Wikipedia:Reference desk/Archives/Science/2013 March 20

= March 20 =

Two CFL bulbs on same line
I have two CFL bulbs on a single line split to two sockets. Sometimes both will light up. Usually, only one lights up. Both work fine. If I remove one, the other always works. I've had the ground checked. It is good. I've had the wiring checked. It is correct. Checking load while lights are on shows nothing unusual. So, after three electricians checked it, all three said that sometimes you can't put two CFL bulbs on the same line. Seriously? I've been searching and searching and I can't find a single webpage that discuses this kind of thing. Does anyone have information about what could cause this problem? — Preceding unsigned comment added by 71.204.230.66 (talk) 01:23, 20 March 2013 (UTC)


 * Here ya go: http://www.thecircuitdetective.com/


 * 99.250.103.117 (talk) 03:56, 20 March 2013 (UTC)


 * (ec) Possibly related to the power factor ? See CFL#Power quality.  If so, perhaps different CFLs (with different electrical ballasts) may solve the problem.  For example, are these both instant-start CFLs ?  If so, you might have better luck with at least one slow-start CFL.


 * If I understand the issue, it's that while the total power requirements for the two CFLs are well within the limits for the circuit, the instant-start circuit draws quite a power spike to warm it up. Your circuit can handle that power spike for one, but not two, CFLs, so the second one fails to "ignite".  If both were on separate switches, so you only tried to start one at a time, then you'd probably be OK, too. StuRat (talk) 04:05, 20 March 2013 (UTC)


 * Yes, it's not the amount of power drawn that is the problem, but the distortion of the voltage caused by the starter mechanism. This local voltage distortion (no longer sinusoidal) seems to interfere with the starting circuitry of the other CFL.  Connecting a small resistive load such as a conventional light bulb on the same circuit will possibly solve the problem.    D b f i r s   22:56, 20 March 2013 (UTC)


 * It would be interesting if 71.204.230.66 could get hold of an oscilloscope to find out what the voltage waveform actually looks like. I've look at some circuit diagrams for CFL and 'those' don't suggest this problem. However, many different companies make CFL and so maybe the OP is  using ones that don't work well together. Has he tried CFL's from a different manufacturer. Dbfirs comment would then make more sense to me, if it turns out they are  really cheap CLF's. Then, as  Dbfirs and others might agree, a resistor might correct this problem but  a home owner  should not have to mess about like this. Knowing what the wave form actually looks like and knowing the make of said CFL, will help the rest of us steer clear of buying such luminaries, as I can not duplicate this fault with the bulbs at my disposal. (P.S. Do the bulbs say 'Made in China' ?) We should have stuck with tallow candles. Just put a  tapper to them and they were away. They did not flicker during electric storms nor suffer from brown outs, in a hash winter if one  ran short of food, one could always eat them, (with getting mercury poisoning) and you could always light more than two at once – so much for progress.--Aspro (talk) 19:43, 21 March 2013 (UTC)
 * I would advise very very strongly against someone with limited experience of electrical installation connecting a scope to the mains. See Lord Finchely. Tevildo (talk) 20:06, 21 March 2013 (UTC)


 * Be that as it may. He just needs Test probe's of the right impedance. The owner of the 'scope' would surly have these. Or and do it with one's own PC. It is not going to blow up, as several times I unintentionally  touch probes to the HT of a line-out-put valve (tube) and that was about 25 kV. Just got nice pretty blue sparks and a strong smell of ozone.--Aspro (talk) 19:06, 23 March 2013 (UTC)

Panstarrs help for tonight?
I've been trying all month to see this comet, and either weather has been bad or I just don't know where to look. I'm at 40° latitude, the sky is clear, it's 8 pm and it's finally getting dark. A forum on cloudy nights said this comet sets about 8:50, and I think they're reporting from Grand Junction.

I can see the moon and Jupiter. Any ideas where to go from there? I don't think there's a panstarrs plug-in for Stellarium (I'm on a mac anyway) and the lack of solid helpful advice in the news is pretty sparse.Reflectionsinglass (talk) 02:01, 20 March 2013 (UTC)


 * oh, Grand Junction is in Colorado, latitude-wise beneath me. If i have the moon, Jupiter, and the Pleiades all in a line, is Panstarrs anywhere near there? Reflectionsinglass (talk) 02:09, 20 March 2013 (UTC)
 * Try this site: . Shadowjams (talk) 02:33, 20 March 2013 (UTC)

Oh my god thank you, Wikipedia should send you a check. Well the comet is behind a stupid mountain directly to the west and has been behind that stupid mountain for hours probably. I'll probably never get to see it. This is turning out to be a week of disappointments. Reflectionsinglass (talk) 02:39, 20 March 2013 (UTC)
 * Get one of those handy Marvin the Martian Illudium Q-36 Explosive Space Modulators – and get rid of that annoying mountain! —74.60.29.141 (talk) 03:08, 20 March 2013 (UTC)


 * I used to be an amateur astronomer, and I still know the constellations very well. Last night (March 19), I went out at sunset and tried to find the comet, using this for my star map.  I also had a finder scope from my telescope.
 * As the sky darkened, but before I could see any stars, I scanned the western horizon both by eye and by scope. No luck.  About 40 minutes after sunset, it was dark enough to see some of the stars, so I made wild guesses about what stars I was seeing, and used my star map to find the comet's location to within a few degrees.  Still, I couldn't see the comet.  20 minutes later, the sky was much darker, and every guess I made about the stars' identities turned out to be correct.  I still couldn't see the comet, either by eye or scope.  10 minutes later, my hands were freezing, and I gave up.
 * This comet is certainly not very easy to see. Oh well, it's coming back in 106,000 years.  Maybe we'll see it then.  --140.180.249.152 (talk) 21:51, 20 March 2013 (UTC)


 * EDIT: I just saw the comet! It was obvious in my finder scope, and had a very bright tail pointing away from the Sun.  It's definitely not visible with the naked eye, even though I knew exactly where to look.  To see it, you need a finder chart, familiarity with the constellations, binoculars (I used a finder scope), and a chair to stabilize your hands.  On top of that, you need to get pretty lucky with pointing your binoculars--the best strategy is to scan the sky around the place you expect the comet to be.  --140.180.249.152 (talk) 00:23, 21 March 2013 (UTC)

The Second Law Of Thermodynamics
Entropy is defined as $$S=-k\sum_{i=1}^\Omega P_i \ln P_i$$. But, really this is a property of our knowledge of the system and not of the system itself. The Second Law says never entropy decreases over time, but The Second Law is a consequence of the fundamental postulate of statistical mechanics (aka ergodic hypothesis) which is not really true in general. I feel it must be possible to decrease entropy, if we have better knowledge of the system. I feel that it must be possible to create an engine which decreases entropy if we consider more factors than the blunt quantities of pressure, volume, and temperature.

150.203.115.98 (talk) 07:09, 20 March 2013 (UTC)
 * Its defined in the first law that the laws of Termodynamics are set for a "closed system". Its a typical model view that reduces or blends out variables of reality to establish mathematical methods that can be managed. --Kharon (talk) 11:30, 20 March 2013 (UTC)

In thermodynamics, entropy is indeed (in principle) a subjective quantity, it is the amount of information about the system you don't have. So, if you have an amount of gas in a box, and all you know about its physical state can be described using a few bits of information (like it's volume and energy, both specified only to a fandful of significant figures at most), then given that limited information you do have, you would have know a huge amount of additional information to know the exact physical state the gas is in. For isolated systems, one can then argue that all the vast amount of states the system can be in given what you know about it, are equally likely. The ergodic hypothesis is not a necessary condition for statistical mechanics to work.

The reason why entropy increases can be explained as follows. At the microlevel, if you did have all the information about the system, you would not lose any information as the system evolves in time. The reason why entropy cannot decrease (in the thermodynamic limit) is then easy to understand. Suppose a system can be in N possible states s1, s2, s3,...,sN, given the information you have about it. Then while the exact time evolution is unknown, this is known to be unitary. This means that after a while, a state sj will have evolved to some state vj, such that vj and vk will be different states if j is not equal to k. So, the number of possible states the system can be in will stay the same, it certainly can't decrease as that would violate unitary time evolution.

Now, this suggests that the entropy will always remain the same, so how can it increase? To see why this happens, we have to consider that after the time evolution when the set of states are the vk, there will be many more stastes the system can't be in, but which have the same macroscopic properties. So, given the macroscopic properties of the system extracted from any one of the vk, like the volume and the internal energy specified to some limite precision, the number of states will be much larger than the number of states vk.

Of course, there is a hidden assumption here, you assume that when extracting the macroscopic properties from a given miscoscopically specified physical state, you throw away information. Entropy can thus only decrease if under time evolution a subset of the sk would evolve to states that would have distinct macroscopic properties compared to the other sk. But due to the random nature of the coarse graining process that yields the macroscopic properties of a system from the exact physical state, that's highly unlikely. Count Iblis (talk) 12:33, 20 March 2013 (UTC)


 * Wow! That's a really neat explanation! I like it a lot.


 * So to pick a simple enough system to understand, let's imagine a strictly deterministic system like Conway's Game of Life.


 * In that game, all of the information about the system is perfectly contained in the initial state of the system. From that initial state (at T=0), all of future time can be perfectly predicted - so there cannot possibly be any more information in the subsequent (T=1,2,3,4,5...) states than we have in the T=0 state.  Entropy cannot decrease because everything you know at T=20 (say) was known from the information at T=0.  Worse still, since any given T=n state can be arrived at from several possible previous states, it's definitely possible for information to be lost because the information you have at T=20 (say) doesn't always allow you to perfectly calculate what the system state was at T=19.  But the information you had at T=0 is enough to perfectly predict the state at T=19 and T=20.  Hence there is more information about the system present at T=0 than at T=20 - so information will generally be lost and entropy will increase.


 * In a system with random elements (suppose you flip the state of one randomly-chosen square of the Conway life game at every time step), your ability to perfectly predict the future from the T=0 state has gone. But T=0 does contain *some* information about future states - albeit only of a statistical nature...but at T=20, you still have more uncertainty about the past than you do about the future - so information was still gradually lost over time.


 * SteveBaker (talk) 13:30, 20 March 2013 (UTC)
 * Unless you also have an excellent model for predicting the (pseudo?) random-number-generator that controls the bit-flipping! But a model of a random-number-generator requires you to store several bits to describe the PRNG state and algorithm (even if you're using a pseudorandom approximation of a truly random physical process)... and so we enter into a Gödel's theorem-esque conundrum: any self-contained system is necessarily less complicated than a complete description of the system.  Nimur (talk) 15:51, 20 March 2013 (UTC)


 * The OP should take a look at our article on Maxwell's demon, which is directly concerned with this issue. Looie496 (talk) 17:32, 20 March 2013 (UTC)

what are these 2 chemical bonds?
those in the picture here. Thank you ! 79.181.122.146 (talk) 11:06, 20 March 2013 (UTC)


 * See Chirality (chemistry). These are single bonds that have been cosmetically modified to represent how the atoms are arranged around the central atom in three dimensions. Plasmic Physics (talk) 11:14, 20 March 2013 (UTC)


 * Okay, that helps ! when u say atoms u mean to the Carbon and it's hydrogens right?
 * could u address me to a 3-dimentional image that shows this cosemtic arrangement u described (the one which the is shown in the chart i attached). it will help me very much. Ben-Natan (talk) 14:27, 20 March 2013 (UTC)


 * Skeletal structure talks about the specific geometric meaning of the "wedged" and "hashed" bond stylings, and also has 3D models of some examples. DMacks (talk) 17:08, 20 March 2013 (UTC)

Question about DC current
I learned that in DC electricity: V = I R

Voltage is Current times Resistance. Rearranging, we have I = V / R.

Now suppose I have a resistor with an imaginary resistance. I also have another resistor with negative imaginary resistance.

If I put them in series R_combine = R_1 + R_2

Where R_1 = i and R_2 = -i

Then would I get an infinite current? I = V / R_combine and R_combine is equal to zero. 68.171.109.163 (talk) 14:05, 20 March 2013 (UTC)


 * Your reasoning is more or less correct. However, what you are doing is dividing an assumed finite numerator by a zero, which has no practical application.  There are applications in electronics and electrical engineering where a positive reactance (which is the correct term for what a positive imaginary resistance is) is in series with a negative reactance at some frequency (the resonant frequency) but there is always a small resistance remaining in pratice.  It is possible to synthesise with active devices a positive or negative resistance that is independent of frequency, however the active devices will of course be driven to overload if the current is not limited by some other component, so no infinate current will occurr. Ratbone 121.215.5.1 (talk) 14:36, 20 March 2013 (UTC)


 * Note that unless you are using superconductors, the resistance of your wires is not zero. If you are using superconductors, they will only carry a certain amount of current before they cease to superconduct. Additionally, your analysis assumes that you have a perfect voltage source. There is no such thing. All real sources will have some finite output impedance, which will cause the voltage to drop as the current increases.--Srleffler (talk) 16:59, 20 March 2013 (UTC)
 * "All real sources ... voltage will drop as current increases"? Not true.  Many practical devices increase voltage as current increases (ie have a negative output resistance - examples are certain configurations of feedback circuits, alternators feeding reactive loads, and more), and it is quite easy to construct a device with zero output resistance, with active devices (magnetic circuits, transistors, tubes, etc) - of course this behaviour will only occur up to a current that overloads (goes into limitting) the active devices.  Note that the words "imaginary" and "real" have special meaning in electrical engineering.  (The former means a something with voltage in-phase or phase reversed with current; the latter means voltage leading or lagging the current by 90 degrees.  I have assumed the OP has been exposed to such terminology but has not a clear understanding of it.)  This terminology comes from Argand mathematics, which considerably simplifies understanding AC electrical cicuits.  Where SrLeffler used the word "real" he probably meant "practical" or "physically realizable". Ratbone 120.145.8.64 (talk) 23:54, 20 March 2013 (UTC)


 * Why bother with imaginary resistances? All you need is a negative resistance. --ColinFine (talk) 17:50, 20 March 2013 (UTC)
 * Isn't negative resistance called "acceptance"? "Imaginary resistance" would be "devil's advocate" or "paranoia". DMacks (talk) 22:09, 20 March 2013 (UTC)

Tasked with getting a digital copy from a cassette tape
So, I've been landed with the task of getting the contents of a cassette tape into digital form...an MP3 or something. I rapidly discovered that cassette tapes are sufficiently obsolete that I no longer own a tape player(!). So a quick trip to the Goodwill store and the princely sum of $9.99 gets me a Technics dual cassette deck with a pair of phono plugs on the back marked "OUTPUT". I clean the heads, plug it into my home theater sound system - and I get good, clear sound from one of the two tape decks. (One of the two decks plays but doesn't rewind...the other rewinds and plays but doesn't produce any audio...I'm sure you can imagine the fun that this causes!)

So I dig into my big box of cables and find one that goes from phono plugs to a 3mm jack - I stick it into the microphone input on my laptop, fire up "Audacity" and get nothing...zip...nada. I plug an actual microphone in there and it works fine. So maybe it's a dead cassette player (what? For $9.99?  How is that possible?!)...but I hooked it up to the "aux" input of my home theater system - and I hear the tape playing just fine. So I do a continuity test on the cable and it seems fine.

I suspect some kind of ikky impedance mismatch between "line" output of the tape player and "mic" input of the laptop. I have no idea how to fix that.

Any thoughts? SteveBaker (talk) 15:10, 20 March 2013 (UTC)
 * No idea how well it worked (sound quality and all), and I'm not trying to advertise for them, but I know my brother was able to move songs from cassette to his computer with a device called a Cassette2USB Converter. His was from a company called ClearClick. I'm guessing the output from a regular tape player won't just work to plug into your laptop and change formats. --Onorem♠Dil 15:29, 20 March 2013 (UTC)
 * Two thoughts:
 * Does the laptop have a "line in" rather than "mic" input? If so, use it.
 * Hook the tape player to the "aux" input of the home theatre system, and connect an output of the theatre system to the "mic" or "line in" input on your laptop. (This essentially uses the home theatre's electronics as an impedence-adjusting device.--Srleffler (talk) 17:03, 20 March 2013 (UTC)
 * No - the laptop only has "mic". I'll take a shot at using the home theatre's line output (assuming it has one). SteveBaker (talk) 01:02, 21 March 2013 (UTC)

Mic inputs are usually around 0.005 to 0.005 volts. Line outputs are usually around 0.3 to 2.0 volts. Find a PC with a line input and use that to make the transfer. Also, make sure that your 3mm jack is stereo, not mono. --Guy Macon (talk) 17:08, 20 March 2013 (UTC)
 * Line-in inputs are typically blue instead of the pink for a microphone. Dmcq (talk) 17:11, 20 March 2013 (UTC)
 * Sadly, on a laptop, all of the sockets are a tasteful shade of black to match the case! But, no - using a "line" input was my first thought.  I guess I'll have to wait until I get home and can use my desktop PC. SteveBaker (talk) 01:02, 21 March 2013 (UTC)


 * If your cassette deck has a headphone jack, try going from the headphone jack to the mic input rather than from the phono jack to the line input. Not sure it will work, but the odds are better. Looie496 (talk) 17:30, 20 March 2013 (UTC)
 * Nope - no headphone jack. SteveBaker (talk) 01:02, 21 March 2013 (UTC)


 * If none of these suggestions work, you could just play the tape and use the microphone to pick up the sound and input it to the laptop.
 * That's exactly the kind of thing I'm trying to avoid! SteveBaker (talk) 01:02, 21 March 2013 (UTC)


 * Sometimes digital transfer software won't accept things that it recognizes as copyrighted, so this may be the problem.--Wikimedes (talk) 17:59, 20 March 2013 (UTC)
 * If that's the case it should be easy to fix by modifying the digital transfer software. Whoop whoop pull up Bitching Betty 19:57, 20 March 2013 (UTC)
 * Since Steve is using Audacity, which is free software and available with sources under the GPL, I think the chance that it enforces some kind of DRM is negligible to begin with. --Stephan Schulz (talk) 20:31, 20 March 2013 (UTC)
 * No DRM coding is possible because this is analogue transfer, not digital.   D b f i r s   08:17, 21 March 2013 (UTC)
 * While this is besides the point since it's clearly not the issue here, I think the statement 'No DRM coding is possible because this is analogue transfer' needs clarification since it depends what you mean by 'DRM'.
 * For starters there's nothing stopping copy protection systems being developed for analog systems depending on various factors (usually taking advantage of the different interaction between components when on is recording vs playing back). There are well known copyright restriction systems which work for analog video signals made by Macrovision and also DCS Copy Protection which often worked initially (helped in the US by the DMCA and the illegality of device able to circumvent it ). I don't believe any such systems exist for analog audio tapes and perhaps none were possible, but it doesn't negate the fact you can't assume it isn't possible without sufficient understanding of the system and the way the various components interaction. (Note that a number of digital copy protections systems for audio and video are not some sort of encryption and software enforced protection but again taking advantage of the way the components interact, check out [[Copy protection].)
 * And of course, as highlighted by the DMCA and later digital systems, there's nothing really stopping the development of copy protection systems in entirely analog systems relying on some intentional feature enforced by the law, contracts or pressure on manufacturers. It may be easier to circumvent and you do have to find some way to force manufacturers to implement it but it doesn't mean it can't exist, there is for example CGMS-A. (These usually rely more on contracts and pressure as even the DMCA doesn't generally require devices to positively respond to measures to prevent copying although again this doesn't negate the possibility of such laws.) Again, these were easier to implement in video since there was already part of the signal not intended to be outputted directly to the end user, but it doesn't really negate the possibility of developing an analog only system for audio which has such protections.
 * Anyway Wikimedes said 'sometimes digital transfer software won't accept things that it recognizes as copyrighted' not DRM or analog only. As Wikimedes hinted by this, for cases where digital processing is involved, the digital device or software could refuse to play if it recognises some sort of sign/signal the content isn't 'supposed' to be played back. A notable example here is Cinavia in the audio of theater soundtracks and Bluray media. As other respondents indicated, it is perhaps unlikely free software will enforce such protections, although probably not impossible if it's made optional (it's been done at least for PDFs .) And there is also the risk the hardware may do it in such a way the free software can't bypass it (without different hardware).
 * P.S. This is all discussed to some extent at Analog hole.
 * Nil Einne (talk) 17:04, 21 March 2013 (UTC)


 * How about a decoupler attachment from line out jack to red and white audio component? 165.212.189.187 (talk) 19:53, 20 March 2013 (UTC)
 * That's just the same as trying a different cable. It is possible that the jack is somehow not connecting.  I would suggest connecting the Line Out across a 1000 ohm and 100,000 ohm resistor in series, then take the output just across the 1000 ohm resistor.  Other than bad connections, the only explanation I can think of is that the input is switching off because of a mis-match in voltage or impedance (though usually you just get distortion, as mentioned above).     D b f i r s   08:17, 21 March 2013 (UTC)


 * it doesn't produce a clear picture of what's wrong, which is always a bad sign. you know the law, if you aren't 100% sure you've found the problem, then you haven't found the problem. if the line output were too high for the mic input, you wouldn't get nothing, you'd get distorted noise, so that's not it. you have output at the line output, so that's not it. the mic input works, so that's not it. it's  possible the phono to miniphone cable is bad in both channels, but what are the chances and you checked the continuity (although it might be shorted;did you check that?). even if the phono to miniphone had a mono plug rather than stereo, it would still carry signal into one channel so that's not it. i doubt that the software is smart enough to recognize casette analog output as copyright so that's not it (although I'm still stunned by my discovery that analog video can contain a no-copy flag, but video's a lot of bandwidth compared to audio cassette). so the answer is clearly that nothing is wrong, and you only think it's not working. (hey, if that works for mechanics and computer tech guys, then why not for us here?) Gzuckier (talk) 20:24, 20 March 2013 (UTC)


 * Nah - it's not copyright. It's my g/f singing at age 6 and her father noodling around on guitar...there is no way that some evil DRM snitch is involved...and (as I said) I'm using Audacity which certainly doesn't have anything like that. SteveBaker (talk) 01:02, 21 March 2013 (UTC)
 * Another possibility is that the player is stereo and the tape is mono and you're only getting the side that wasn't recorded rather than both tracks. I don't see how the home theatre system worked though if you used the same plug. Maybe the audio system is set up to only take one side rather than both, it is worth looking at the sound control in the control panel and checking what it says it supports. It is also better than Audacity for checking things are working in the first place. Dmcq (talk) 11:28, 21 March 2013 (UTC)


 * A method that I have used, is to plug my cassette player into the aux of my DVD recorder and make a DVD (picture is irrelevant). I then take the DVD and rip the sound on the PC. Why you ask? My HiFi and PCs are in different rooms and it is much easier to carry a DVD. --TrogWoolley (talk) 13:58, 21 March 2013 (UTC)


 * reminds me of the old days, when people would so often find that the fastest network protocol was sneakernet. Gzuckier (talk) 17:09, 21 March 2013 (UTC)


 * Try a web search using "connecting line output to mic input". Here's a link that seems relevant: . Good luck! --108.36.120.166 (talk) 01:36, 23 March 2013 (UTC)

Botany - A wonderful Gothic flower !
can you identify ?! — Preceding unsigned comment added by 79.180.120.107 (talk) 20:09, 20 March 2013 (UTC)
 * They are decaying Fuchsias. See Fuchsia. μηδείς (talk) 20:52, 20 March 2013 (UTC)


 * It is a Fuchsia, but I don't think it's decaying. I think it's one of the varieties bred for dark petals, such as the 'blacky' or 'black prince' seen here . SemanticMantis (talk) 00:50, 21 March 2013 (UTC)


 * I was wondering if that might be the case, but the petals do darkens as they die. μηδείς (talk) 17:29, 21 March 2013 (UTC)

Buying a small amount of food-grade HCl
Does anyone have suggestions for how I might purchase a smallish (i.e. not huge industrial-sized) container of food-grade hydrochloric acid for some food chemistry experiments? I searched for food-grade NaOH also, and I found a number of websites where you can simply order it online, such as http://www.essentialdepot.com/servlet/the-Sodium-dsh-Hydroxide-dsh-Lye-dsh-Food-dsh-Grade/Categories. It would be great if something exactly like that existed for HCl. Any ideas? —Keenan Pepper 20:35, 20 March 2013 (UTC)


 * In the UK I've used “analytical grade reagents” such as HCL - which will be suitable for this use. I don't know what you mean by bulk but I buy Winchesters of 2.5 litres. HCL is very useful for unblocking drains, rust removal and other stuff, so anything left over never goes to waste.--Aspro (talk) 21:43, 20 March 2013 (UTC)


 * These folks make, sell and distribute Food Grade HCl:, and: -74.60.29.141 (talk) 22:02, 20 March 2013 (UTC)
 * You might check at a laboratory supply store in your area. University chemistry labs might use food grade in "smallish" amounts. 74.60.29.141 (talk) 21:54, 20 March 2013 (UTC)
 * Any university lab that will let a member of the public have any access to any of their chemicals will potentially be in big big problems. I wouldn't count on it. Fgf10 (talk) 08:12, 21 March 2013 (UTC)
 * Sounds like something that a local home-brew store or club could help find. DMacks (talk) 22:08, 20 March 2013 (UTC)
 * NaOH is used to make lutefisk but what food dish is HCl used for? Rmhermen (talk) 02:42, 21 March 2013 (UTC)
 * I don't know, but 13% of HCl produced by ReagentChemical is used for food: —74.60.29.141 (talk) 03:07, 21 March 2013 (UTC)
 * Cheese can be made with HCl to coagulate the milk proteins. In a bakery were real pretzels are made NaOH is available. The brown crust is formed when you dip the pretzels into 3-4% NaOH  and then baked.--Stone (talk) 12:07, 21 March 2013 (UTC)


 * A local pharmacy would be a good bet - particularly if they're a compounding pharmacy (rather than something like a convenience store that happens to sell drugs). Although they might not have it in stock, they'll frequently be able to order pharmaceutical-grade (safe for consumption) chemicals. -- 71.35.100.68 (talk) 16:58, 21 March 2013 (UTC)


 * Okay, I found that the magic letters to search for are "FCC" (referring to "Food Chemicals Codex"). In catalogs it doesn't say "food-grade", it says "Hydrochloric acid, FCC". Example: http://www.sigmaaldrich.com/catalog/product/aldrich/w530574 —Keenan Pepper 19:38, 21 March 2013 (UTC)

twin paradox - calculating how many years difference
The twin paradox article says that if someone travelled away from Earth at high speed and returned, they would have experienced only a short time passing but on Earth people would have experienced a longer time passing. The math there is beyond me, but there is a simple example about a person travelling at 80% of the speed of light 4 light years away and back, for an elapsed 6 years for the traveller and 10 years on Earth. I would like to know how to calculate similar numbers given only the time elapsed on Earth. (You can assume 80% of the speed of light, or 95%, or anything you like.) So, if 20 years elapsed on Earth, is it possible to calculate how far the space traveller went and how many years elapsed for the traveller? And for 200 years on Earth, do both time and distance just get multiplied by 10 or is it more complicated than that. Thanks. 184.147.116.201 (talk) 22:42, 20 March 2013 (UTC)
 * It's more complicated than that, because you've said "for a = b+c, if a = 10, what are b and c?" Specifically, there will be an infinite number of distance traveled / time elapsed pairs for each "x years on Earth" scenario you want, ranging from "x years, 0 distance" (not moving) to "limit approaches 0 years, limit approaches infinite distance" (asymptotically approaching c).  If, though, you specify two data points (say, stationary twin's time and moving twin's distance), a single answer can be provided (subject to some assumptions like constant velocity on the part of the moving twin). &mdash; Lomn 22:52, 20 March 2013 (UTC)


 * Sure. If we assume the moving twin travels at 80% the speed of light, then 20 years would correspond to a traveled distance of 16 light years, since d = vt.  So the target would be 8 light years away.  Time dilation for the moving twin is by a factor of 1/gamma, where gamma = sqrt(1-v^2/c^2) = 0.6.  So the moving twin is (20 years)*gamma = 12 years older upon his return.  --140.180.249.152 (talk) 23:12, 20 March 2013 (UTC)


 * The only formulae you need are $$t = \frac{1}{\sqrt{1-\tfrac{v^2}{c^2}}} \tau .\ $$ which relates the time past on Earth $$t$$ to the time past on the ship $$\tau$$ and the speed of the ship $$v$$, and the simple relation $$v = \frac{2d}{t}$$ where $$d$$ is the distance traveled and the factor 2 comes from assuming a round trip. Dauto (talk) 23:31, 20 March 2013 (UTC)

Thank you, thank you and thank you. Lomn, I can give you two data points, I think - I was interested in calculating, for fun, the legend of Herla. Suppose his three days in the dwarf kingdom were actually three days travelling in a superfast spaceship. On his return 200 years had passed, so can one calculate how far would he have gone in space? 140.180.249.152, thanks. Do your numbers vary according to Lomn? Or is it always the case that 20 years on Earth = 12 years of space travel? (And 200 = 120?) Duato, are you saying the same as Lomn, that I need to know both t and T to find d, or both t and d to find T? Also, what is c? 184.147.116.201 (talk) 00:18, 21 March 2013 (UTC)


 * c is the speed of light - it's a constant, and so long as we know the value of it (which we do), we don't have to worry about it. That leaves three variables in the equation: t, T and d.  It is a rule in mathematics that no matter what the equation, you can only get the answer if there is no more than one unknown variable.  So if you want to know 't' - then you have to specify what 'T' and 'd' are...if you want to know 'd' then you have to specify both 't' and 'T' and if you need 'T' then you've gotta specify 't' and 'd'.  That's basic algebra - nothing specific to physics or relativity or anything.  *any* equation (except for a few very trivial ones) have that problem. SteveBaker (talk) 00:51, 21 March 2013 (UTC)


 * Thank you Steve. OK, so this means 140.180.249.152's solution for me is wrong? And if (for the Herla story) I have t = 200 years (say 73,000 days) and T = 3 days, then I can solve for d. Or at least I could if I can remember how to algebra, but I can't. Please help me further. I can write out Dauto's equation:
 * 73000 = 3 x (1 / square root of 1 - ((2dx73000)squared / csquared) )
 * but what next to get d all by itself on one side of the equation? Thanks for your patience. 184.147.116.201 (talk) 01:33, 21 March 2013 (UTC)

You can get $$v$$ from the first equation
 * $$t = \frac{1}{\sqrt{1-\tfrac{v^2}{c^2}}} \tau $$


 * $$\sqrt{1-\tfrac{v^2}{c^2}} = \frac{\tau}{t} $$


 * $$1-\tfrac{v^2}{c^2} =(\tfrac{\tau}{t})^2 $$


 * $$\tfrac{v^2}{c^2} =1-(\tfrac{\tau}{t})^2 $$


 * $$\tfrac{v}{c} =\sqrt{1-(\tfrac{\tau}{t})^2} $$


 * $$v =c \sqrt{1-(\tfrac{\tau}{t})^2} $$

Than you can take $$v$$ from above and use the second equation of my previous post to find $$d$$.


 * $$d =\frac{ct}{2} \sqrt{1-(\tfrac{\tau}{t})^2} $$

for the numbers you have from the movie/game/whatever-it-is, the time frames are so disparate that the square root is almost exactly one giving a distance $$d \approx \frac{ct}{2} $$ which is a distance of 100 light years.

Dauto (talk) 01:56, 21 March 2013 (UTC)


 * Great, I see it now. Thank you very much Dauto. (It wasn't a movie by the way, just an Old English legend/myth/fairytale that I was wondering how would look with a (real math-based) sci-fi angle to it.) I can now plug in my other values of t and T for other options. Many thanks again. 184.147.116.201 (talk) 11:31, 21 March 2013 (UTC)


 * what if the travelling twin just went around the earth inside the atmosphere and they used cell phones to communicate constantly??68.36.148.100 (talk) 02:21, 21 March 2013 (UTC)


 * Ignoring gravitation (which I really shouldn't), the one circling the earth ages slower. Any radio signal from the orbiting twin will be redshifted when received on the ground, and any signal from the ground will be blueshifted when received by the circling twin (both by the same factor that aging and everything else is slowed down). The cell phone standards are really not designed to deal with this, but ignoring that, the orbiting twin will send audio packets to the ground at a lower-than-expected rate, which would probably be decoded as short snatches of sound interleaved with silence, while the twin on the ground will send too many packets to the orbiting twin, so parts of the audio will be skipped on playback.
 * The general rule is that the elapsed time is the length of the worldline, where the length is given by a formula similar to the Pythagorean formula but with a sign change: $$\sqrt{(c\Delta t)^2 - \Delta x^2}$$. In Euclidean geometry a straight line is the shortest distance between two points, but the sign change means it's the longest distance instead in special relativity—so if one twin never accelerates, that one will always age more. A helix wrapped around a straight line is longer than the line by a factor of $$\sqrt{1 + m^2}$$ where m is the slope of the helix. Likewise, the circling twin, who has a helical worldline, ages less by a factor of $$\sqrt{1 - (v/c)^2}$$, where v/c is the slope of the worldline (i.e., the speed). -- BenRG (talk) 05:36, 21 March 2013 (UTC)