1. Limited time only! Sign up for a free 30min personal tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

I Humidity and discharge

  1. Apr 18, 2017 #1
    It is known that increased humidity helps discharge a charged body. I was wondering what is the molecular mechanism of that process. It is not clear to me why increased number of well separated uncharged water molecules in the air will help the process of discharge. Does the polarity of water molecules help? How? Does self ionization play any role, although the chance of two water molecules interacting with each other in air is so low? I am having difficulty with this question specially because liquid pure water is not a good conductor. Am I correct in assuming that increased humidity cause higher discharge rate? Thank you for your answer.
  2. jcsd
  3. Apr 18, 2017 #2


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    The main reason is that humid air has a much lower resistance than dry air. This allows the charge to leak away at a much lower breakdown voltage.
  4. Apr 18, 2017 #3


    User Avatar
    Gold Member

    An hour ago, I would have believed you. But my googling indicates otherwise:

    Measurements of the Electrical Conductivities of Air over Hot Water
    1 March 1988
    Page 5
    Contrary to the predictions by Carlon, however, no increases in the conduction currents through the air in the system were ever observed resulting from increases in the humidity of that air. Our findings are that the electrical conducting ability of the air within our measuring apparatus is invariably decreased by the introduction of large quantities of water vapor.
  5. Apr 19, 2017 #4


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    that goes against everything else

    the higher conductivity of humid air is why you DONT get a problem with electrostatic discharge during the summer that you do on dry winter days
    The higher conductivity also reduces coronal discharge problems associated with very high electric and electrostatic fields

    you will find dozens of references on the net to back that up. And surely you must have noticed a difference between static discharges in winter vs summer ??

  6. Apr 19, 2017 #5


    User Avatar
    Gold Member

    And that's why I spent several hours researching this yesterday, as I had a hard time believing it myself.
    My only answer to this, is that there is another mechanism that causes this effect.
    This is probably caused by yet another mechanism.
    I couldn't find any. But I wasn't looking for "coronal discharges" specifically.

    But I did search around the forum to see if a similar question had been asked, and I found:

    How does humidity affect conductivity of air?

    Where I did not find an answer.
    But berkeman asked; "Do you find some graphs when you Google the subject?"
    Which prompted me to look for graphs, and I could only one, on the document I referenced:

    The graph requires a bit of head scratching, but it can plainly be seen that as humidity goes up, conductivity goes down.

    My next quest is: "How on Earth does one measure pico(?)amps?

    Yes, but see my answer to [2] above.

    Bullet point [1-5] numbering is mine, as this is complicated.
    My head almost exploded trying to figure this out.

    ps. As I've pointed out before, I engage in threads, either to learn or to teach. This revelation, if true, really boogers up a hypothesis of mine from a few years ago regarding Is fire a plasma?
    So other questions are:
    "How does minor ionization of air relate to plasma conductivity?"
    "Should we introduce how smoke detectors work?"
    "Does air containing hard-boiled water vapor contain more or less ions than evaporated hot water vapor?"
  7. Apr 19, 2017 #6


    User Avatar
    Gold Member

    Ok, I googled that specifically, and found:


    The effect of humidity on a corona discharge in air is investigated and
    found to produce a decrease in discharge current of about 20% for a change in
    humidity from dry to saturated conditions at atmospheric pressure and room


    Being both vertically and horizontally dyslexic, and somewhat slow in my old age, I have to break this down, bit by bit, to see what they are saying;

    "decrease in discharge current" = current goes down
    "for a change in humidity from dry to saturated conditions" = when humidity goes up

    ∴ current goes down when humidity goes up
    ∴ resistance goes up when humidity goes up
    ∴ conductivity goes up when humidity goes down
    ∴ Lower humidity implies higher conductivity
    ∴ humid air has a lower conductivity
    ∴ humid air has a much higher resistance than dry air

    Which is contrary to the original assertion:
    "humid air has a much lower resistance than dry air"

    ps. Fun thread. Thanks @greensky !
  8. Apr 19, 2017 #7


    User Avatar
    Science Advisor
    Gold Member
    2017 Award

    that's right because with higher humidity, the leakage is higher and therefore the voltages cannot build up to the higher levels to produce the coronal discharge.
    This is the same for any electrostatic discharge. Higher humidity means that the voltage cannot build up high enough to produce that high voltage zap when you reach out to metal door knob, light switch etc as the voltage is leaking away at the much lower level

    Ahhhhhh have just discovered something important to revise my thinking. The paragraph I just wrote above is correct but not for the reason I was thinking. I have been looking at it the wrong way ..... thankyou for making me delve deeper :smile:


    and a couple of other places have lead me to realise that the humidity is not causing the leakage through the air, RATHER it is the buildup of moisture from that humidity on the insulating materials ( dielectric materials) ... carpets, clothes, body and other surfaces that is allowing a lower resistance path for the leaking away of the electrostatic before it can build up to the voltage levels that can zap across the gap between fingertip and metal plate etc, as the voltage becomes high enough to cause dielectric breakdown of the air gap

    A " lightbulb/eureka " moment :smile:

  9. Apr 19, 2017 #8


    User Avatar
    Gold Member

    One of the most delightful things I read yesterday, had nothing to do with this problem, but was more a commentary on how wonderful science was/is:

    The story of atmospheric electricity is a detective story in the realm
    of nature, arresting in the wealth of material already in hand, and allur-
    ing through the promise of rich finds just ahead. This story is told in the
    accurate language of the scientist in the foregoing work by on of the
    chief investigators in this important branch of physics.

    --- W. J. HUMPHREYS​

    ref: A book review from 1929.
    2017-1929 = 88 years

    Book: The Electrical Conductivity of the Atmosphere and Its Causes by Victor F. Hess, 1928

    Victor F. Hess
  10. Apr 19, 2017 #9


    User Avatar
    Gold Member

    I am having a hard time figuring out his circuit.

    For example
    1. With a short across the corona discharge tube, signifying a high ionization, or current flow across the gap, there should be NO current I1.
    The plotter should record then No current. Vy is also = V
    2. With infinite resistance across the corona discharge tube gap, the current I1 should be maximum.

    Is the guy doing it correctly to measure the corona current?
  11. Apr 20, 2017 #10


    User Avatar
    Gold Member

    Be careful not to revise your thinking "too" much. In my quest to publish first, I missed the following:

    page 10

    A study of corona loss from a 400 kv power transmission line by Bailey[3]
    had shown an increase of discharge current with increasing water vapour pressure,
    while the effect is seen here to be a decrease. However, Bailey's work was
    carried out on conductors of about 50 mm diameter
    at much higher voltages than
    the present investigation, which might explain this apparent contradiction.
    Also, since Bailey's test line was in the open air, the effect of humidity may
    have been masked by the presence of water droplets on the conductor; the
    influence of weather generally on corona loss is discussed by Boulet and

    The present observation that the critical voltage is lowered by the
    presense of water vapour agrees with a similar observation by Waidmann,
    working with a pointed electrode; and explanation of the effect has been offered
    by Loeb[14].
    [3] B.M.Bailey I.E.E.E. trans. Vol. PAS-86, 1141 (1967)
    [4] L. Boulet, B. J. Jakubczyk Trans. Eng. Inst. Canada, Paper EIC-64-ELEC 4, July 1964
    [14] L. B. Loeb Ref.1. p.225
    [1] L. B. Loeb, Electrical coronas. University of California Press, Berkeley and Los Angeles (1965)

    "For the bulk of this investigation a wire of 13 µm diameter was used"
    From figure 3, it appears the applied voltage ranged from 1.5 kV to 3.0 kV

    So comparing apples to oranges, 400 kV vs 2 kV, and 55 mm vs 13 µm wire, it would appear that you get opposite results.

    Me too. :blushing:
  12. May 11, 2017 #11
    Thank you all for taking interest and time to think about the problem I had posted. It seemed very simple and almost so silly that I was very hesitant to post it. That is why I am so grateful that you all took the question seriously. I read all the comments and the references to understand more before posting anything further. That is the reason for my long silence, but I enjoyed every minute of reading your postings. This what I think now, which again probably is incorrect or at least partially incorrect.

    1. One of the question raised (and I myself was not sure as wrote in my question) is if increased humidity cause higher discharge rate. So, I did some experiment by charging two conducting balls with a 6000V power supply on a rainy day. I barely could charge the balls as they would not repel much. Whatever the low repulsion they gained, was lost quite quickly with time. However, later in the day, when it was sunny, I could easily charge the balls nicely and repulsion held for much longer period of time. In both cases I opened the windows and used a fan to exchange the inside air with the outside air.

    2. Since I could not charge the balls with the Power supply, I used a Van de Graff generator to charge the balls. I noticed (when raining) that the coronal discharge (the spark) was not taking place unless I take the ground probe much closer to the dome than normal. However, I was getting normal spark when the day was drier. In other words, the maximum spark length was much longer with drier air. This indicated the voltage was not high enough with higher humidity. to produce long spark length. This is in line with some of the reports you mentioned.

    3. I now feel that coronal discharge takes place through the air, but the loss of charge I am thinking about is not through the air.

    4. I feel (after reading others) that water condenses everywhere, whether charged or not. This means water molecules attaches on the insulating rod or string that keeps the charged balls away from the ground. This condensation depends solely on the relative humidity and not on the absolute humidity.

    5. Although invisible, this layer of water provides a conducting path for the charge on the balls to go to ground.

    6. This explains many things including why coronal discharge was less in moist air. However raises a question if condensed water has enough conductivity or nor not. There are two points I would like to mention - 1. self ionization of water will provide some conductivity and 2. dissolved carbon dioxide from air will also increase the conductivity.

    7. So the property of water involved here is that it can stay liquid at room temperature (which other air molecules cannot) and it can condense to the insulating rod.

    8. I would like to know the mechanism of the condensation itself. How can I stop such condensation? Antistatic dryer sheet, Water repellent RainX ( although it may just
    increase the size of the droplets)?

    I will appreciate any thought or suggestion of experiment (for or against) you might have . Thank you so much.
  13. May 11, 2017 #12
    Right now this makes most sense to me. I need to learn more about this condensation process and the conductivity of water. Interested also to how to stop such condensation. Thanks.
  14. May 12, 2017 #13


    User Avatar
    Science Advisor

    Shine a Heat Lamp on the ball supports.
  15. May 12, 2017 #14


    User Avatar
    Science Advisor
    Gold Member

    Re : Original question in #1 .

    Couldn't this be due to the water droplets in the air being good charge carriers ? Combined with a convection / repulsion mechanism the droplets just transport charge away from the source ?
    Last edited: May 12, 2017
  16. May 12, 2017 #15
    I like the idea. I have to be careful as there will be mass movement of air. I was thinking if I coat the insulating rod with hydrophobic molecules, will that stop water molecules from sticking to the rod or they will the water molecules still stick but will form larger drops reducing the area of contact with the rod. I feel I do not clearly understand the condensation process at he molecular level. Thanks for the idea.
  17. May 12, 2017 #16
    Initially I was thinking that the water molecules will be attracted to the charged body, come in contact with the ball, pick up some charge and then fly away taking some charge with it. However, I could not understand why the same will not be done by the nitrogen and oxygen molecules which are more abundant in air. Also I feel (I have no data or reference though) the loss of charge depends on the relative humidity and not on absolute number of water molecules. Your suggestion is slightly different though as you are suggesting droplets and not molecules. Other air molecules cannot form droplets. Formation of water droplets also depends on the relative humidity. So, what you suggest is quite possible.

    Sometimes I have found that even after a tinsel touches a hand held Van de Graff generator (Fun Fly Stick), the tinsel does not fly away. It remains stuck to the wand. That baffles me, as that is not what I expect. Is it because, there was no transfer of charge to the tinsel as a layer of dust (non conductor) is stopping the charge transfer to take place? Sticking of a charged balloon to a wall comes to mind ( the balloon does not get repelled). In the winter time, the synthetic clothes stick to our body - they do not get repelled. Such observation makes me think that I do not quite understand the process yet. So, will the water droplets actually get repelled after touching the charged surface? I do not know. If that is true, that may be a way to avoid condensation of water on an object.

    It is also possible that the cohesive force between the charged body and the water is high and will not generate enough repulsive force for the droplet to leave the surface of the body. Again, I do not have any answer.

    I am not sure if I can treat the movement of free electrons in a conductor and the movement of ions (much larger in size, and consequently need more force for drift velocity) in water the same. So, I am not really clear . It seems that more I think about it, I find more questions than answers.

    Thank you.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook

Have something to add?
Draft saved Draft deleted