Voltage peak analyzer

Discussion in 'The Projects Forum' started by spidy634, Oct 16, 2014.

  1. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    Hi all, I am looking forward to build a simple test rig for following test.
    Test: Disconnect the appliance from the supply mains at the instant of "voltage peak". One second after disconnection, the voltage between the pins of the plug is measured.

    I have an Oscilloscope to analyze voltage however as you know since the supply mains freq is high 50Hz, we cannot determine exact voltage peak occurence time (for every 5ms which is small time). Is there any instantaneous voltage analyzer?. I need to prove that I have disconnected the appliance socket exactly at voltage peak?? May I please know how could I achieve it.

    Any suggestions would be appreciable. Thank you.
     
  2. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    what kind of aplience? if it has a motor or transfo0rmer in it, you will get a swurge or spike when you disconnect at peak. 1 second later, nothing will be happening, ac cannot be stored to bleed off like dc.
     
  3. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    Hi alfacliff,
    Thank you for the reply.
    Appliance comes under household appliances category and has both transformer and motor in it and is fed by mains supply. Since the capacitance in the circuit has the capability to hold charge, the aim of the test is verify if the pins of socket plug holds any charge after removal of plug from mains socket (safety functionality test to avoid humans getting shock if they touch plug pins immediately after removing from socket). Its true that after 1 sec (whch is long time for ac pulse) the charge might not be retained, but the time limit 1sec might be set coz it might be estimated that it takes atleast one sec for human to touch the pins after removing from plug. My problem is I need to prove that the socket is pulled off or switch turned off exactly at the mains ac peak voltage. I can connect a voltmeter across socket pins while plugged into mains and verify voltage across the pins immediately within 1sec while power input turned off however unless I prove that I have turned off exactly at ac voltage peak my test is invalid. Issue is how to prove that I have turned the switch off eaxctly at ac voltage peak.

    Please let me know if I am unclear and pardon me if there are any mistakes.
     
  4. Nykolas

    Member

    Aug 27, 2013
    87
    31
    When you state "voltage peak" do you mean the crest of the mains sine wave? If so, you could use a zero-crossing detector and a 1/4 wave delay of your mains frequency and an electronic disconnect (allow for any delay time in the circuit). A relay would not switch fast enough and the switching cycle times vary.
    If your scope is mains powered you could do serious damage to the scope hooking the probe across line and return (I speak from experience!), so, as always, use an isolation transformer! E
     
  5. alfacliff

    Well-Known Member

    Dec 13, 2013
    2,449
    428
    the primaqries of the transformers and windings of the motor will bleed off any voltaged retained by the cvapacitors in the circuit before 1 second has elapsed. so will any lamps or other loads. power cannot move backwards through power supplies to supply source when the power is removed.
     
  6. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    Define what you mean by "exactly". For instance it might be relatively easy trigger a disconnect at a voltage within, say, 0.5V of the peak voltage or within a few milliseconds of the predicted next peak. But it gets harder and harder as you narrow in on either the voltage or the time.
     
  7. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    Hi Wayneh,
    Theoretically its exactly after 5ms of the wave start as the freq of the mains is 50Hz.
     
  8. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    Hi Nykolas the solution seems feasible, however unfortunately I got limited experience in circuit design and this is my starter project. I have looked into some zero cross detectors, however could you please suggest me about wave delayer and delay time cutoff switch. coz I found it hard to find a switch to cutoff at that time.

    However I am just wondering, cant I connect the oscilloscope to the mains and the socket, turn off the power and read the wave characteristics instantaneously. Might need to do trail and error tests, repeat couple of times to get desired waveform.
    Please advise.
     
  9. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    That could be a good brute force approach. If you manually switch it off enough times, you may end up with a few attempts that are close enough to the peak to satisfy your needs.

    But if you need something better, it would not be terribly difficult to 1) Detect a set point voltage, say 10V, 2) Upon that detection, initiate a timer set to trigger in the right amount of time, a few ms, and 3) Trigger a disconnect. It would take some tweaking to get it to hit the "exact" moment you need, but it is completely doable.

    For a one shot test, you'd get there faster with the brute force approach.
     
  10. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    Hi guys thanks for your help. atlast I came to a solution which I would like to share - might be helpful to anyone. please let me know if you got any better solution.

    Solution: I worked out this circuit in simulator and haven't tried physical circuit yet. I connected mains voltage directly to load and negative of load to a MOSFET switch. I controlled this MOSFET switch with a waveform generator generating wave with 50Hz freq and 25% duty cycle. So when I turned the circuit on, the switch is ON only for 25% cycle of the input wave and turns off immediately when input wave reaches the peak value of the waveform.
    Hope thats clear.
     
  11. wayneh

    Expert

    Sep 9, 2010
    12,145
    3,055
    Oh dear, that might get this thread closed.

    How will your circuit follow the waveform in the real world, as opposed to your simulation? I guess you're planning to replace the signal generator with a divided-down signal off the mains?
     
  12. spidy634

    Thread Starter New Member

    Oct 16, 2014
    6
    0
    To be honest due to my limited experience, I got no idea yet, I am still doing research. I just managed to work it through simulator till now,.Now the exciting task begins which is to build it phisically and efficiently. Even though I can use waveform generator to trigger the transistor all the time in my lab, I would love to use some kind of crystal oscillator which works at freq 50Hz with 25% duty cycle. Unfortunately I cannot get hold of oscillator at that frequency.
    Any support would be appreciated.
     
Loading...