[TCML] For the SISG experienced out there

[bartb]

Hi James,

Sounds like your doing all the right things. When I run my pig, I have a 
ridiculously elaborate safety system involved. I use a PLC which uses 
signals from a deadman switch, motor, enable, alarm, etc.. all of which 
culminate to main contactor control (some if it is just because I can 
and it's fun). I'll probably end up doing something similar with the 
SISG. You've given me some ideas. I like the fan temp control. Easy 
enough to implement. If I do similar, I'll likely keep the fan on 
continuous and implement an indication LED for "ouch, too hot" at which 
point I'll shut it down.

I don't have the means to go over 80nF at the moment. I'll probably go 
with 60nF to start with. Thanks for mentioning that the sisg setup seems 
to like high C. I'll keep that in mind.

The IGBT's I'm using are not the original robust components Terry Fritz 
found. I'm using the ones mentioned a few weeks ago (decent specs but 
smaller die area which means heat is a concern). However, I populated 6 
boards at about $100, so if they fail, it's not like I'll commit 
suicide. I'll just chalk it up to experience and continue on. If they do 
work fine, well, we then have a low priced IGBT readily available.

Thanks for an informative reply,
Bart



James wrote:
> Hi Bart,
>    The SISG coil seems to like a "large" tank cap. When run on the 3
> MOT's, I can set the tank cap up for 200nF. Running on the pole xmfr, I
> configure the cap for 133nF @ 30KV. I’m gonna buy a bunch of 942C's and
> add another 100nF to it. I watch the firing with an O'scope, and then
> tune the primary accordingly. With the SISG coil, the proper tune point
> is VERY easy to see. The firing voltage is approximately 26KV peak to
> peak. Of course the 1st fire pulse is 14.4KV(16 modules), the ring down
> pulses are of lesser voltage. I have the SISG modules in a plexi duct
> with a fan. They run hot as hell. I am VERY careful to use the fan as a
> warning. If I hear the fan, I don't touch. I have an IR thermometer to
> check the transistors. The 1M resistors across the sidacs discharge the
> tank cap after 60 seconds or so. I monitored the high voltage for the
> first few runs. I also use a check list to be sure the SISG coil won't
> pop me.
>     I have a large VTTC powered by a 7.2KV pole xmfr and a 3CX3000F1
> tube. The secondary is 6.25" X 32" wound with #28 wire. The pole xmfr is
> FW rectified (same rectifier as the SISG) so the VTTC runs on 120Hz
> full-wave. I consider the Toob coil much safer, even though I have to be
> careful switching the high voltage off. The switch is close to the
> "action". It makes noise and corona at almost any high voltage setting.
> The big VTTC has a spark length of 36-40 inches, while drawing 60A. I
> can crank it a bit higher, but I get external flashover on the tube. 1st
> time it happened, I almost wet myself. Sounded like a Cherry bomb.
> 	I like playing with the SISG coil. It's quite a bit of fun to
> observe the sparks, without that dazzling light from the SG. I have been
> setting the variac on the SISG coil to 220-230V. I may be able to add 1
> or 2 more modules and still have necessary voltage to fire. 18 modules
> will require 16.2KV to fire. We'll see. Later, James 
>   
>