overvoltage protection circuit

Discussion in 'The Projects Forum' started by unlv007, Apr 11, 2008.

  1. SgtWookie

    Expert

    Jul 17, 2007
    22,201
    1,809
    if you really want to ensure that there are no rapid transients, all you really need to do is add a 10 ohm resistor between the current supply and the circuit i posted above, and a 1uf cap from the output side of the resistor to ground. that will remove the possibility of high frequency transients due to the rc time in conjunction with the constant current supply. with no output load, it will take roughly 1.04ms to charge a 1uf cap through a 10 ohm resistor up to 62 volts with a 60ma supply, or 60v per millisecond, or 1 volt per 16.666...microsecond.

    of course, adding a capacitor in the circuit will cause your project to have only an average of 60ma current through it. if the resistance of your project should decrease at a very rapid rate, the 1uf capacitor would discharge it's stored power through your load. in order to avoid that, you could add a 2nd current limiter circuit on the output of the voltage clamp circuit.

    or, you could really avoid all of this by simply using a power supply where you can set the voltage and limit the output current to 60ma, rather that one that has an impractically high limit. as it is, having the voltage clamp after the current supply is a bit like having the cart before the horse. usually, when building a linear power supply, the voltage regulation comes first, with the current limiter circuitry afterwards. that's because the voltage regulation circuit uses large caps on it's output to lower it's impedance. an ideal voltage supply looks like it has zero impedance to the load. a current supply is opposite; it looks like it has infinite impedance to the load. therefore, a good current regulator should not have capacitors on it's output.

    in this circuit, we have to deal with a large, slow transistor due to the amount of power that may be dissipated across it. adding a capacitor will cause current regulation to not be stable if the load resistance is not stable. however, i don't know precisely you need your current to be regulated.

    adding a 65v mov should suppress rapid transients. however, movs also have some capacitance. you would need to read the datasheet for the particular mov you are considering to determine how much it will have.
     
  2. unlv007

    Thread Starter Active Member

    Apr 5, 2008
    44
    0

    i need strict current regulation, that means, the current must strictly be constant up to 60mA. The load resistance in the experimental device gradually increases to 3-4K over 5 sec and then remains constant . This is the state under normal conditions. Under abnormal conditions, the resistance fluctuates pretty badly causing overvoltages. So do you recommend the capacitor, please let me know?
    thanks
     
  3. unlv007

    Thread Starter Active Member

    Apr 5, 2008
    44
    0
    i mean , if i need strictly current source, which one do you recommend
    1) connect the capacitor as RC CIRCUIT
    2)connect MOV and no capacitor. typical specs of a varistor are given below. please advise what criteria i need to consider to select the best MOV, such as capacitance limit.
    example-

    The ZA Varistor Series has a Wide Operating Voltage Range VM(AC)RMS of 4V to 460V, DC Voltage Ratings of 5.5V to 615V.
    Typical parameters for a V220ZA05 Metal oxide varistor:
    • 220VDC nominal (198-253V @ 1mA)
    • 6 joules for a 10/1000 microsecond pulse
    • 360VDC max clamp @ 5 Amp
    • 400Amp max transient surge
    • 180VDC max continuous
    • 140VAC RMS max continuous
    • 0.2W avg power dissipation
    • 90 pF capacitance
    thank you very much.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,201
    1,809
    what is your tolerance for the 60ma supply?

    if you need precisely 60ma all of the time, you are in trouble as soon as you connect anything to the output of the current source. even wire has inductance and parasitic capacitance as well as resistance. so, you must have some tolerance to take such things into account.

    if your experimental load changes resistance as slowly as you say, then using a 10 ohm resistor and a 1uf cap should have a negligible effect; roughly 1 part in 500.
     
  5. unlv007

    Thread Starter Active Member

    Apr 5, 2008
    44
    0
    i can allow tolerance of about +/- 10% of current rating.
    so you would recommend the MOV the RC circuit as well as zener, if i understand correctly.
     
  6. SgtWookie

    Expert

    Jul 17, 2007
    22,201
    1,809
    wait - in order to get a 60ma current through a 4k ohm load, the current source will need to output 240vdc.

    if you limit the voltage to 62v, then the maximum value your load can have to sustain a 60ma current is 1033 ohms - or if the load is 4,000 ohms, current through your load will be about 15.5ma and the limiter will sink the other 44.5ma.

    so, what realistically is the maximum 'good' value of your load?
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,201
    1,809
    i am heading for bed - will follow up later.
     
  8. Caveman

    Senior Member

    Apr 15, 2008
    471
    1
    Just a quick note. A cheap $5 multimeter from Harbor Freight is sufficiently accurate and has a max voltage of well over 1000VDC.
     
  9. SgtWookie

    Expert

    Jul 17, 2007
    22,201
    1,809
    caveman,
    actually, their 'cheapie' meters' highest dc range is 1000v. i have two of them sitting right here. that would still be more than enough for the op's purpose.

    but maybe instead of a clamping circuit, they need an autoranging circuit to keep from supplying too much voltage to the system they're using to plot the voltage data. using a 'cheapie' meter would mean that they would have to manually take datapoints over time, which would be very difficult to do accurately.

    even just a simple voltage divider may suit their purpose; for example, ten 1-meg resistors connected in series across the load. if the output of the current source maximized to 600v, such a divider network would only draw 0.06ma. taking readings across the resistor connected to the return would give them a 1/10 scaled voltage, or limited to 60v.
     
  10. sridhar.k

    New Member

    Apr 16, 2008
    1
    0
    I LIKE TO GET THE STARTER CIRCUIT WHICH IS USED IN OVER VOLTAGE AND CURRENT PROTECTION, AND ALSO ABOUT THE AUTO STARTING OF GENERATOR WHEN EB FAILS, MY MAIL ID IS krissri_85@yahoo.com
     
  11. unlv007

    Thread Starter Active Member

    Apr 5, 2008
    44
    0
    I was asked the maximum value of load under normal condition. Actually it is 1K but that is an experimental device and i am more worried when it goes bad and throws up a high resistance. Thank you very much for your time and explanations. They are definitely helping me to protect my laboratory instruments from damage apart from giving me some hands on experience in circuit design.
     
Loading...