ScanTesla - Tesla's Colorado Springs Coil

["Tesla list" <tesla@xxxxxxxxxx>]

Original poster: Terry Fritz <teslalist@xxxxxxxxxxxxxxxxxxxxxxx>

Hi All,

I have incorporated Antonio's new TRSSTC calculation engine into 
ScanTesla-7.40:

http://drsstc.com/~terrell/modeling/ScanTesla-TRSSTC-740.ZIP

I found with MandK, that the K13 coupling was 0.0292 and the K23 coupling 
was 0.0532.  The new coupling made a significant difference and it appears 
that K13 should be a negative coupling.  The negative coupling is not 
significant since it depends on how the model defines current flow.  It 
appears that Tesla should have wound all three coils in the same way and 
direction.

At first, the model did not like to tune well.  It liked an enormous output 
capacitance of about 350pF!  But "I think" Tesla's third coil may have been 
connected to that 140+ foot tower with the ball on top.  I ran that through 
E-Tesla and came up with 331.75pF!  So that is what I used:

http://hot-streamer.com/temp/TeslaCSthridCapacitance.gif

I got most of the number from Dave's data that Richard Hull figured out:

http://hot-streamer.com/temp/CScoilInfo.txt

And MandK and E-Tesla6 Modeling.

http://hot-streamer.com/TeslaCoils/Programs/Programs.htm

The numbers all seem to fairly well agree with each other.  The R1, R2 and 
R3 losses are questionable, but I went light on them so the coil would have 
the best output.  Csecondary is questionable too since that coil was a 
giant in a small close building.  I did not change it since I assumed the 
wood was really dry ;-)

The transformer was rewound for 25kV so we will assume the firing voltage 
was 1.414X or ~35.3kV

At Cprimary =138nF, the coil seems to tune for maximum voltage of 
502kV.  The input power is 50kW  The primary frequency is 56.2kHz.  The 
data is given at the end here.

The output power seems low still at 7.6kW.  For a normal 50% efficient 
coil, that would be a streamer length of 17.5 feet.  To get 32 feet, he 
"should" need a streamer power of 25.5kW

Using mode 2, the program can graph many things vs. Cprimary tuning:

http://hot-streamer.com/temp/power.gif

The 50kW power point is in the center where all the lines meet and the 
parameters are normalized.  If he raised the primary voltage to where the 
input greater and retuned,  He would gain more streamer power.  He would 
have had to raise the BPS some too get the power he needed for 32 foot 
streamers.  The power draw would have been enormous, but he did blow up the 
power plant at one point too ;-))  If he did this, he would have had to run 
in short bursts...

The system seems to "tune best" at Cp = ~85nF.  But it still does not seem, 
even with high BPS, that he could have gotten near a streamer power of 25kW.

The Load Energy Rise Time is higher at lower Cp values.  He coil seems to 
have been in the ~250uS range which is not too bad for a coil that has a 
cycle time of 17uS or about 14 cycles  However, about 50% of the power is 
delivered within 25uS as shown:

http://hot-streamer.com/temp/Voltage.gif

The top voltage has a powerful start and a high peak.  This provides a very 
short powerful burst which is very good for LERT and high BPS 
operation.  Most of the burst is "over" in 100uS!!  It is not know how much 
that would help streamer length.  It may be dramatic!!

We don't know what effect the very short LERT had on streamer length at 
this time.  Like all coil's, the streamer length is fairly directly 
proportional to firing voltage.  If Tesla used a higher voltage and higher 
power than we think (3.4X power 1.8X voltage), he would have had the bang 
energy to get to 32 feet.  It was probably a combination of using high 
voltage and the very short LERT that got the coil that far.  In the second 
file below, I ran a case where the firing voltage was 65kV with and 
efficient tuning and higher BPS.  The coil has 25kW of streamer power in 
this case which could easily hit 32 feet.  The input power is very high at 
155kW!!  Tesla could probably easily raise the primary voltage and BPS rate 
until something blew up (the power plant?).  A short burst at 155kW would 
have been fairly easy.  He also would have had a fairly wide tuning range 
in this case of "throw a lot of power at it"...  He probably could have run 
60nF to 200nF of Cp and still pushed 32 foot arcs.  A higher Cprimary would 
have driven the LERT down also which may have certainly given better 
performance overall...

Take all this with a big grain of salt ;-))

Cheers,

	Terry


Coil Output Data
----------------------
ScanTesla  V-TRSSTC-7.40  June 23, 2005  Terry Fritz
........
Cprimary = 1.380000e-007
Lprimary = 5.800000e-005
Rprimary = 3.000000e+000
K12 = 0.550000   K23 = 0.053200   K13 = -0.029200
Csecondary = 1.500000e-009
Lsecondary = 9.622000e-003
Rsecondary = 6.100000e+000
Ltertiary = 1.591100e-002
Ctertiary = 3.317500e-010
Rtertiary = 9.400000e+000
Cload = 3.200000e-011
Rload = 2.200000e+005
BPS = 576.000000
Dwell Time = 3.000000e-003

Ilprimay Maximum = -1683.990371
ICprimary RMS = 116.958365
VCprimary Maximum = 35255.285148
VCsecondary Maximum = -163863.037412
VCtertiary Maximum = 501655.667015
Coil Power = 49679.047138   Primary Bang Energy = 86.248346
Load Power = 7616.217639   Load Bang Energy = 13.222600
Primary F0 = 56255.707590
Load Energy Rise Time (Sec) = 2.369537e-004
Models Tested = 1
--------------------

Second case with higher input voltage and higher BPS
-------------------------
Cprimary = 8.500000e-008
Lprimary = 5.800000e-005
Rprimary = 3.000000e+000
K12 = 0.550000   K23 = 0.053200   K13 = -0.029200
Csecondary = 1.500000e-009
Lsecondary = 9.622000e-003
Rsecondary = 6.100000e+000
Ltertiary = 1.591100e-002
Ctertiary = 3.317500e-010
Rtertiary = 9.400000e+000
Cload = 3.200000e-011
Rload = 2.200000e+005
BPS = 863.000000
Dwell Time = 3.000000e-003

Ilprimay Maximum = -2559.496625
ICprimary RMS = 205.505341
VCprimary Maximum = 64816.393975
VCsecondary Maximum = -241876.001888
VCtertiary Maximum = 654553.421935
Coil Power = 154962.437500   Primary Bang Energy = 179.562500
Load Power = 24987.984775   Load Bang Energy = 28.954791
Primary F0 = 71679.781368
Load Energy Rise Time (Sec) = 2.689740e-004
Models Tested = 1
-----------------