[Home][2014 Index] Re: [TCML] Dc charging and power arcing [Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCML] Dc charging and power arcing

Hi Jim,

I have been looking at your 24 kV power supply and possible options over the past few days. The 8020's and the high inductance choke limit your options for a DC charging system, especially with the tank cap sizes you're looking at.

In your current 6-pulse bridge rectifier setup, the combination of tube plate dissipation of the 8020's and the full-wave bridge configuration limits the RMS output current of the supply to ~200 mA (even though each tube is only rated for 100 mA). Also, the tube manufacturer specifies a peak current limit of 1.5 Amps. These specs add some additional limitations for designing a high-power DC resonant Tesla Coil system. These are discussed below, along with some suggestions.

Because of the relatively high series resistance of your 8020's (about 1000 ohms/tube), I'd recommend adding a DC storage capacitor on the tube side of the choke that is at least 10X the capacity of your tank cap. You'll also need an appropriate HV blocking (dequeuing) diode in series with the choke. As you have noted, the charging frequency (Fo) of your tank cap and choke lead to relatively low break rates for full resonant charging of 59, 84, or 118 BPS for 0.2uF, 0.1uF, and 0.05uF tank caps respectively. However, that's not the _only_ problem. At full output voltage and maximum break rate, the RMS charging current for your three tank cap sizes will be:

RMS Current: Tank Cap:
0.555 A      (0.2 uF)
0.408 A      (0.1 uF)
0.297 A      (0.05 uF)

So, even with the smallest tank cap, the HV supply RMS charging current is still too high for sustained safe operation of your rectifiers - you risk overheating the anodes, possibly damaging/destroying them. Operating at lower DC output voltages will reduce the RMS current to below 200 mA, but this also reduces the maximum power delivered to your coil. For example, following are the maximum DC output voltages and output powers that will keep tube RMS current at or below 200 mA. Unfortunately, the output power for all three cases is probably much lower than what you;re looking for. :

16.0 kV, 2.8 kW   (0.05uF)
11.7 kV  2.0 kW   (0.1uF)
 8.7 kV  1.4 kW   (0.2uF)

Some possible options that still use your Raytheon supply:

1. Use a smaller tank cap and a higher break rate. For example, using a 0.022uF tank cap will allow you to run up to 175 BPS in a DC charging configuration while still limiting RMS charging current to 200 mA. This will deliver about 4500 watts of power to your coil when running at 24 kVDC at 175 BPS. Your tank cap must be rated for 90 - 100 kVDC and you'll need to run with wider turn-turn spacing on your primary to avoid flashovers. Unfortunately, low tank circuit inductance will also cause tank currents to be significantly higher, so spark gap electrode erosion and high tank cap RMS current may become problem areas. Operating with a 48 kV tank circuit and rotary will also create another set of challenges.

2. Substitute solid state rectifiers for your 8020's with higher RMS and peak current ratings. The 8020's "look" like ideal rectifiers in series with 1000 ohm resistors - they are quite lossy and fragile compared to series stacks of silicon avalanche rectifiers. Although you'll ultimately be limited by joule heating in your 3-phase HV transformer and charging choke, you should be able rely on the thermal lag of the copper windings and push it to 2-3X the nominal current rating for short-moderate run times (i.e. 500-600 mA range). This will allow you to run with larger tank capacitance than with your tube rectifiers. (I'm assuming that the charging choke has sufficient design margin so that it does not begin saturating under these conditions). This may allow you to run at lower voltages and larger tank capacity while still processing decent power levels.

3. Use a more complex gap configuration and DC storage cap across the output of the Raytheon supply to charge the tank cap during one presentation and then discharge it across the tank circuit on the next. This will allow you to run at a maximum of 24 kV in the tank and gap which is still high, but not nearly as high as 48 kV. Steve Young's approach could also be used, but this charge/reverse charge system also results in 2X the tank circuit voltage and associated set of design challenges.

Regarding "Trailing Arcs" (in DC resonant charging systems):
As long as your charging choke does not begin saturating, you shouldn't encounter problems with trailing arcs - even when running at break rates that are significantly higher than 2X the resonant frequency of the charging system. Trailing arcs usually occur when the charging inductor is too small or if it begins to saturate, or the rotary speed is too low. In these cases, the tank cap recharges much too quickly, allowing the separating gap electrodes to reignite and then continue to arc with inductor follow-through current. Similar behavior is seen in "overpowered" pig-powered systems (where the tank cap is too small (versus available recharge current) versus rotary speed.

As your break rate begins to exceed 2X the recharge frequency (Fo), the gap will continue to break properly. However, the charging choke will still be recharging your tank cap when the next gap presentation occurs. Instead of being charged to approximately 2X VDC, the tank cap will be partially recharged to a lower voltage. There are two possible results:

1. If the tank cap is recharged to a voltage sufficient to fire the gap during the NEXT presentation, the result is a reduction in bang size, but the gap still fires at the new (higher) break rate. Although the system can repeatably work under these conditions, the output power no longer scales linearly with break rate. However, the reduction in bang size is partially offset by the increased break rate, so output power to your coil flattens out. At higher break rates, output power eventually begins to decline.

2. For wider gap spacing, or higher rotary speeds, the tank cap may not recharge enough to fire the gap during the next presentation, and the tank cap continues to charge until the next gap presentation. If the cap voltage is now sufficient, the gap will fire. If not, charging continues until the next presentation. Eventually, the cap becomes sufficiently charged to fire the gap. For equally spaced gaps, as rotary speed is increased, the gap will begin firing on every other presentation, then every third presentation, etc. However, real-world variations in electrode spacing will introduce uneven firing. Although coil firing may no longer be smooth, no major problems should occur. Behavior is similar to that of underpowered systems with larger tank caps, underpowered recharging supplies, and excessively high gap presentation speeds. The effective firing rate is significantly lower than the mechanical presentation rate.

Hope this helped and best wishes,

Bert Hickman
Stoneridge Engineering
World's source for "Captured Lightning" Lichtenberg Figure sculptures,
magnetically "shrunken" coins, and scarce/out of print technical books

Jim Mora wrote:
Hello List,

I have been letting the issue of DC charging rattle my brain a few days and
it always comes back to Burnet's comments on Arc trailing. If my Inductor
has more energy then is needed, the Spark cap is likely going to take on the
power in long arcs between electrodes. What experience do you folks have
with this issue? I'm tempted to let go of this and proceed on with pig
power. I do have a virgin 18" square of 5/8" G10 Fr4 that should quench a
dragon though best maybe to let this DC coil simmer on the back burner for
awhile longer. I seem to remember Steve Young made some form of charge than
isolate gap though.

Thanks for your patience; I'm just getting back into coiling again and have
12" PVC form waiting to be made first anyway!

Jim Mora

Tesla mailing list

Tesla mailing list