So I've been experimenting with creating high voltages and have a circuit generating 1kv ac, 20khz, about 40 watts. I hooked it to a 20w fluorescent tube, and it lights, but the arc inside moves in a spiral motion along about half of the tube. I know I'm pumping more power than its rated for, it was a burned out tube from my old kitchen counter lights so I don't care if it will be damaged. Note that it does not heat up much. It doesn't appear brighter than a new tube on a normal ballast. I'm just curious why the arc is moving like this? I have a video on youtube here:

any ideas?

 

Also this effect seems to go away after a few minutes.

 

So I've been experimenting with creating high voltages and have a circuit generating 1kv ac, 20khz, about 40 watts. I hooked it to a 20w fluorescent tube, and it lights, but the arc inside moves in a spiral motion along about half of the tube. I know I'm pumping more power than its rated for, it was a burned out tube from my old kitchen counter lights so I don't care if it will be damaged. Note that it does not heat up much. It doesn't appear brighter than a new tube on a normal ballast. I'm just curious why the arc is moving like this? I have a video on youtube here:

any ideas?

Short (i.e. qualitative) answer: The arc follows the path of least electrical resistance (technically; the path of most readily ionized fill)-- said path being determined by, among other factors, convective flow within the fill gas --- hence the observed phenomena are owed, primarily to thermodynamic effects - but 'recursive' in that, owing to corollary heating and charge distribution, the arc path is both its own 'cause and effect', if you will --- Add to this the fact that, at 20kHz on a linear tube, there will be significant capacitive coupling to nearby conductors and, to a lesser extent, radiation effects...

Also this effect seems to go away after a few minutes.

The arc stabilizes as the fill gas and tube wall approach thermal equilibrium/mean uniformity...

Hope you find this helpful
Best regards
HP

PS: I probably don't need to explain this, however by 'radiation effects' I mean that the plasma (i.e. arc), in small measure, acts as an antenna --- No worries about ionizing radiation!

 

Short (i.e. qualitative) answer: The arc follows the path of least electrical resistance (technically; the path of most readily ionized fill)-- said path being determined by, among other factors, convective flow within the fill gas --- hence the observed phenomena are owed, primarily to thermodynamic effects - but 'recursive' in that, owing to corollary heating and charge distribution, the arc path is both its own 'cause and effect', if you will --- Add to this the fact that, at 20kHz on a linear tube, there will be significant capacitive coupling to nearby conductors and, to a lesser extent, radiation effects...


The arc stabilizes as the fill gas and tube wall approach thermal equilibrium/mean uniformity...

Hope you find this helpful
Best regards
HP

PS: I probably don't need to explain this, however by 'radiation effects' I mean that the plasma (i.e. arc), in small measure, acts as an antenna --- No worries about ionizing radiation!

Thanks that makes sense I've never seen this in normal fixtures but I am dumping about 40 watts into a 20 watt tube which probably makes the arc hotter and therefore creates more movement inside the tube which would cause the arc to move around. It's kind of fun to watch too.

I'm familiar with radio waves but the arc acting as an antenna never occurred to me. That would explain why the tube glows dimly with only one end connected to power.

 

I'm familiar with radio waves but the arc acting as an antenna never occurred to me. That would explain why the tube glows dimly with only one end connected to power.

I have seen the glow flowing in random patterns in cold tubes such as in an unheated garage. HP's explanation probably works for that situation as well.

 

This is just a guess. It looks to me like the moving arc is due to local heating, which makes the gas easier to ionize. I've seen this happen in old, weak tubes. As the gas ionized near the electrode, it heats gas in the local vicinity, which in turn ionized, heating the gas in its local, and ionizing it, which in turn heats more local gas, etc., thus creating a 'heat wave' of ionized gas.

 

So I've been experimenting with creating high voltages and have a circuit generating 1kv ac, 20khz, about 40 watts. I hooked it to a 20w fluorescent tube, and it lights, but the arc inside moves in a spiral motion along about half of the tube. I know I'm pumping more power than its rated for, it was a burned out tube from my old kitchen counter lights so I don't care if it will be damaged. Note that it does not heat up much. It doesn't appear brighter than a new tube on a normal ballast. I'm just curious why the arc is moving like this? I have a video on youtube here:

any ideas?

There's a certain amount of capacitance related to the mercury droplets in the bulbs. They light-up from cold cathode emissions and thereby limited conduction within the bulbs.

 

Yeah? You've got me perplexed. I joined this site last week, and I don't know who you had by the name Patrick Malarkey a year ago?

 

This is an old thread that has not seen the light of day in many years. I suggest that if you think this is important enough, send a private message to the one to which you would reply.