RE: [TCML] gaps

["Jim Mora" <wavetuner@xxxxxxxxx>]

Hi Bert,

Please define big and power. My 15/120 used two 1" brass dowels 2" long each
spaced about .5" if I recall (end to end). They were threaded and therefore
easily adjustable. They would run 5-10 minutes and not get hot and cleaning
off the Zink oxide we have talked about was occasionally needed for best
performance. The faces were polished. And the air blast was very focused on
the gap and velocity enhanced by the fire hose like tapered nozzle which had
2.5" input to .5" out tapered over 12" and set very close to the gap. 

I have often wondered how scaling this up would work. I have a 240v air
curtain dual blower...I wonder what would happen at 5KVA with BIG dowels.

Jim Mora
 

-----Original Message-----
From: tesla-bounces@xxxxxxxxxx [mailto:tesla-bounces@xxxxxxxxxx] On Behalf
Of Bert Hickman
Sent: Monday, July 28, 2008 3:24 PM
To: Tesla Coil Mailing List
Subject: Re: [TCML] gaps

Hi Bart,

And, with gaps and NST's, YMMV...  :^)

bartb wrote:
> Hi Bert,
> 
> Excellent assessment as I always expect to see in any of your postings!
> 
> Your first two paragraphs sort of stuck out to me and didn't quite match 
> my experience (the note about the most efficient static gap appearing to 
> be a single gap). For that particular issue, I disagree. If within a 
> single or very quick time frame, then I would agree. But there are 
> various types of coilers (and by various I mean how they run their 
> coils). For me, I like to run "long runs" and for others, 20 seconds is 
> a long run. For "those" extremely short runs, a single gap would be 
> efficient, but for those of us that run the 5 minute or longer runs, a 
> single gap will never do well except maybe at low power.

 From a practical standpoint, I agree with you. I've also had very good 
results using multiple static gaps. As you suggest, the challenge with a 
single gap is cooling. Simply using high velocity air flow is not 
sufficient, particularly at higher power levels. A successful high power 
single gap (be it sucker type or air blast) requires using comparatively 
massive electrodes (i.e., low thermal resistance and high thermal 
capacity) combined with sufficient overall cooling. The electrodes need 
to be comparatively massive, with little distance separating the active 
region and the main body. The key is to prevent forming incandescent 
regions on the electrodes since, once this occurs, the gap then loses 
its ability to efficiently quench, and coil output plummets.

Static gaps often use thin or pointed electrodes. Since either have 
higher thermal resistance, overheating may occur within seconds. The 
electrode tips simply can't get rid of heat quickly enough, even with 
heroic levels of air flow. To achieve low thermal resistance, electrodes 
must be comparatively large and made of copper or copper tungsten alloy 
so that heat is rapidly extracted from the active region. Cooling air 
then removes heat from the larger electrode assembly as well as removing 
hot gases from the gap itself.

> 
> The gap has to find a thermal equilibrium. That's when heat 
> dissipation/time becomes important for an efficient static gap. 

I agree. It ultimately comes down to power level, thermal resistance, 
thermal capacitance, and sufficient air cooling. You still reach 
equilibrium, but at a lower electrode operating temperature, with better 
quenching.

> A single 
> gap would only do well in this situation at low power. 

It ultimately depends on the robustness of the gap design. When properly 
designed, a high power single static gap can work very well, even at 
high power and high rep rates.

BTW, there have been

> Higher powered 
> NST coils need quick thermal stability. For myself, I've found largish 
> copper tubing to be good for a static gap. 

I've seen the same thing. And, I suspect that most coilers would see 
better performance when using a typical multiple gap versus a typical 
single static gap. But I still contend that a properly designed single 
gap can ultimately outperform a multiple gap by a measurable amount.

> It's quick to heat but quick 
> to dissipate also. With the right air flow, rather high current can be 
> used efficiently. I have tried the same air flow on 1/2" tubing and the 
> gap was lousy! Then tried the same on solid stock. Great at first, but 
> then suffered immense losses due to heat build up in the material (about 
> 15 seconds into the run, efficiency dropped like a rock). Surface area 
> makes a big difference. But even 1.25" tubing surface area without the 
> right air flow is no better than small 1/2" tubing. The air flow and 
> tube size are very important for an efficient static gap that will rival 
> the best SRSG.

I agree, and bigger is usually better... :^)

> 
> As far as "chaotic firing" of static gap systems, I "sort of" disagree. 
> It is a bit "higher" bps than one might calc, but I have not found it 
> all that high. There is a pattern to the break. 

Interesting. I've found that at some variac settings, the gap "sort of" 
synchronized, showing similar sequences of firing patterns over 2-3 
mains half cycles. Then, at other settings, there seemed to be no 
discernible pattern, while input mains current fluctuated wildly. 
However, I made these measurements on a system that used a mains 
resonant tank cap, and that may also have worsened the degree of 
instability. It wouldn't surprise me to hear that LTR systems exhibit 
more uniform behavior.

> Because of that, it's 
> not so "chaotic" in my measurements (so far). I still need to measure 
> current and bps concentrically to get a handle on it. My most basic 
> thoughts are current is dramatically increased (ferro resonance?), the 
> gap breakdown voltage decreased for a moment (?), or the gap had a 
> transient occur resulting in a hv situation that forced breakdown. 
> Always, this occurrence occurs directly "after" an expected break down. 
> This is the not so chaotic behavior. There is a reason for it. It's just 
> a matter of determining if it is a sudden current or a voltage 
> controlled issue.

I mostly agree. Seemingly chaotic behavior oftentimes has underlying 
structure. We may be seeing something closer to "deterministic chaos".  :^)

> 
> Take care,
> Bart

Best wishes,

Bert

> 
> Bert Hickman wrote:
>> Hi Mike,
>>
>> As with life, many Tesla Coiling questions have no simple black or 
>> white answers... :^)
>>
>> A well-designed static spark gap can provide excellent performance. 
>> The most efficient static gap appears to be a single gap with high 
>> velocity air flow (a "sucker gap" or air blast type). Because of 
>> individual voltage drops within the gaps, multiple static gaps can 
>> have higher overall voltage drops (thus being lossier), but they can 
>> have superior quenching capability at less than heroic air velocity. 
>> Under similar quenching conditions, the single gap may slightly 
>> outperform a multiple gap that has a similar breakdown voltage. Static 
>> gaps are also recommended for first-time coilers for reasons discussed 
>> below.
>>
>> Static gaps can't wring out maximum performance in NST systems. Static 
>> gaps tend to fire chaotically when used with an inductively 
>> current-limited transformer (such as an NST or ballast-limited pig). 
>> This causes multiple bangs during each AC mains half cycle that are of 
>> varying (and suboptimal) size. In order to get maximum spark length 
>> versus input power, best results are obtained using a synchronous 
>> rotary spark gap (SRSG), or a synchronously triggered static gap, 
>> combined with a larger size tank capacitance (see below). A properly 
>> adjusted SRSG consistently forces the gap to fire so that each bang is 
>> of the same, optimal, size - once on every incoming half cycle of the 
>> power mains.
>>
>> In earlier days of coiling, many folks destroyed their NST's when they 
>> converted from static to rotary gaps. These were also the days when 
>> the accepted practice was to design coils where the tank capacitor was 
>> "tuned" to resonate (at mains frequency) with the NST's leakage 
>> inductance, a practice known as "mains resonant charging". Indeed, 
>> many TC design tools defined this as the optimal tank capacitor size 
>> for a given NST voltage and current. For a 15 kV NST, this was about 
>> 0.01 uF for every 30 mA of output current for a 60 Hz supply. 
>> Unfortunately, if the main gap was set too wide, the rotary gap was 
>> improperly adjusted, or the RSG was merely running too slowly, the 
>> voltage cross the NST could rapidly grow to ridiculously high 
>> voltages. This usually resulted in overvoltage failures of NST's, or 
>> for pig-driven systems, tank capacitors.
>>
>> Over the years, it was learned that, by using a larger sized capacitor 
>> (called Larger Than Resonant or LTR), mains-resonant overvolting could 
>> be avoided. Adding a properly adjusted safety gap placed directly 
>> across the NST output terminals protects the NST from overvolting 
>> under other/abnormal circumstances. Terry Fritz's "Terry Filter" adds 
>> an additional layer of protection, especially against high speed 
>> transients that can cause inter-turn corona damage within the 
>> outermost turns of the NST windings. It even protects against an 
>> improper safety gap setting.
>>
>> The bottom line line:
>> A system using a synchronous rotary gap, an LTR tank cap, a properly 
>> set safety gap, and a Terry Filter is every bit as reliable as a 
>> system using a static gap. The SRSG system also provides higher 
>> performance than a similar static gap system.
>>
>> Bert
> _______________________________________________
> Tesla mailing list
> Tesla@xxxxxxxxxxxxxx
> http://www.pupman.com/mailman/listinfo/tesla
> 


-- 
***************************************************
We specialize in UNIQUE items! Coins shrunk by huge
magnetic fields, Lichtenberg Figures (our "Captured
Lightning") and out of print technical Books. Visit
Stoneridge Engineering at http://www.teslamania.com
***************************************************
_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla

_______________________________________________
Tesla mailing list
Tesla@xxxxxxxxxxxxxx
http://www.pupman.com/mailman/listinfo/tesla