From: Bert Hickman
Date: Tue, 04 May 2004 08:06:43 -0500
Subject: Re: [o-o] [TSSP] Modeling with ANSOFT + attached picture
Marco, The distributed liquid divider you've described sounds very interesting! Do you recall which book described this device? Since you will be measuring only RF voltages, you may be able to use a simple capacitive divider instead of trying to fabricate a compensated divider. Malcolm's suggestion of using a chain of capacitors down the center of the resonator is excellent, and I seem to recall that Greg Leyh tried this approach using a chain of ceramic (doorknob?) capacitors to add bulk C to a small topload. However, I don't think he ever tried using the chain as a voltage divider and I don't recall if the chain was immersed oil within an insulated pipe (probable) or if it used grading rings. In order to obtain greater measurement accuracy and reduce dC/dV effects a chain of low loss film caps could be used. The coaxial geometry of the resonator and central divider chain should also make electrostatic modeling a bit easier. Best regards, -- Bert -- -- -------------------------------------------------------------------- We specialize in UNIQUE items! Coins shrunk by Ultrastrong Fields, Lichtenberg Figures (electrical discharges in acrylic), & Scarce OOP Technical Books. Stoneridge Engineering -- http://www.teslamania.com -------------------------------------------------------------------- Marco.Denicolai@tellabs.com wrote: > Hi, > > Suddenly it came to my mind, does this list trash automatically every > post with something attached? Just in case, here is again my email dated > 3.5.2004, without attachment. > > --------------------------------------- > > Hello all, > > I have been using more Maxwell and it feels huge, old and instable. Then > > I started using Bela and I like it more. The attached picture is copied > from its output. There I modeled only the toroid, the primary and a > double guard rail. On one side of the toroid a 25 mm sphere to start the > > streamers. That replaced my previous copper pipe bump because it is > easier to model. > > I'll try to get the secondary modeling equation from the Etesla source. > > In the HV lab they have giant voltage dividers, but they work either > with DC or 50 Hz AC. Already to calibrate my SMPS we built a resistive > divider using one 138 MOhm arm-sized resistor and a 150 kohm usual one. > The transient response was simply orrible. We had to wait for the > voltage to stabilize to take a good reading. > > On one book I saw a picture of a simple balanced resistive divider. > Balance came from geometrical symmetry. It was just a column of liquid > with Rogowski profile top and bottom caps. A grid was positioned inside > the tube, near to the base to achieve a balanced resistive, capacitive > and inductive ratio. The electric field was also uniform inside the > tube. What do you think about that? > > Best Regards > > >>-----Original Message----- >>From: paul@abelian.demon.co.uk [mailto:paul@abelian.demon.co.uk] >>Sent: 30. huhtikuuta 2004 21:31 >>To: tssp@abelian.demon.co.uk >>Subject: Re: [o-o] [TSSP] Modeling with ANSOFT >> >> >>Marco, >> >> >>>I have still a lot of thinking/reading/measuring >>>before I can tell you anything valuable. It takes time... >> >>Have you anything with which to measure the operating >>topvolts of the coil, to the extent of capturing a >>calibrated waveform into a scope? >> >>To be useful, the measured signal would have to report >>specifically the toroid voltage and not just sample the >>E-field flux from the vicinity of the topload. >> >>This signal, combined with that from a coil-top current >>probe, would allow us to calculate the total streamer >>charge, as a function of time, through the breakout >>event. >> >>We have Q = CV for the topload and its breakout load. >> >>So the current into the topload from the coil is >> >> dQ/dt = CdV/dt + VdC/dt >> >>The measured signals give us dQ/dt, V, and dV/dt, so >>that we determine dC/dt and the 'breakout charge' VdC/dt >>as functions of time. >> >>If we just sample the flux, we get a signal roughly >>proportional to Q, which isn't very helpful. >> >>This is the point on which we are a bit stalled at the >>moment. Sometime ago we discussed the prospects of >>a shielded capacitive divider, perhaps running down inside >>the secondary. >> >>Sorry I'm not familiar with the commercial field modelling >>packages, so can't really help with that. But I think it >>would be easy if you just model the system with charged rings, >>discs, toroids, spheres, etc, and set them to appropriate >>voltages (eg Terry's voltage distributions). >> >> >>>I would like to verify the postulate of 5 kV/cm required >>>to elongate up to, say, 1.5 m and 1 kV/cm required to elongate >>>further. >> >>These average field strengths look reasonable. It shouldn't >>be too hard to verify them. You could take a topvolts >>measurement and then use a closed formula for the field >>around a sphere or toroid - ignoring the coil altogether. >> >>I think Antonio and Godfrey were passing a few suitable >>formulas back and forth a while back, or were they just for >>surface fields? I should think you could just pretend the >>topload is a sphere as far as the average E-field gradient >>is concerned - once you get a little way from the topload it >>doesn't matter much what shape it is. >> >>It would be interesting to see how your 5kV/cm to 1kV/cm varies >>with risetime, bps, etc. >> >>But I think you will need to measure topvolts at some point. >>-- >>Paul Nicholson >>-- >> > > > > ----------------------------------------- > ============================================================ > The information contained in this message may be privileged > and confidential and protected from disclosure. If the > reader of this message is not the intended recipient, or an > employee or agent responsible for delivering this message to > the intended recipient, you are hereby notified that any > reproduction, dissemination or distribution of this > communication is strictly prohibited. If you have received > this communication in error, please notify us immediately by > replying to the message and deleting it from your computer. > > Thank you. > Tellabs > ============================================================ > > . >
Maintainer Paul Nicholson, paul@abelian.demon.co.uk.