----- Original Message ----- From: "Robin van Spaandonk" <[EMAIL PROTECTED]> To: <[email protected]> Sent: Wednesday, June 07, 2006 5:54 AM Subject: Re: Battery shapes
> In reply to jonfli's message of Tue, 6 Jun 2006 10:18:22 -0700: > Hi, > [snip] >>Since I'm not involved in any type of plasma electrolysis and the > [snip] >>Also, please correct me if I'm wrong, but the typical electrolysis cells >>appear to be most nearly "resistive" relative to their waveforms with very >>little or no phase shift. > [snip] > Plasma electrolysis cells have extremely high frequency and very > spiky voltages and currents. It is impossible to speak of phase, > as the frequency components are pretty much random. > That's why I have in the past suggested that the best way to deal > with them is to use a very low pass filter between the actual cell > and the power supply. Absolutely Robin :) This is the only way to go when the frequency contents exceeds measurement capabilities, which is the case in GDPE's nanosecond scale energetic events. > E.g. two large capacitors with a heavy > ferrite core inductor between them (also a small high frequency > capacitor on the cell side). That way, voltage and current (or > power) can be measured on the supply side, and should be > reasonably accurate. You would get LC ringing with this, plus losses in the ferrite. I have proposed a simple no-ringing lossless power smoothing scheme to the CMNS group, it should be tried soon. I can't tell you more for now as the situation is getting complicated wrt publishing. Michel > At worst, input power measurements will be a > little too high due to resistance losses in the inductor. > Regards, > > Robin van Spaandonk > > http://users.bigpond.net.au/rvanspaa/ > > Competition provides the motivation, > Cooperation provides the means. >
