That's why I wrote "so far, I just started the Baysian training..." The whole point of a Baysian filter is that it will train itself to ignore an attribute when that attribute is paired with an attribute that is unique to a poster that I want to read. If you wish, I will post the filtering rules after it gets trained.
Yes, and I don't see where that is a trick. I thought that was common?
any ways, we can't removed the ripple in this application. At the end of the stack, a capacitor coupling is driving a RF xformer used to power the heater on the beam tube. yes, the rectifier stack not only supplies the B(-) for the cathode, it also heats the heater via the ripple that is still present in the DC component. Motorized dell rods go down through the middle of the stack where they turn rheostats to adjust the heater voltage. The Rheostats are mounted on the high end of the stack.
When setting up the process, the operator needs to just the heater level that goes with the voltage level applied. There is a problem with this kind of system, there is a minimum point of operation due to the design of the heater system. Other systems use motorized generators etc..
The capacitors in the multiplier array/stack are corona half rings that sit very close to the vessel wall. Spacing and gas density plays a roll in the value of the capacitance of each ring.
It probably is, but I have seen physics students build them from the original 1951 paper by James D. Cockcroft and Ernest T. S. Walton with equal value capacitors. I probably should look up the original paper and see if it mentions larger capacitors near the bottom of the stack.
OK, I looked it up. The 1951 date is when they got the Nobel prize. The first Cockroft-Walton voltage multiplier was it seems, made in the 1930s. It looks to me from the picture that it had equal capacitors.
[
It is called stored energy, and the caps in the multiplier string as well as any storage placed across the outputs changes the ripple figure, not that you could place a cap across a 5MV tap.
The ripple will be there regardless since there is no easy way to incorporate a "smoothing cap" across the output with any great ease or success at those voltages.
It's like trying to capture a continuous lighting bolt.
What are you exciting at 5MV anyway?
Far out. Got some pics?
Have you seen the rectifier stacks at Bonneville? They are like 50 feet tall. Huge thyristors 'plates'.
THIS particular multiplier design has equal potentials across ANY one segment in the string. ALL the way thru any number of stages.
There are, however, multiplier configurations that push more power, and there are those that pump more voltage per stage, and there are those that fire out a center tapped dual voltage output.
Depending on exactly how the stages are configured, voltage at each component in a stage may or may not grow.
With the C-W configuration, it is the same across all components in all stages.
That is why it is the most common choice in design. No need to use different diode and cap elements along a string of multiplication stages. The amount of energy stored in any one stage should also be equal. The capacitance should be increased as much as possible, but one should not store more energy than one plans to use. So the best way to reduce ripple or PARD here is finding the right frequency of operation, and the right amount of energy storage to make everybody warm and fuzzy inside. There is a happy point.
So, you take your diode drops and you subtract that from your transformer HV side out (secondary) from your step up transformer voltage to get you per stage multiplication factor to get near what your output voltage should be.
Yes,all equal sized .
Modern applications, however are low wattage,switcher fed front end devices that do not take up a hundred cubic feet of space.
Many take up the space of a couple packs of cigarettes.
You can place the final cap to common at the end if you can find one that can handle the voltage.. I don't think a 5Mv one exist. I could be wrong there !. :)
In any case, we wouldn't be able to energize the heater in the beam tube at the end of the HV stack if we didn't have a ripple in it. Its funny how a hindrance actually aids the design in other ways.
100Khz was selected in the design so that low value cap size could be used along with litz wire pie RF xformers.The oscillator design is a armstrong type. Tickler coil is actually the plate coil and feed back coil which sits in the pie rack assembly as part of the feed back tanked with some caps to select the 100 Khz freq.
The heater in that tube takes aprox 260 amps at 12 volts AC to warm up. Supply voltage is variable up to 15kV DC coming from SCR 3 phase firing stacks.
I am curious with the pipe gone, do passing trains and forklifts still disturb the beam (with the field cancellation turned off)?
That's a great "mine is bigger than yours" photo!
One would think so. Afer all it is a huge ferro-active mass.
we'll know that most likely by the end of this week, we must give it some time to determine that.
Sky Shine, 1. Noun, reflected radiation off a metal roof. 2. Noun, sun light from holes in a old roof. 3. Look on a couple'sjoining the mile high club.
Which is it?
Steve
"Rad Shine" sounds more appropriate then.
Def number1 should read "reflected SOLAR radiation...".
How long do you figure to give it? .
Have something to add? Share your thoughts — no account required.
Ask the community — no account required