Wikipedia:Reference desk/Archives/Science/2013 November 7

= November 7 =

Close Encounters of Jupiter
I am writing an article about the comet 40P/Väisälä, and one of my references say that the comet will pass 0.09639 AU from Jupiter in 2127. What would such a close approach to Jupiter do to the orbit of this comet? If possible, could I see the equations used so I can calculate orbital perturbations in the future? Thanks, Carbon6 talk 04:02, 7 November 2013 (UTC)
 * First, 0.1 AU is still 15 million kilometers, or ~40 times the distance from the Earth to the Moon. So while it is close, it is not extremely so. Jupiter has ~300 Earth masses, so in sum the gravitational attraction will be be about 5 times less than that between Earth and Moon. Since this is a general n-body problem, there is no closed solution - as far as I know, the analysis of such situations is often done using step-wise iterative simulation, basically applying Newton's law of universal gravitation and his laws of motion, although I wouldn't be surprised if modern models also incorporate secondary effects. --Stephan Schulz (talk) 10:37, 7 November 2013 (UTC)
 * Looking at Carbon's reference led me here which will calculate the comet's ephemeris at relevant times. However, to what extent the orbital elements will change, I don't know. Thincat (talk) 12:02, 7 November 2013 (UTC)
 * You can change the ephemeris type in that calculator to elements and get orbital elements to till year 2200... so inclination is expected to change from 11.5 degrees to 9 degrees. Close encounters also act to magnify observational errors on current orbital elements, but I don't think it matters much here. Wait long enough (few Myr?) and everything is chaotic. 88.148.249.186 (talk) 14:01, 7 November 2013 (UTC)


 * You need to know how fast the comet will be moving as it approaches Jupiter in order to calculate how much Jupiter will affect it. See http://www.bogan.ca/orbits/gravasst/gravasst.html for a description of the basic calculation. Looie496 (talk) 15:40, 7 November 2013 (UTC)
 * If that's the case, then it might help to know that the comet is moving at a relative velocity of 10.23916 km/s. What result would that give? Carbon6 talk 01:59, 8 November 2013 (UTC)

Digital to analog signal
I have a wire that has a 5v digital signal on it. For an example, think of an HD activity light on a computer. I have an LED on it and it blinks on/off. I want to get a better feel for if it is blinking a lot or a little. My goal is to connect it to an analog panel. I already did that. It pegs at 0 and 25 (the maximum value on the meter) with nothing in between, except that it sweeps from left to right very quickly when pegging from one side to the other. My assumption is that I can place coil or inductor on the line to inhibit the digital change and make it sweep slower. I know the voltage is 5v. I do not know the amps, but it has to be ridiculously small because it only drives an LED. If an inductor is the proper solution, what size? Is there a way to easily adjust the voltage so the meter won't peg hard when it hits 5v? I assume a resistor will do. — Preceding unsigned comment added by 209.149.113.217 (talk) 17:46, 7 November 2013 (UTC)
 * A resistor and capacitor are commonly used for this sort of situation. Basically, you're trying to filter the square wave coing in into a smoother analog signal. The 5V will slowly charge the capacitor, and the resistor slowly brings it back down. (Slowly being relative - slow compared to the near instant hange of the digital signal) If it is going on and off quickly enough, the simple filter will average it out to a pretty steady voltage somewhere in between. K ati e R  (talk) 18:18, 7 November 2013 (UTC)


 * With analogue panel meters, the scale does not necessarily correspond to the full scale deflection (FSD), particulalrly if the meter is second hand or has been removed from equipment. For example, a common FSD is 1 mA; 50 microamp is also common.  If the scale is marked 0 to 25, a 1 mA FSD meter may, for example, been used as a 25 volt voltmeter by means of a series resistor of 25 kohm.  Or it may have been used as an ammmeter by means of a shunt resistor.  You need to do the following steps:-
 * 1 Determine the FSD current of the meter.  Remove any shunt resistor.  In a good quality panel meter, the FSD current will be marked on the bottom of the scale in an unobtrusive way.
 * 2 Calculate the series resistance required to get the FSD current at 5 V.  For a 1 mA FSD meter, you need 5 kohm.
 * 3 Choose two approximately equal standard resistor values that in series make up the calculated value or just slightly greater. Using 5 kohm as the calculated example, choose say 2.4 kohm and 2.7 kohm.
 * 4 Wire the two resistors in series with the meter.
 * 5 Connect an electrolytic capacitor between the mid point of the two resistors and the opposite terminal of the meter.  The capacitor should be sized about  c = 4000 T/R  where c is in microfarads, T is the response time you want in seconds, and R is the total resistance in kohms.
 * Don't connect a capacitor directly across the terminals of an analog panel meter as you will get an undesirable bouncing of the pointer.
 * 124.178.58.238 (talk) 23:55, 7 November 2013 (UTC)


 * How fancy? Oversampling and decimation is a very common technique to convert a single bit stream into an analog signal.  It's the sampled version of an integrator mentioned above.  Basically, when the signal is 5V, each sample puts a fixed charge on the integrator while each 0 subtracts.  See Delta-sigma modulation. Or even Pulse-width modulation for ideas.  The tradeoff is the switching noise is pushed into the sampling band allowing a LP filter to have a higher cutoff (smaller caps) or keep the same cutoff for more accurate reading.   --DHeyward (talk) 08:12, 8 November 2013 (UTC)