Announcement

Collapse
No announcement yet.

When you're building something, read the datasheet 100 times.

Collapse
X
Collapse
 
  • Page of 1
  • Filter
  • Time
  • Show
Clear All
new posts
Previous template Next
    #1

    When you're building something, read the datasheet 100 times.

    I made a PCB for my little 3842-based flyback PSU which uses a 5vsb transformer from an ATX PSU. It didn't work. The 3842 didn't output Vref, so the oscillator wouldn't start, hence no drive to the MOSFET -> no output.

    I checked the PCB a dozen times and found nothing suspicious. I spent over 3 hours trying all the 3842s i had, i think i went thru 8 or 10 of them. Note to self: sockets.

    Then, before going to sleep, i thought i'd look at the UC3842 datasheet another time. UVLO threshold for the UC3842 is 16v, and i had a 15v zener on the 3842's Vcc pin. Son of a... Sure enough, i disconnected the zener and the supply fired right up. Now, i don't recall why in the world did i put a 15 volt zener there, but i did. It says so on my schematic too.

    So... when your circuit won't start, read the datasheet one more time. You might just catch that brain fart you had when building the thing.
    Originally posted by PeteS in CA
    Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
    A working TV? How boring!
    Tags: None
    #2
    Re: When you're building something, read the datasheet 100 times.

    Yea it's often very difficult to catch your own mistakes, it's just the way our brains work...
    For this reason at my work the guy making a CAD drawing for a machine may never be the same guy to measure the part after it has been produced, he will just make the same mistake when measuring as when drawing... (Been there, done that)

    So if it's possible for you to get a second opinion from someone else that's usually the best bet
    "The one who says it cannot be done should never interrupt the one who is doing it."

    Comment

      #3
      Re: When you're building something, read the datasheet 100 times.

      At least you didn't blow up the UC3842, like I did the other day!
      Muh-soggy-knee

      Comment

        #4
        Re: When you're building something, read the datasheet 100 times.

        We have progress! Circuit is nice and stable with a 1A load on the output (quick n dirty compensation too, haven't even tweaked it yet, will try other loads too as it should do 3A), and i've arrived to the final values for the RCD clamp on the primary.

        Now the last thing i need to do is a RC snubber for the diode as well, and we'll have a really clean PSU. Ripple is already very very low, as a matter of fact, the only thing i see on the output, after the pi filter, is the spike from the diode.
        Originally posted by PeteS in CA
        Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
        A working TV? How boring!

        Comment

          #5
          Re: When you're building something, read the datasheet 100 times.

          One general way you can check stability and transient response would be to create a dynamic load with a load resistor in series with a MOSFET driven by a 555 timer free running at somewhere around 100Hz-1KHz. That assumes you have access to an oscilloscope. Looking at the O/P voltage in AC coupling, if you see oscillation, ringing or a large overshoot in both directions, you probably have marginal stability or the circuit is not stable. On the other hand, if you see undershoot (when the MOSFET turns on) and overshoot (when the MOSFET turns off) your bandwidth is too low (i.e. the regulator is too slow). It's not as definitive as doing stability plots, but it's easier and quicker if you want a general idea.
          PeteS in CA

          Power Supplies should be boring: No loud noises, no bright flashes, and no bad smells.
          ****************************
          To kill personal responsibility, initiative or success, punish it by taxing it. To encourage irresponsibility, improvidence, dependence and failure, reward it by subsidizing it.
          ****************************

          Comment

            #6
            Re: When you're building something, read the datasheet 100 times.

            I know about that. I just drive the MOSFET connecting the load from my computer with audio generator software.
            Originally posted by PeteS in CA
            Remember that by the time consequences of a short-sighted decision are experienced, the idiot who made the bad decision may have already been promoted or moved on to a better job at another company.
            A working TV? How boring!

            Comment

            Previous template Next
            Working...
            X