Connect with us

Help, processor reset when power on large transformer.

Discussion in 'Electronic Design' started by Rock, Apr 2, 2005.

Scroll to continue with content
  1. Rock

    Rock Guest

    I've got a simple pic based controller that turns on power via a relay
    to a 1000W transformer used for high energy lighting. When I turn the
    power on the processor ends up reseting. I've traced this back to about
    a 2 micro second low going pulse break in the power to the processor.
    Adding capacitance, and even putting in diode based filter has not
    helped. I'm guessign the emf field from the transformer is causing
    inductance via the wires on the prototype (this design has not gone to
    board yet). The power glitch happens on the output of the LM2940 5 volt
    reg I'm using, and it is supplied by a 24 volt DC off line power supply
    that doesn't glitch. I've tried various caps on the regulator with no
    help. I'm switching the power anywhere in the AC cycle, and it doesn't
    always reset. Anyone have any ideas of things to try?

    I rewrote the software so that a reset is part of the normal power on
    and off sequences for this device, if it happens. While this works I
    find it a real cop out.


  2. Most pics take very little power. Perhaps you could use a resistor +
    bypass to 'filter' the power supply? If your glitch is 2us, then you
    don't need much to keep it above water. If you use a 10 ohm resistor
    from the power supply to your Vdd pin, and a 1uF cap from there to
    ground, the supply will only drop to 82% of Vdd during the 2us glitch.

    Robert Monsen

    "Your Highness, I have no need of this hypothesis."
    - Pierre Laplace (1749-1827), to Napoleon,
    on why his works on celestial mechanics make no mention of God.
  3. Joerg

    Joerg Guest

    Hello Rocky,

    Do you have a full ground plane? If not it may be time to provide one.
    The uC and other chips should all have at least a 0.1uF ceramic to GND
    very close to their pins. No more than very few millimeters.

    Also, some low drop out voltage regulators can become a bit fickle when
    their input line impedance changes, even if these changes are caused by
    EMF. Since you have a 24V supply you shouldn't need low drop out. I'd
    consider a more classical regulator such as the LM317 series. These are
    the ones I use a lot and I never had any problems even in the vicinity
    of serious RF fields.

    Another issue to watch out for is where the lines from the uC go. If
    there aren't any series resistors or better yet, RC networks, it is
    possible that surges are induced in those lines. Then the current spikes
    often have no other path than to run into the substrate diodes of the
    uC. From there it's either into GND or VCC. If you hook up a battery
    such as four NiMH cells and the problems are still there you might have
    signal lines picking up stuff.

    Regards, Joerg
  4. J M Noeding

    J M Noeding Guest

    a reset facility has been used in almost any circuits I've seen with
    an 1M and 10-100nF capacitor grounding the reset pin when powered on,
    can't understand that such is unknown

  5. Mark

    Mark Guest

    a couple of things..

    #1 you may have a measurment issue, when then power transformer turns
    on, there is a large EMI spike and you may be seeing that on the scop
    and it is not really on the +5. If you put your scope probe on the
    ground, do yo usee the same spike?

    #2 big power xformers can saturate at turn on due to residual flux.
    This causes them to draw a very large current spike. look up
    trasnforer flux stauturation at turn on.. the only sure cure is to
    soft start it first. Maybe you can use two realys, one through a
    resistor, or maybe you can use a PTC soft start device in series.

    #3 when the xformer draws the large current spike, this is somehow
    being injected into your up. Try to keep all the power wiring away fom
    the up low level circuits. think about the current loops in the power
    wiring and arrange ther components to minimize the area of the loop.
    Example, keep the relay close to the transformer and bring the low
    level relay drive to the relay instead of the other way round.

    good luck..
  6. I agree with your distinction here, but would add: Anytime you
    see big spikes or glitches, either they are really there or some
    fast changing field (most often magnetic) is inducing the observed
    signal in the probe/scope/circuit-under-observation circuit. Even
    if the latter is responsible, it can be a good clue as to what is going
    wrong, especially if the circuit being examined is not built in a way
    that tends to make it immune from ambient fields.
    That is a good point, and very relevant. However, I will quibble
    with both the cause and the cure. When a line transformer is not
    designed to handle about twice the peak magnetic flux that occurs
    during steady state operation, and is switched on near a voltage
    zero-crossing, it can saturate, often with an audible "whump". In
    addition to the cure you mention, two others are available. One
    is to switch on near a voltage peak. (This is counter-"intuitive",
    so please consider carefully before contradicting this.) Another
    is to specify a transformer with enough more iron that it will not
    saturate upon turn-on regardless of the starting phase.

    Another aspect of transformer saturation often overlooked is
    that the external magnetic field from a saturating transformer
    will far exceed that which occurs when the flux is much more
    confined to the intended, highly permeable magnetic paths.
    This effect can be used as one of the tell-tales of saturation.
    All good advice.
  7. Fred Bloggs

    Fred Bloggs Guest

    More generalized bullsh_t from a narcissist who imagines himself to be
    the great teacher- but fails miserably because he doesn't have the
    intellectual horsepower to put anything into practice.
    Total bullsh_t once again- every damn transformer made will draw
    considerably more current and nearly saturate when starting from zero
    flux density...most reasonable people make their circuits tolerant of
    this effect rather than over-specifying the transformer, idiot.
    Well- yeah, yeah, yeah- ya' friggin idiot, it is a simple field divider
    circuit where the field divides in accordance with the reluctance- this
    is simple Electricity 101...
    And as usual, you add nothing but garbage hot air and non-information...
  8. Fred Bloggs

    Fred Bloggs Guest

    This is almost certainly being caused by the relay and its solenoid
    field cutting the area formed by the 5V regulator to Vdd to GND and back
    to the regulator. You can verify this by supplying power to the relay
    through two twisted wires as well as moving it away from the PIC circuit
  9. Pooh Bear

    Pooh Bear Guest

    Sounds like it's possible poor pcb design ( loop areas for example ) or
    lack of attention to general EMI issues.

    There's tons of stuff written about how to do it right but I guess you need
    to read some.

    You won't be the first to experience this kind of issue for sure.

  10. Pooh Bear

    Pooh Bear Guest

    Hmmm, thinks. Is the 5V regulator close to the PIC or far away ?

    It *should* be close to the PIC.

    Watch out for common ground current paths too.

  11. Derf transform applied.

    Many "made" line transformers do saturate when turned
    on near a voltage zero-cross. Depending on the regulation
    of the transformer, currents 20 times larger than the inrush
    under more favorable conditions, or more, will flow. This
    is not usefully similar to what happens "near saturation".

    The problem is one that leads to reduced fuse life, (even
    when slow-blow fuses are used), and is clearly addressed
    by those vendors who do design the magnetics to handle
    worst case starting peak flux. I have looked at a number of
    transformers when dealing with the problem, and discussed
    it with at least two transformer designers who worked for
    reputable transformer vendors.
    The point, perhaps not clear to you, is that the OP may
    want to use the effect to see if saturation is occurring.
    As for it being elementary, I have worked with many
    competent engineers who had not worked out all the
    details and ramifications of how magnetic devices work.

  12. Fred Bloggs

    Fred Bloggs Guest

    Yeah- so what- they're line operated and the saturation is "soft"- no
    one seems to have a problem with it except a dodging pseudo-intellectual
    and unproductive weakling like you- looking for *any* excuse for not
    Yeah- right- BULLSH_T!

    Oh yeah- there you go again with your assumptions about everybody being
    P.O.S. ON S.E.D...BULLSH_T...

    Optimized manufacture of line transformers requires that saturation is
    approached from zero residual core flux- anything else would be as much
    of a waste of material as you are of time. GFY.
  13. Genome

    Genome Guest

    Well Larry, once again your delivery leaves a lot to be desired.

    You agree there is a problem but you want to quibble about the cause
    and the cure. Then, not only do you not properly explain the source of
    the problem or why your cure works. You choose to issue a challenge to
    anyone who questions your statements.

    Surprisingly when Fred calls you on it you choose not to take the
    opportunity to rub his nose in it by explaining the details behind the
    cause and cure. No, not you, you decide to start talking about your
    experience with other 'experts' in the field and bring up some other
    subsidiary effects.

    Do you want to explain the cause and cure or shall I do it for you?

    Oh, I will assume you don't know if you decide to respond with
    anything which is not a correct and succinct answer.

  14. [Previous quotes repaired for inadvertent line-wrap.]

    The reason for quibbling about the cause is because it has
    clear implications regarding the possible cures.

    The "proper explanation" of the phenomenon is easily seen by
    anybody who understands the intregral relationship between
    voltage applied to an inductor and current flowing thru it. For
    purposes of getting the OP's problem solved, there is no need
    to launch into some tutorial on such a topic. In my opinion,
    when I posted and now, there still is no such need. So I deny
    your claim that I did not "properly explain" the problem.

    What you call my "challenge" is nothing more than a request
    for some thought before anybody jumps in with ill-considered
    "common sense" or other forms of "what everybody knows".
    The notion of either educating Fred or shaming him with his
    occasional ignorance is humerous. Your surprise shows
    only your misunderstanding of my objectives here.
    My mention of some people who could be presumed to be
    experts (and in fact were experts) was in support of my claim
    that some transformers are designed to handle the doubled
    flux level that can occur upon turn-on. I have no idea what
    "other subsidiary effects" you are going on about..
    Go ahead if it thrills you or you believe anybody really
    wants to see that sort of education.
    Assume away. It has no bearing on what I know.
  15. Genome

    Genome Guest

    Very kind of you, we certainly like to follow what's being said in a
    Larry Brasfield thread.
    So, I don't suppose you'd like to explain those 'clear implications'.
    You didn't quite do it but you practically told Mark he was wrong. Are
    you now saying that he may have had a point? Do you want to include it
    in the explanation you are not going to give?
    But you wanted to, and did, talk in terms of flux. Fair enough there
    is a relationship but, again, you're choosing to avoid a full
    explanation. You start to give a partial explanation but next..... you
    say you won't.
    Hang on, you were responding to Marks post. There's a not too subtle
    difference here. You tell Mark he is wrong and then you refuse to
    explain why because someone else asked a different question.
    But...... you didn't..errr.... so you can't. Deny it that is.
    Like I said 'Your delivery leaves a lot to be desired'. I have
    considered the problem and I did not jump in. The way you phrase
    things just invites it. You could have explained yourself and avoided
    getting into a nice long thread. (cough, but you enjoy it)
    I wasn't looking for objectives or motives. Just clear precise
    answers, I am quite foolish..... aren't I?
    OK, on second reading I'll give you that. However you might like to
    group your information rather than spread it about. It makes
    interpretation and understanding easier.
    That's a bit pretentious isn't it? I'm sure there are people who would
    like to know.
    Well that appears to be a half hearted attempt at something that
    wasn't a correct answer. And it wasn't succinct. So you don't know.

    By the way, how do you manage to write so much stuff with so little
    content so quickly?

  16. I read in that Genome <>
    He's under then impression that those laxative tablets you sent him are
  17. I find his long posts hard to follow. I tend to nod off, and wake up
    with a start while reading Fred's inevitable reply.

    However, I'm guessing he feels like he has to explain every little
    detail in order to forestall attacks. It's like reading a legal brief.
    He is preparing for the inevitable court battle. I wonder why he cares
    so much? As Larkin pointed out, usenet is just an RPG.

    Robert Monsen

    "Your Highness, I have no need of this hypothesis."
    - Pierre Laplace (1749-1827), to Napoleon,
    on why his works on celestial mechanics make no mention of God.
  18. Fred Bloggs

    Fred Bloggs Guest

    As is usual with you wise-ass morons who aspire to *teach*, you can't
    *do* a f_cking thing...this is too classic.
    "Humerous"? So in addition to being totally incompetent and uneducable
    in technical matters such as electronics- you can't consistently spell
    at the 4th grade level. It has been pointed out to you time and again
    that what you consider a revelation in theory to be announced to the
    whole NG is trite and elementary trash. You are a pretentious little
    zero who can't produce anything worthwhile.
    Yeah- well whatever- those would be custom jobs for such a small number
    of applications that they are representative of nothing and no
    manufacturer could afford to maintain off-the-shelf inventory for them.
    This is just another example where you have to allude to something that
    makes you quite *special*- when all you do is confirm that you are an
    e-*special*-ly worthless reject.
    You don't really "know" anything- it is clear that a long time ago you
    started compensating for the lack of intelligence enabling you to
    *comprehend* with some sort of schizoid fantasy you live by which you
    "know"- pathetic really.
  19. message
    [Apparently quoting the OP]
    Most relays have a closed magnetic path going through
    iron everywhere except for small construction gaps and
    the gap which varies to operate the relay. Allowing much
    flux into the surrounding air is bad for the efficiency, and
    would be a sign of a poorly designed relay, if it happened.

    I have used quite a few relays in close proximity to other
    circuits without seeing the magnitude of effect you claim,
    including circuits made with single-layer etch for cost
    reasons. So I am quite skeptical that your "certainly"
    is well founded in reality.
  20. Cute. While your insinuation is superficially plausible,
    my note reflects only my intention to quote accurately.
    Your post, when quoted normally, was a mess. I've
    done the same repair again here. Why not just adjust
    your line-length limit and get past this?
    Is that not obvious by now? Do you honestly believe that
    anybody who cared was unable to make that connection?
    I disagreed in a minor respect with his assignment of cause.
    I thought Mark had a good point, and stated as much.
    Yep. The light appears to dawn. I am not going to be giving
    tutorials on magnetics, basic RLC circuit theory, or anything
    else without good reason and a favorable cost/benefit ratio.
    I had no reason to believe Mark even wants or needs such
    an explanation. For all I know, once he thinks about the
    issue for awhile, it is as obvious to him as to me. Maybe not.
    If not, he is certainly free to ask why I quibble with his cause.
    I did deny it. You may not be happy with my proof,
    but my denial is in plain English. Get a dictionary.

    [Silly baiting banter cut.]
    That paragraph could have been clearer. With more
    time invested, it might have been.
    If you are so sure, why do you not follow thru on your
    implied promise to educate them? My own estimate is
    that few if any who read it would learn anything from it.
    And playing your silly game is not rewarding enough to
    justify the time expenditure.
    So you say. Of the dozens of reasons I might have for
    not expounding on something, you choose 'ignorance'
    without the slightest evidence. That choice reflects your
    state of mind, not mine.
    It's partially a function of the audience. An ARRL member
    once explained to me, as a boy intrigued by his transmitter,
    why receivers were much more challenging to design.
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