[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Weird SRSG turn on



Original poster: "Bart Anderson" <tesla111-at-sbcglobal-dot-net> 

Hi Gerry,

Are you running 120 bps? Sounds like the electrodes have moved past and 
away beyond the reach of the gap distance with given firing voltage (a 
timing issue). Curious also about the phasing method your using?

If your measuring a firing signal against the input or charge voltage, do 
you know where the electrodes are when the firing occurs (before, on, or 
after alignment)? We always assume before alignment, but this is also 
dependent on the firing voltage and gap distance at the firing time. It is 
possible to fire on and even after alignment (but after will fall off 
quickly and move to safety's firing). As you attempt to phase later along 
the 1/2 cycle, the electrodes have moved even further away. A slight 
adjustment with the disk itself in the "late" direction should identify if 
this is the problem.

Just a thought. Take care,
Bart B.


Tesla list wrote:

>Original poster: "Gerry Reynolds" <gerryreynolds-at-earthlink-dot-net>
>Hi SRSG experts,
>
>I'm in the process of turning on a SRSG system and I'm getting unexpected 
>results.  Operation is smooth when the phase is set before peak or as late 
>as 0.5ms after peak.  Any later than this and operation becomes rough 
>coinciding with the safeties firing.  So far, I've used 2.0*Cres and 
>2.5*Cres for the Cp (I'm working my way up to 2.8*Cres).  I'm using a 
>15/60 NST, the terry filter, and a 3 terminal safety gap feeding the 
>SRSG.  The safeties are set to just not fire when the variac is set to 
>140VAC with no load on the NST.  I calculate that this is approximately 
>25KV but I should probably measure it.
>Simulations showed that optimum phasing would be about 1.5ms after peak.
>
>With 2.0*Cres and while scoping one side of the safety to ground, I 
>delayed the phasing to just before rough operation occured.  At this point 
>(0.5ms after peak) the safety voltage was measured 11KV and then doubled 
>to account for the other bushing for a total Vfire=22KV.  So far OK
>
>I switched to 2.5*Cres and repeated the phasing adjustment and measurement 
>and Vfire would only get to 18KV and of course performance was worst.  The 
>phasing was still only 0.5ms after peak and any attempt to delay the 
>phasing to allow the Vfire to increase resulted in the safeties firing and 
>rough operation.
>
>I then stopped the SRSG and positioned it so it would not fire.  I then 
>scoped the safety voltage while slowly ramping up the variac output.  I 
>was able to increase the voltage to the safety gap to about 10KV (or 20KV 
>total) and everything was OK.  Any increase above this level resulted in a 
>runaway where the voltage snapped to about 15KV (or 30KV total) and the 
>safety would fire.  I do need to repeat the measurements and document the 
>readings more accurately since the original values are from memory.  I'm 
>wondering if the bi-modal behavior is the result of something saturating 
>and if the single ended measurement should be replaced with a differential 
>measurement.  Right now I can't charge anything larger than 2.0*Cres and 
>get full voltage without the safeties firing.
>
>
>Any insight into this behavior would be appreciated.
>
>Thanks
>Gerry R
>
>