Re: [TCML] mmc cap protection by spark-gaps-any ideas?

[Quarkster <quarkster@xxxxxxx>]

Ray -
   
  Can you elaborate a little on your statement that connecting capacitors in series to gain a higher working voltage rating is "bad practice"?
   
  Many high-reliability, high-performance (low ESR/low ESL), high voltage capacitors are designed by the manufacturer in just this way, with multiple series-connected modules contained in a common case. Perhaps you are thinking of situations where someone does not really have a capacitor that is suitable for a particular application, and tries to meet the voltage requirement by stringing together a bunch of random ("junk-box") capacitors that don't really meet any of the other functional requirements (peak current, RMS current, ESR, ESL, operating temperature, operating life, etc).
   
  Modern MMC designs have become quite sophisticated, taking all of these factors into consideration, and selecting the capacitor configuration to simultaneously meet all the desired requirements.Powerful circuit simulation tools like MicroSim can easily determine the peak current and voltage requirements that an MMC will be subjected to in a given Tesla coil circuit, and can yield MMC designs that are reliable and cost effective.
   
  Regards,
  Herr Zapp
   
  

Ray von Postel <vonpostel@xxxxxxxxxxx> wrote:
  
On Aug 19, 2008, at 10:08 AM, S&JY wrote:

> Ray, all,
>
> Yes, for DC applications, like power supply filter caps in series, the
> resistors certainly do help equalize the voltages across each cap. 
> Thus
> they do add some protection to the caps.

I hope by implication you got my point that connecting capacitors in 
series to achieve a given d.c. working voltage is/was considered a bad 
practice. The use of the resistors was just to compensate for some of 
the makeshift technical work-a-round or lack of funds.
>
>
> But in TC usage, the voltage across the series strings of caps is RF,
> changing polarity on the order of 100,000 times per second. In this
> situation, the RC time constant for a bleeder resistor across an MMC 
> cap is
> so large (seconds)in comparison to the roughly ten microsecond cycle 
> time
> that the resistors will do practically nothing to equalize the voltage
> across the caps.

What has the frequency got to do with any thing? By place a resistor 
across a capacitor you have created the equivalent of a capacitor with 
a low leakage resistance. If you apply an a.c. voltage to a parallel 
circuit consisting of a capacitor and a resistor, the instantaneous 
voltage is the same across both elements.
>
>
> So in Tesla Coil usage, the bleeder resistors are there as just ONE 
> of the
> protections against dangerous shocks, NOT to protect the caps. But 
> as you
> say, that should not be relied as the ONLY means of preventing 
> shocks and
> electrocution. The cap bank should ALWAYS be discharged manually 
> with a
> shorting wand or equivalent before handling the caps. If there are no
> bleeder resistors on the MMC array, then to be absolutely safe, each
> individual cap should be discharged manually as well.

When you say bleeder resistors are there as one of the safety devices 
and that is a reason for them being there, your logic completely 
escapes me. Safety is a very incidental factor. There is no such 
thing as a non dangerous shock. Given the right circumstances any 
shock can kill you. You are right. Before you put your grubby little 
fingers on a capacitor in the MMC you should discharge it manually.

> I highly recommend adding resistors across caps in an MMC as one of 
> several
> human safety precautions. Cap protection is achieved by using caps 
> of all
> the same type and having an ample voltage margin, e.g. run the caps 
> at 1/2
> or 1/3 of their DC ratings.

I have no issue with the desirability of the resistors or any other 
circuit details of the MMC. My issue is giving some one the idea that 
just because there are resistors across capacitors it is safe to touch 
them. It just isn't. The d.c. working voltage of the capacitors is 
beside the point.

There are many on this list who have never had any experience with 
anything before solid state and its typical low voltage. Tesla coils 
are a different thing entirely in that all voltages encountered are 
typically lethal. I don't think any one intends to give the idea that 
bleeder resistors are a fail safe device. They won't be until some 
one invents one that has a direct short failure mode.

Ray

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

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