[TCML] Dream come true for Singing Tesla coil project

[David Petzer <davepetzer@xxxxxxxxx>]

My dream came true yesterday. I received a large budget for my singing Tesla 
coil project. I am six months into my project to build a solid state driver for 
an already built  RSG Tesla coil and make it sing. Pics here

http://www.flickr.com/photos/39813220@N08/sets/72157624539910485/

When I realized just how much this would cost I thought it would never happen, 
but the gods smiled on me somehow.
I have built a small (10A 100V) H bridge with variable dead time modulation to 
drive a car ignition coil, and that works and sings great with a 2inch arch.

I used it to power the large Tesla coil. Obviously 10A (this requires about 20V 
DC bus) into the large Tesla coil produces no sparks at all, not surprisingly, 
but it does produce high voltage that can be picked up with a non connected 
oscilloscope probe acting like an antenna! 


Now the electronics need to be scaled up a little bit. Just a word of thanks to 
all on the list, it has been a wealth of information and entertainment, Xrays to 
LCR meters it has it all!

Question 1
So I have tentatively picked the parts I want the question to the list is can 
anyone see anything wrong with these parts? Do they know of something better for 
large singing Tesla coils?

FF900R12IE4 IGBT half bridge leg (two will make up a full bridge)
http://www.infineon.com/dgdl/DS_FF900R12IE4_2_4.PDF?folderId=db3a304412b407950112b4095b0601e3&fileId=db3a30431f848401011febc3dd603eec


IXDD430 MOSFET driver (As seen on easternvoltage.com)
http://ixdev.ixys.com/DataSheet/99045.pdf

Control signals generated with a MCU (Probably a PIC18F4520 or some such) that 
will also handle self tuning and protection (Over current and over temperature).

My main concern is switching times of the IGBTs I estimate that at 60kHz 
operating frequency 15% of my switching cycle is taken up with delays, turn on 
delay; rise time; turn off delay; fall time the total is 1.23uS. Adding some 
minimum dead time into this pushes this percentage up still further. These 
things have to be lived with I guess but if anyone has any suggestions I would 
love to hear them.

Question 2.
The DC bus voltage could be made by rectifying 3phase AC, this voltage will have 
some ripple to it however even with filter capacitors, at 300Hz (We run on 50Hz 
here in Australia). 


Question is does anyone know if this frequency is heard on large singing Tesla 
coils or is the DC voltage produced some other way? If this is a known problem 
then my next move would be to build a buck converter to regulate the dc voltage 
removing the ripple. This is a massive project all in itself. Or build a large 
filter for the mains and then use very large filter capacitors. Making the 
filter capacitors very large causes large pulse currents which the mains 
protection will not accept, and the capacitors may not like.

Thanks again,

David



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