Your Industrial Art For The Day
Turns out it wasn't the filter on Nancy that was the problem. Not willing to undergo more petrol exposure, I took it to the dealer for them to mess with. Hopefully they'll have it fixed by Tuesday. I rented a Chevy Cobalt in the mean time. When I picked the old Lady up from work on Friday I told her I had traded Nancy in for the new Chevy. I though she was going to have a breakdown right then and there.
posted by nate @ 1:19 PM
7 Comments:
Fuel pump?
vacuum tube on a rotary table
When Nancy breads down, then you pay for a temporary replacement.
If CK breaks down...?
tehe
When CK breaks down she'll take you with her, like any good woman would. :)
That isn't just any old vacuum tube, as will be revealed somewhat, hopefully, when Nate's (that piker!) close-up lens kit comes in the mail. That's a Sylvania (discontinued 1977) 5722 noise tube. Yep, a tube specially developed to generate (Shot as opposed to thermal) noise. How, you ask? Well, the 'space charge' effect, which normally diminishes quantum effects in tubes (the effect of surplus electrons is to make them sorta spread into a field) is supressed by making the anode large and proximate to the (directly heated 'hot') cathode. This is a diode tube, so there is (are) no screen grid(s) and, in operation the C+ (super) potential is held fixed. The only way to modulate through current is to modulate the heater! This is made of un-thoriated tungsten, 'cause the secondary emmision from the radio-active element would add another term to the noise function -- don't let 'em tell ya' otherwise, kiddo.
Well the heater, especially when (1200K--second hottest thing in the room, usually!) energized, ain't no wimpy Cooper-paired, Bose-Einstein condensate, no siree! It's a Fermi sea, if there ever was one. Neglecting spin pairing, the (Fermions NOT Bosons) particles (here, electrons) in a Fermi sea cannot have the same energy, as you well know. 'Shot' noise is named after Dr. Schottky (google that genius!) and is not a mere stastical (population-based stochiochastic) effect! Contrary to popular belief. If the Zout of the circuit is held fixed (which it typically is) output noise is directly proportional to filament potential.
Why? Originally as a war tool to jam radios and radars. Then very soon to test the sensitivity of recievers. Because one can modulate the noise power, one can test in two regimes, and impute 'excess noise' to the D.U.T.! As an aside, if we have two noise sources, that are uncorelated, A and B, of equal integral amplitude, and sum them, the resulting signal would have the power sqrt(2)(AorB), rather than 2(AorB) as one might expect. Think about it.
But our use is different. The setup we are using is this circuit, then a fixed LNA, as the noise power from the tube, as with any noise artifact, is quite low. Then a variable gain amplifier, then an RF (14-bit) DAC (our bandwidths are on the order of .5GHz topping out) then a high-speed fifo memory with ping-pong, finally into a PC. Don't worry, this has only set me back $1500, the trick to economy in this is to use Analog Devices evaluation boards rather than test equipment, which would have bumped costs an order of magnitude. Could probably have made it cheaper by a similar margin by scratch building, but couldn't justify the design effort for a research one-off. So in short, the goal is to construct a random-number generator. If I have been prolix, I apologise. My rat-bastard kids never call me, and I have been left alone all day in an egregious case of parental neglect. They say no one knows the state of one's finances better than one's bartender. . .
Wanna see a recent pic?:
http://blog.myspace.com/index.cfm?fuseaction=blog.view&friendID=2245097&blogID=211122900&MyToken=9520e307-2ca9-4767-bb67-427add08ab01
Kinda convouted link to the pic, Its Ben's MySpace page, signaljammer. Then I am to be found in the links to the Beloit pics.
Post a Comment
<< Home