the uppermost bit setting in the instruction register.

    Neither flip flop was in any way related, yet when one set, so did the other. The common element was the neon indicator driver card that handled both flip flops.

    The cure was a kludge. We swapped both flip flops (A14 & B12 in chassis 4) with aother pair of flip flop cards in the computer. The problem went away for several months. It did come back, but we just made up a chart of swapped cards and kept swapping. We weren't about to chase any wiring problems (had that experience in another computer. Found out how handy a crochet hook can be).

    You don't need a formula, but you do need to understand how to read a schematic and predict what things should look like if they are working.
     
  4. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    The problem with your question is that, essentially, the answer is many years of experience and learned lessons, along with a well-grounded education.

    There isn't a set of magical formulae that can solve every problem. Well, unless, you truely understand Maxwell's equations and quantum mechanics...

    Steve
     
  5. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    so why did i learn all those electronic formulas in college , if you can't use them to troubleshoot or verify something is wrong?

    To do calibrations, i did duty cycle formula to verify
    i had to adjust the trim pots
     
  6. Wendy

    Moderator

    Mar 24, 2008
    20,765
    2,536
    To give you feel when something isn't right. Troubleshooting is a catalog of skills, not just one big skill set. The more you know, the easier it is (but sometimes it will never be easy). There have been times I needed theory to argue an engineer into acknowledging his design had a flaw. Not often, but it happens. You need to be able to speak the language before you can communicate.
     
    Last edited: Nov 12, 2008
  7. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    I used

    Ohms Law: to confirm DC biasing voltages for transistor circuits.

    Gain Formula: Vout/Vin

    example 8 volts AC output / 500mV AC input = a gain of "16"

    8 volts AC output/ 800mV AC input = a gain of "10"


    TIME CONSTANTS formula

    example : R is 10K and C is .01uf

    R X C = Time

    10K X .01uf = 100micro seconds or 0.1 milliseconds

    frequency is 9.21 HZ ? ( the inverse of .1 milliseconds)

    Time constant is 63% of 100 micro seconds or .1 milliseconds

    .1 milliseconds multiplied by 63% = 63 microseconds or .063 milliseconds

    10 volts multiplied by 63% = 6.3 volts

    Time constants= 6.3 volts at 63 microseconds OF 10 volts at .1 milliseconds at a frequency of 9.21 HZ


    I got a Scintific Calculator that i don't know how to put into engineering mode or even if it has the opposition

    Of my inverse to time to frequency is messed up it seems
     
  8. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    Rise time of 10 volts is

    10 volts X 10% = 1 volt
    10 volts X 90% = 9 volts

    Rise time of 5 volts is

    5 volts X 10% = .5 volts
    5 volts X 90% = 4.5 volts
     
  9. scubasteve_911

    Senior Member

    Dec 27, 2007
    1,202
    1
    Yes, there are a nice set of equations! For now, these are built in my head and programmed. I don't even think of them as formulas anymore, as long with many others. They become second nature when you are using them all the time, kind of analogous to calling multiplication a formula. I am sure many will agree on this.

    Steve
     
  10. relicmarks

    Thread Starter Active Member

    Oct 13, 2006
    355
    0
    if the time period is 1mS to get the time constant how?

    1ms divided by 5 = time constant ? 200uSec microseconds

    frequency is 1K

    frequency cutoff of 200uS is 796.1HZ
     
Loading...