Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Tools needed to analyze circuit boards

  1. Dec 27, 2011 #1
    I come across many circuit boards at my job and I while I have to understand exactly what the board should be giving as inputs and outputs I don't have to understand what's the board is doing within it. For instance, I know that a board should have a high voltage switch close when another area of the board has a control voltage applied so if it doesn't do that I simply replace the board. But I never really understand why it's not working, mainly because that's just not my job. Also we aren't allowed, by manufacture's orders, to repair the board but only replace it.

    With that said, over the last couple weeks I've brought home several boards to try to analyze them but it's not going well. I have a multimeter and some transformers to apply certain voltages but I'm at a loss when it comes to getting into the "brain" of the board like timers, transistors and the programming of it. What tools can help me dig deeper into a circuit board? Thanks.
     
  2. jcsd
  3. Dec 27, 2011 #2
    To do simple troubleshooting it would be very helpful to have a digital signal source. The simplest thing I can think of is a 555 running in astable mode and a binary counter that uses the same logic levels (voltage) as the logic on the boards. The 555 can drive the binary counter which will give you multiple outputs to feed into the various ICs on the circuit board. You could stop the binary counter at any combination of inputs to the IC that you want. Then with your multimeter you can check the output of the gate to verify that it has the output it should have. Do you have the ability to build a circuit like that?
     
  4. Dec 27, 2011 #3
    I have some doubts building something like that. Are there any instruments that test a board in the way you described? Anything I've found online seems to lead to a logic analyzer but I'm not exactly sure if that's what I need. I'm willing to throw some money at some tools because I'm an EE major and can only benefit from learning this stuff.

    Also, I'm unsure of how to test transistors with a multimeter. Is there a neutral or ground I can test voltage from to the transistor or is there a certain way to test them? The reason I ask is because I have one very simple looking circuit board and the other 2 are much more complicated so I want to savor this simple board to try all my tests before things get way over my head.
     
  5. Dec 27, 2011 #4
    I don't have enough information to suggest specific equipments. You mention more than once you have transistors, power supplies, in general, you need a oscilloscope, a function generator, good soldering iron, a digital multi-meter as a minimum start. Unless you have absolute need, logic analyzer is not needed. decent digital oscilloscope will do the job most of the time. I just bought a multi function generator for like $100 used. You can get a new one for under $200 and that will get you pulses. These should cost you around $1000 or a little over if you get in surplus store.

    To troubleshoot your boards, try getting a set of schematics for the boards.
     
  6. Dec 27, 2011 #5
    Agreed 100%.
    If the boards have much SMT, a good magnifier will help a lot as well.
    You can start reverse engineering the board as practice. By using your DMM and looking up datasheets, you should be able to draw the circuit (for the most part).
     
  7. Dec 27, 2011 #6
    Most gates have more than one input and that's why I suggested a signal generator with multiple outputs. However he could bias one input high or low with the power supply and inject a signal from a signal generator into the other and check the output with a multimeter. Personally for digital circuits, I prefer an analog voltmeter because the actual voltage isn't as important as knowing whether it is high or low and that's a lot easier to see with an analog meter.

    If you want an oscilloscope, one way to go would be to get a multimeter with an oscilloscope function, for example: http://www.storeinfinity.com/new-digital-multimeters-scopemeter-oscilloscope-ut81b.html

    I have no experience with that model or any other and am not recommending anything in particular, but a multimeter with a scope is a combination you probably won't outgrow.

    When I was studying electronics, I had a second shift job repairing logic boards for a company that made electrostatic copiers. I learned more on the job about digital logic than I did in school.
     
  8. Dec 27, 2011 #7

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    you'll have to get schematic drawings and figure out how the board responds to every input.

    then build a tester that applies such inputs.

    you don't own a machine until you've taken it apart and put it back together, learning what every part does.
     
  9. Dec 27, 2011 #8
    I asked the manufacturer for schematics but they politely denied giving me them. They told me it comes down to a safety issue because the boards that I see are used for HVAC systems and have many connections to safety devices which they don't want technicians messing with.

    Are there any specific oscilloscopes and function generators I should be looking for? They seem to have quite a range of settings. The boards I work on are 120V main power with 24V controls, I'm not sure if that's relevant information or not.
     
  10. Dec 27, 2011 #9
    It's all about budget!!! Go to surplus store to look around. I am more on analog side, I am happy with my old Tektronics 465 scope. I won't buy a lot of equipments just for a specific project. You have plenty of chance to use them when you work in the future.
     
  11. Dec 27, 2011 #10

    es1

    User Avatar

    This is very relevant as probing main power can be dangerous to both you and your equipment. You should check out this thread and the links in it. If you're probing a board powered by mains and you don't know the schematic you'll need to be doubly cautious.

    https://www.physicsforums.com/showthread.php?t=505849

    Since the board is wall powered I bet it has what is basically an adapter on it.
    Check out this. Lot's of good trouble shooting information in it.
    http://www.repairfaq.org/sam/aapsfaq.htm

    Also, since you have a multimeter put it into ohm mode and while the board is unpowered check all the fuses for continuity. I'll bet you can repair many of them by just swapping in a good fuse. Won't really help with the understanding part though. :)
     
  12. Dec 27, 2011 #11

    jim hardy

    User Avatar
    Science Advisor
    Gold Member
    2016 Award

    are you speaking in-circuit or with the device removed and in your hand?

    here's how for ordinary transistors, but not FET's.

    if you have the part unsoldered from board and in your hand,,,

    IF you are using an analog multimeter (Simpson 260 is my favorite)
    set meter to RX100 scale NOT to RX1 scale.

    If using a digital multimeter set it to the diode scale


    read resistance between all 3 leads, both directions. With Simpson it's handy to use the +/- switch to swap polarity.

    none of them should be shorted
    emitter to base should read open in one direction
    and about 0.6 volts in other.
    DMM on diode scale reads volts directly so open circuit is probably 2 volts (4 on some meters),
    , analog Simpson should show about 2/3 scale one direction(a few hundred ohms), almost no deflection in other(very high ohms).

    collector to base should read very nearly same as emitter-base .

    With analog Simpson, collector to emitter should read very high ohms both directions.
    With DMM, it should show same as when leads are not connected to anything.

    Now with analog Simpson you can do one more trick.

    moisten your index finger.
    Touch wet fingertip across meter leads and note small deflection of needle.
    Now connect meter from collector to base, and touch wet fingertip across collector-base. You should see meter needle deflect to right more substantially.
    That's because transistor amplifies the leakage current your finger allowed into its base.
    When you get good at this you can estimate current gain from the wet finger test.

    If you swap polarity of meter and repeat wet finger experiment you can tell that the transistor works better forward biased than it does reverse biased. I dont know what use that is, but it's interesting trivia that a transistor will show some current gain even when soldered in backwards. I used it once to identify polarity of an unmarked transistor....

    that's your most basic transistor test out of circuit and it's a go-nogo for finding failed parts. Transistors usually fail by shorting a junction so zero ohms is a giveaway. If the circuit allow enough current flow to melt the internal wires you'll have an open circuit.

    some old high quality equipment uses transistors with metal cans (2n2219 series) which you can cut open with a Dremel and look inside. With a strong magnifying glass sometimes you'll see the wire burnt in two, sometimes just broken. Now that's getting right down to failure mode.....

    i hope that's what you asked, and please excuse the digression. old firehorse in me - spent too many years troubleshooting and like to help newbies develop skill.

    testing them in-circuit is more complicated but not difficult, and another subject.


    will stop now. old jim
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Tools needed to analyze circuit boards
  1. Analyzing Circuits (Replies: 8)

  2. Circuit boards (Replies: 5)

  3. Circuit Board Problem (Replies: 24)

Loading...