Can an A/C motor swiching circuit using Triacs activate itself

Discussion in 'General Electronics Chat' started by Alive, Jun 3, 2011.

  1. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    I have an intermittent problem with a hospital bed controller circuit. Unfortunately I don't have a schematic so I will attempt to describe with the little electronic experience I have.
    It is powered by 120V a/c and swiths 3 a/c motors using Triacs. The voltage is transformed and rectified to 5V d/c, and then sent to the hand held control for switching between 6 functions. The signal from the hand held is then sent to a microprocessor that outputs to an optoisolator triac driver. The triac then switches the 120V a/c to the appropriate motor for head up or down, bed high/low or foot up/down. There are snubbers on each of the out puts.
    The patient in the bed claims a ghost is pushing the buttons and the bed, foot or head will just move to full high position on its own. I have several beds now doing this but not often and some only once. When I put the suspect contoler boxes on the test bench I can't get them to malfunction.

    My question then (eliminating the hand control as the culprit) Can these strait forward simple circuit boards switch themselves to activate a motor?
    I have seen triacs switch on and stay on or not switch at all but what could be causing them to occasionally activate?
     
  2. ErnieM

    AAC Fanatic!

    Apr 24, 2011
    7,395
    1,607
    They put as micro in this thing? Oisch... if that hiccups then anything can happen. Hopefully whoever designed it did their homework and that is not the problem.

    I would guess it is the mechanical switches in the hand held. How old are these units? An intermittent switch is the hardest thing to find as they only exhibit the problem, well, intermittently.

    We almost lost my wife's car when the seat position switch shorted out. It drained the battery dead, then heated again when I put it on the charger. Good thing it buzzed on the charger or I would never have found that!

    That your motors only go one direction (UP) sounds like a significant symptom, but of what I don't know.

    If it was the triac failing I would expect it to fail either open or short, not intermittent.

    Can you swap the hand held controller from bed to bed? If so you can see if the problem stays in the bed or follows the controller, as long as your patients don't mind being "experimental subjects."
     
  3. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    If the snubber is not well designed dV/dt transients can trigger the triac.
     
  4. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    Thanks for the reply ErnieM.

    The reason I do not suspect the hand control is that if you remove the power from the board and plug it back in you regain normal operation of the bed.

    Tom66 - I suspect the snubbers but the boards are possibly 5 years old and working correctly until now. I have tested the circuit and there are transient voltages at the outputs - enough to light a small test lamp.

    Is it possible that the Snubber capacitor takes a surge large enough to put a hole in the dialectric and then reseals itself? Excuse me if I am demonstrating my ignorance.
     
  5. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    I don't know that much about triacs, but could glitches on the AC mains cause issues? If you have a lot of beds being turned on and off you could have spikes on the mains which could be triggering the triacs.

    Yes some types of capacitors can heal themselves, although they do lose a small amount of their capacitance.
     
  6. #12

    Expert

    Nov 30, 2010
    16,346
    6,835
    I just logged on to confirm the self healing capacitors. I had a new poly-film cap (Sprague Orange Drop) actually blow a hole in its outer plastic coating when it first met 315 Vdc. The circuit never glitched and I would never have known about the incident if the cap wasn't facing the right way to see the hole.

    Oil filled caps also self-heal, but I've never seen the evidence in real life.
     
    Alive likes this.
  7. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0

    These caps are also poly film .1k 630VDC. If they heal that quickly I can't see how the current could pass through the cap and a 39 ohm resistor to complete the circuit and stay engaged long enough to raise the bed. The Triac must be getting triggered and held in a latch position some how, as Tom66 indicated. These beds are always on, meaning there is power to the boards constantly waiting to be switched through the triac. A surge in the mains could be causing the problem.
     
  8. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    I should test the mains surge theory. Anyone have any ideas on how? I can pick a board that has acted up but I don't want to fry it - they cost $300.
     
  9. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    Use a 115V/230V to 12V transformer and an oscilloscope.

    Carefully connect the ground of the scope to one tap and the input of the scope to another tap of the transformer. DO NOT EVER connect the scope to the mains directly; you will likely cause a small explosion and wreck your scope. If you have a digital scope you may be able to key in the division factor of the transformer so you can measure the waveform exactly. You should be able to see whatever noise is on the mains using this set up, although the transformer will attenuate some, so consider getting a proper 100:1 scope probe (about $50.)
     
  10. #12

    Expert

    Nov 30, 2010
    16,346
    6,835
    The problem with scoping the mains is that the ground clip of the probe is, well, grounded! If you connect it wrong, the scope will blow the 120 volt fuse (or open the 120 volt circuit breaker). You have to be very careful about that or just use ways to never get connected directly to the mains. Better safe than sorry.

    Even with a 10 to 1 scope probe, I don't have enough attenuation in my scope to see the 120 RMS wave and the ground clip is still a direct short to ground.
     
  11. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    Looks like I need to go shopping for a probe. Thanks for the replies. Repairing these boards are paying for my electronic education. There are over 250 at this long term care facility. I want to understand these boards inside out. "They are proprietary" says the manufacturer and will not supply any assistance.

    thanks again
     
  12. #12

    Expert

    Nov 30, 2010
    16,346
    6,835
    Welcome to the world of reverse engineering.

    My method is to draw a schematic of the board by looking at both sides and drawing every part and connection. It takes a long time and a lot of paper as you add parts and paths, then re-draw them in an orderly fashion, and when you get done, the microprocessor will still be a "black box". With 250 boards, it is probably worth the effort.

    Good luck. You have your work cut out for you!
     
  13. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    Transformers are isolated magnetically. There is no harm in connecting your scope ground to a tap (assuming the transformer is good! It could have a short from primary to secondary, though I'd wager that would be a rare thing); it just ties the tap to that potential. But if you connected it to live or neutral you could cause some sparks to fly.

    I suggest a 100:1 because 115V becomes 1.6V amplitude (3.2Vp-p) and any scope can display that. However, a decent one is crucial (I suggest BK Precision as I've heard good things about them, but around $68 for a good 100:1.) If a cheap one breaks down, all 115V will go straight into your scope. Not pretty.
     
    Last edited: Jun 3, 2011
  14. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    Also, you should be okay in leaving the ground clip disconnected with a 100:1 direct to mains, but can another member confirm? I have never been brave enough to probe mains directly with my scope.
     
  15. Ron H

    AAC Fanatic!

    Apr 14, 2005
    7,050
    657
    Does hospital equipment generally have secondary (DC) ground connected to earth ground? if so, doing what you suggest could short out a transformer winding or some other component.
     
  16. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    Good point! Make sure to test this with a multimeter.

    If you use a transformer ripped from an old stereo or something you could get away with it as it would be separate.
     
  17. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    Scoping the mains may be getting away from my expertise. I might be better off consulting with the Building Electrical Contractor before I proceed or loading up a board until I get a triac triggered. .
     
  18. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    What is your budget, you can pick up a good isolated multimeter type scope for less than $200 these days.
     
  19. Alive

    Thread Starter New Member

    Jun 3, 2011
    7
    0
    Do you have any particular suggestions on a scope? I'm committed to do and get what I need. I have a couple of multimeters and an LCR meter that does ESR measurement. Although I've been doing industrial switching circuits to earn some cash I want to move in the direction of audio electronics. I knew that a scope would be the next purchase.
     
  20. tom66

    Senior Member

    May 9, 2009
    2,613
    214
    Look around for a electronic equipment surplus store, I found one within 10 miles of my place and I bought my HP 54501A for £150 there. :)
     
Loading...