Corona discharge and polarity

[mbrooke]

This is from an EE book and I am having trouble understanding what I am seeing. It seems as though DC corona varies in magnitude by polarity? Or is something else at work?

 

[GoldDigger]

Discharge between two identical electrodes will be identical for both polarities of DC, while AC may be different because the current passes though zero repeatedly.
If the two electrodes are not identical there can be asymmetries. Air will be ionized where the field gradient is highest, such as thenear sharp points. And the results can be different depending on whether those ions are positive or negative.

 

[mbrooke]

Discharge between two identical electrodes will be identical for both polarities of DC, while AC may be different because the current passes though zero repeatedly.
If the two electrodes are not identical there can be asymmetries. Air will be ionized where the field gradient is highest, such as thenear sharp points. And the results can be different depending on whether those ions are positive or negative.

But does air emit light at the positive or negative electrodes? I ask, because in cases such as fluorescent tubes operated on DC, mercury will migrate toward one end.

 

[junkhound]

I have a hard time understanding where the electrodes are in those photos - is it 2 adjacent plates a cm apart (or narrower at lower voltage and just the field strength given v.s actual voltage?) Or, is it wire above a ground plane not in the photos, and the ground plane is the anode -- some of those old text are confusing, such as straight polarity and reverse polarity in welding. Straight polarity is with the workpiece + and the stick negative. Kinda opposite what one would think right off being as the workpiece is often thought of as 'ground'.

I suspect that the top electrode in the photo orientation are +, as they show streamers, and the the bottom electrode may be a plate or chassis. Assuming that, read below.


With 2 electrodes, as electrons are pulled to the anode (+), a small cloud of positively charged ions is ‘left behind’. This creates a highly localized electric field at the cathode (called space charge region) that prevents all electrons from reaching the anode.

At the cathode (-), there is a dense cloud of positive ions, and electrons are rapidly freed from gas molecules and recombined with the dense ion
cloud. The glow at the cathode is what is commonly seen on devices such as the common neon glow lamp. This is a diffuse area.

At the anode, one can visualize the flow of electrons flowing toward the anode as streams coming together into a larger creek and then a river. This produces the faint glows as seen in the first picture, and the tendrils are called streamers.
There are no streamers at the cathode as there is the diffuse glow, plus the positive ions are much heavier and moving much slower.

As the voltage increases, there are obviously more electrons. If the camera shutter were stopped down, the streamers could still be seen vs. the big white spots on the photos. At high enough voltage, the full arc occurs, which is a plasma bridging the electrodes as all the molecules on at least one streamer path are fully ionized between the anode and cathode.


If both electrodes are identical, reversing the polarity would basically just flip the pictures upside down, but that is obviously not the case. Hard to describe just what is going on without seeing a 'daylight' picture of the electrode setup.

ac is simply reversing the polarity at the frequency of the ac.

 

[junkhound]

add in for where the light comes from from added query.

To dark with the camera set to capture the brighter streamers a 10s of kV/ cm field strength, so the cathode glow is not seen in the photos.
The light is a function of electrons and ions being torn apart and re-combining - in a cloud at the cathode, bright streamers in the electron flow path at the anode. As the electrons in the; molecules change state, they give off a color consistent with the distance they jump (very colloquial description) - thus air is blue because of the size of the air molecules, neon is yellow orange, red for other gases, neon tube makers take advantage of the different sizes of argon, neon, helium, etc. to produce different color signs.
If one put dc on a neon sign, all the color would be at one end of the tube, the other basically dark because of low current and the streamers would not be bright enough to be seen. Gets more complicated on what is seen when different tests are run at different gas pressure, etc., suffice it to say at low pressures and current (few electrons flowing) the cathode glow is most visible, at high pressures (e.g 15 psi) it is the streamers that are most visible.

The photo below is a common neon glow tube on dc. The LH terminal is + (note cathode ring on the diode). This is at only a few mA so the glow is seen but streamers are invisible. The far left photo is in normal room light, the middle photo is with darker room and auto camera shutter. If one looks VERY closely, you can see small streamers from the tip of the anode.

Hope this helps vs. confuse farther?


PS: fluorescent tube is full plasma arc thru the mercury vapor, ac or dc. Light is from UV from the mercury arc hitting the phosphors, not from streamers or cathode glow as described here.

 

[gar]

160126-1416 EST

junkhound's scope picture was interesting. I had never plotted one before. I was curious as to why the bottom of the plot was not flat. I could approximately duplicate this with a 10 M scope probe, but my curve was a little flatter. By paralleling 1 M with the scope probe I could flatten this.

This upperward slope of the bottom flat is probably a result of the equivalent shunt capacitance of the series diode used for half-wave rectification.

Attached is my waveform.
View attachment 14249


A useful reference for NE-2s is:
http://vacuumtubes.biz/documents/Glow Lamp Specifications.pdf

.

 

[gar]

Here is the plot:



.

 

[gar]

169127-1218 EST

Although the neon bulb waveform is a digression from the original post it is interesting.

My new experimental circuit consists of the following items in series and parallel:
1. 1N4006 diode in parallel with a 10 pfd mica capacitor.
2. A 100k 1/4 W resistor in series with the diode-capacitor parallel circuit.
3. Connected in series to the lower end of the 100k is a NE-2 type neon bulb.
4. The lower end of the NE-2 goes to common (AC neutral).
5. In parallel with the NE-2 is a scope probe. Input impedance 10 M paralled with 12 pfd.
6. 125 V 60 Hz is applied between the top end of the diode and neutral.

This circuit differs from my previous circuit of posts 6 and 7 by added capacitance across the diode, and removal of the 1 M resistor that was paralleling the NE-2.

The zero base lines of the two traces were adjusted to coincide when no voltage was applied. If there is no offset by overdriving the scope channel 2 amplifier, then the scope picture should have a good zero base line.

This picture shows the applied input voltage and the voltage across the NE-2.

Now a curve is displayed with a similar characteristic to that shown by junkhound.




.

 

[rlundsrud]

add in for where the light comes from from added query.

To dark with the camera set to capture the brighter streamers a 10s of kV/ cm field strength, so the cathode glow is not seen in the photos.
The light is a function of electrons and ions being torn apart and re-combining - in a cloud at the cathode, bright streamers in the electron flow path at the anode. As the electrons in the; molecules change state, they give off a color consistent with the distance they jump (very colloquial description) - thus air is blue because of the size of the air molecules, neon is yellow orange, red for other gases, neon tube makers take advantage of the different sizes of argon, neon, helium, etc. to produce different color signs.
If one put dc on a neon sign, all the color would be at one end of the tube, the other basically dark because of low current and the streamers would not be bright enough to be seen. Gets more complicated on what is seen when different tests are run at different gas pressure, etc., suffice it to say at low pressures and current (few electrons flowing) the cathode glow is most visible, at high pressures (e.g 15 psi) it is the streamers that are most visible.

The photo below is a common neon glow tube on dc. The LH terminal is + (note cathode ring on the diode). This is at only a few mA so the glow is seen but streamers are invisible. The far left photo is in normal room light, the middle photo is with darker room and auto camera shutter. If one looks VERY closely, you can see small streamers from the tip of the anode.

Hope this helps vs. confuse farther?
View attachment 14242

PS: fluorescent tube is full plasma arc thru the mercury vapor, ac or dc. Light is from UV from the mercury arc hitting the phosphors, not from streamers or cathode glow as described here.

I thought the light was caused by electrons moving from a higher to a lower orbit in the atom being ionized. As it drops to a lower orbit a photon is emitted. Inversely if a photon hits an atom with enough energy, it moves an electron to a higher orbit or even releases an electron from the valence orbit if it is greater than the work function value for that atom.

 

[GoldDigger]

I thought the light was caused by electrons moving from a higher to a lower orbit in the atom being ionized. As it drops to a lower orbit a photon is emitted. Inversely if a photon hits an atom with enough energy, it moves an electron to a higher orbit or even releases an electron from the valence orbit if it is greater than the work function value for that atom.

The photons are emitted as electrons move to a lower energy state. They are in a higher energy state in the first place because of the energy released in the recombination process and other collisions.

 

[winnie]

I thought the light was caused by electrons moving from a higher to a lower orbit in the atom being ionized. As it drops to a lower orbit a photon is emitted. Inversely if a photon hits an atom with enough energy, it moves an electron to a higher orbit or even releases an electron from the valence orbit if it is greater than the work function value for that atom.

I think that it is fair to approximate things as having two aspects:
1) Electrons (and ions) moving in bulk between the electrodes, delivering energy to the atoms and ions in the gas between the electrodes.
2) Electrons 'jumping' between different energy states near nuclei and giving off (or absorbing) photons.

In this approximation, the effects being described (which electrode glows, polarity, etc) are predominantly aspect 1). Clearly there have to be photons given off by the bulk process as well, since electrons are accelerating and moving between different energy states, but my guess is that the bulk process photons will be low energy (RF and below). So in this approximation aspect 2) tells us the color of the light and aspect 1) tells us where the light is emitted.

-Jon