Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Spark generation in capacitors

  1. Oct 26, 2018 #1
    If we directly connect the two plates of an ideal charged parallel-plate capacitor, a spark will generate. What is the exact reason of spark?
     
  2. jcsd
  3. Oct 26, 2018 #2

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    If by "connect", you mean that you let the two plates touch each other, then I have 2 comments:

    1. Shouldn't you create a short by touching them together? If this is the case, why would there be "sparks"? You may get sparks as they both get closer, but not after they touch!

    2. A spark can happen even when they are not touching. That's how you can destroy a capacitor.

    This "spark" is a dielectric breakdown in the capacitor, similar to lightning in air. The exact mechanism is complicated, and one may even argue that this is similar to a vacuum breakdown.

    Zz.
     
  4. Oct 26, 2018 #3
    In an ideal parallel-plate capacitor, the electric field between the plates is always constant no matter in which distance they are. Why the dielectric breakdown occurs when we make them closer while it doesn't occur when they are far?
     
  5. Oct 26, 2018 #4

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    This doesn't make any sense. If your capacitor is connected to a constant voltage source (i.e. batteries), then when you make them closer, the electric field in between the plates increases (E = V/d), since the separation distance "d" gets smaller. So your claim that the electric field is always a constant when the distance between them changes is wrong.

    Zz.
     
  6. Oct 26, 2018 #5
    Please consider the case where the plates are disconnected from the battery. In this case, the electric field is always constant.
     
  7. Oct 26, 2018 #6
    By this, do you mean that short circuiting the plates never cause spark and spark is only due to dielectric breakdown?
    I think that by shorting the plates, the electrons gain high acceleration and so there will be great radiation that we observe as spark. Is this right?
     
  8. Oct 26, 2018 #7

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    When you short circuit the plates, the current goes through the contact point, not through the gap. So what spark?

    You asked me if it is right based on something you contradicted with what I said. So why should I tell you that it is right, when you are going against what I told you?

    Zz.
     
  9. Oct 26, 2018 #8

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    Doesn't make any sense either. If the battery is disconnected, then the amount of charge is fixed on each plate. Now what do you think will happen if the spacing gets closer?

    You seem to be making things up as you go along. You never specified this when you started the thread. And now, you're changing your tune. I don't want to play this game anymore.

    Zz.
     
  10. Oct 29, 2018 #9

    Andrew Mason

    User Avatar
    Science Advisor
    Homework Helper

    Perhaps hokhani is thinking of something other than an ideal parallel plate capacitor - i.e. a real one. In an ideal parallel plate capacitor with a fixed charge distributed evenly across the plates right to the edges, the field is constant. V = Ed, so as the plates approach, V⇒0 (potential energy of charge ⇒ 0) and no sparks. But in a real capacitor surfaces will not be completely smooth (say, ± δ). So that as the plates approach and the separation d → δ, the charges will move around and accumulate at the closest point, increasing the electric field (E) at that point which could cause a spark before they touch.

    AM
     
  11. Oct 29, 2018 #10

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    But the problem here is that in the very first post, there IS a spark! That appears to be the central question, i.e. why is there a spark? If there is no spark, this question does not exist, does it?

    During winter or when the air is very dry, we sometime get a jolt when we are about to touch a door handle or a metallic object. This happens when we get very close to it, but NOT when we actually touched it. In fact, touching something metallic is one way to prevent this from happening again soon. The OP never provided any evidence for the observation of "ideal parallel plate capacitor" having a spark when they are "directly connected". If this actually happened in real life, this is not an "ideal parallel plate capacitor". So this whole premise in this thread is full of contradictions, and the OP seems to making it up as he/she went along.

    It was why I decided to disengage from answering his queries anymore.

    Zz.
     
  12. Oct 29, 2018 #11

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    well ... shorting out a capacitor does produce a spark .... have done it many, many times over the years in the workshop
    But I haven't looked at it on the microscopic level to see when the spark occurs

    A 100V 10uF electrolytic cap will produce a healthy spark and "crack" sound when the terminals are shorted out by a screwdriver
    and will leave burn marks on the cap terminal and on the screwdriver tip.

    Assumption is that there must be a gap for a spark to occur ?
    if so, the gap must be really small in that it is impossible to tell doing a quick experiment whether the spark occurred
    with micrometres between the 2 conductors or at the time of contact ?


    Dave
     
  13. Oct 29, 2018 #12

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award


    Indeed, agreed, at the voltages involved, 5000V and much more, the gap is easily seen as is the spark jumping across those several mm.

    But with a screwdriver etc across a capacitor, the gap isn't obvious and it is easy to understand
    that people possibly mistakenly think it happens at the point of physical contact


    Dave
     
  14. Oct 29, 2018 #13

    Andrew Mason

    User Avatar
    Science Advisor
    Homework Helper

    Bringing a conductor across the capacitor leads is a very different scenario than the OP's scenario of bringing the plates physically close together.

    The charge on a charged capacitor plates creates a potential difference, V, between the plates and, therefore, between the leads. V does not change if you change the distance between the leads. So, if V is high enough and if you then bring the leads close enough together, you are going to create an E field that exceeds the dielectric breakdown strength of the air creating a spark.

    On the other hand, if you move the plates together (in an ideal parallel plate capacitor), the E field is constant so V approaches 0 as the plate separation approaches 0 → hence no spark.

    aM
     
  15. Oct 29, 2018 #14

    ZapperZ

    User Avatar
    Staff Emeritus
    Science Advisor
    Education Advisor

    First, let's go back to what we know:

    1. If there is no potential difference, there will be no spark. So when something is shorted, after the potential difference has dropped, breakdown can't happen.

    2. A screw driver touching a capacitor is not an ideal capacitor.

    This is where careful consideration has to be done on what exactly what happened. Did the spark occurred right before contact? Or did it continue a millisecond second or two after contact? Here's the issue, and I've sketched a particular scenario here:

    screwdriver.jpg
    When you first make contact with the plate with another grounded object, the capacitor since won't be fully discharged immediately. The contact point often has a high resistance, and so the time constant for discharge isn't 0. There is still a momentary potential difference between the plate and the rest of the object during discharge. I illustrate 2 points where the potential difference can easily cause further arcing, especially if the object has other sharp points or corners.

    So the arcing is in between the two due to the large electric field developed over a small distance.

    Zz.
     
  16. Oct 29, 2018 #15

    davenn

    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    but it's unlikely to be instantaneous, as in, the discharge occurs over maybe milliseconds
    as you indicate in what I quoted from you below. So doing that short isn't an instant voltage drop
    it occurs over a brief period of time allowing for that arc generation ??



    great response and it goes towards explaining my own ponderings that I was trying to put into words in my previous posts

    thanks :smile:
     
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted