Connect with us

scr latchup CMOS failure?

Discussion in 'Electronic Basics' started by John Fields, Aug 10, 2004.

Scroll to continue with content
  1. John Fields

    John Fields Guest

    Dunno, but...

    1. What are you using for Vdd?

    2. Are you using 4538's or 74HC4538's?

    3. You could be frying the 4538 because you don't have Schottky diodes
    across the the timing resistors (cathode to Vdd) and the solenoid is
    pulling the supply down, allowing the timing caps to dicharge through
    the chip's protection diodes (yes, the dreaded parasitic SCR's rear
    their ugly heads...)

    BTW, did you get the circuit I emailed you?
  2. andy

    andy Guest

    I've been testing the watering circuit i built:

    It was working OK with an LED in place of the coil, and a 10 uF capacitor
    instead of the 1000 uF one. I've just finished building the MOSFET switch
    part of the system, which works OK by itself.

    I just tried testing it with the coil and the 1000uF capacitor, and now
    the 4538 is not working right and getting very hot when I switch the power
    on. Switching it off and leaving it doesn't fix the problem. Could this be
    a case of SCR latchup failure? The only way I can see that this could be
    happening is if the fully charged 1000uF capacitor is discharging through
    the 4538's A1 and B2 inputs. Is this likely to be the problem?
  3. andy

    andy Guest

    after switching off, i mean. the main symptom apart from overheating is
    the 393 can't pull down the A1 and B2 inputs of the 4538 any more - it
    goes down to 10.63V and no further. With the 4538 disconnected, it works
    fine, switching between 0V and +12V.
  4. andy

    andy Guest

    thinking about it, this is unlikely to be the problem - the capacitor
    would also be keeping the supply rails high while it discharged, so it
    wouldn't be driving the 4538 past the rails.
  5. andy

    andy Guest

    a 12V 1.2Ah lead acid battery.
    a 4538.
    That could well be it - when the battery is low, the supply gets pulled
    down to around 7V. Maybe need a bigger battery.
    Yes, thanks. It looks like it would work OK, but I'd rather stick with the
    design I have now.
  6. John Fields

    John Fields Guest

    Why? if 1.2AH will fire the thing under worst case conditions, day in
    and day out, and you can be sure that the PV array will always keep it
    topped off, then you should be OK. Just put some diodes across the
    resistors. You might also think of replacing the lead-acid (SLA?)
    battery with NiCd or NiMH for the lower internal resistance under
    pulse load, but you should install the diodes in either case.
  7. andy

    andy Guest

    I want to build the one I worked out for myself, not just copy
    another design. I'm doing this to practice/learn how to, as much as to
    have a working system.

    and mine has 2 chips to your 3, so nerrr. ;)
  8. I assume you've double-checked that your 4538 (and other) connections
    are wired up exactly as intended? And that the chip is inserted with
    correct polarity! Have you swapped it for a fresh 4538, with same
    result? Have you used the 4538 in a basic test circuit to check it's
    not already zapped?

    BTW, I got this message on trying that JPG link:

    Too Many Users

    There are more clients using this WebWasher than your license allows.
    Please upgrade your license to support more users.
  9. andy

    andy Guest

    No - need to get another one.
    It was working before, so don't see the need for that.
  10. John Fields

    John Fields Guest

  11. andy

    andy Guest

    Is there any reason why they have to be schottky diodes? The only ones I
    can find in Maplins are meant for power rectification. (1 amp plus max
    current). I'm thinking of using germanium signal diodes instead.
Ask a Question
Want to reply to this thread or ask your own question?
You'll need to choose a username for the site, which only take a couple of moments (here). After that, you can post your question and our members will help you out.
Electronics Point Logo
Continue to site
Quote of the day