Wikipedia:Reference desk/Archives/Science/2009 March 30

Welcome to the Wikipedia Science Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

March 30

why don't sugars taste like alcohols?

For give me, but the alcohol article says nothing about why OH groups (that are not part of COOH groups) on sugars or glycerine don't make their host molecule behave like alcohols. I mean, shouldn't sugars behave like extra strong alcohols, what with their abundance of OH groups? If not, isn't the definition of an "alcohol" as given in the articles inaccurate? I mean, the presence of an OH group on a carbon chain cannot be the only thing that makes alcohols alcohols. John Riemann Soong (talk) 02:06, 30 March 2009 (UTC)

Alcohol is a generic class name for all compounds having a a hydroxy (-OH) functional group. Do not confuse this with ethanol, a specific chemical compound with unique (pharmacological) properties that are caused by the whole molecule, not just by the isolated -OH part. Cacycle (talk) 03:14, 30 March 2009 (UTC)

But, methanols and propanols and cyclohexanols all behave similarly, have similar pKa's ... heck, even my octanol product produced by hydroboration-oxidation smells like all other alcohols I have known. And if I dared, I bet they would taste roughly the same -- pungent, overpowering, and quite bitter. Yet sugars taste sweet. John Riemann Soong (talk) 05:39, 30 March 2009 (UTC)

Sorry can't actually answer your question, but ethylene glycol tastes sweet. --TammyMoet (talk) 09:28, 30 March 2009 (UTC)

Water has an OH group aswell, but that isn't sweet either. The point is that certain molecules will bind to orthosteric site on the sweet receptors on your tongue, these include ethylene glycol, sucrose, sorbitol and many others (mostly sugars though). I personally haven't read into the pharmacology of sweet receptors, but I'm certain there will be papers out there describing the structure of the orthosteric site of those receptors and descriptions of exactly what molecular shape + functional groups are necessary in order to agonize those receptors. --Mark PEA (talk) 13:24, 30 March 2009 (UTC)

Alcohols have one hydroxyl group attached to a carbon chain. Sugars have many hydroxyl groups plus either a aldehyde or a ketone. they are quite different. Dauto (talk) 15:03, 30 March 2009 (UTC)

That doesn't tell me much. Why don't sugars then act like super-alcohols? John Riemann Soong (talk) 15:41, 30 March 2009 (UTC)

What do you expect a super-alcohol to act like? Dauto (talk) 19:28, 30 March 2009 (UTC)

Sugar alcohols generally taste sweet as well. Also while I'm not an organic chemist, I'm pretty sure sugars do of course show some of the same reactions that all alcohols do presuming the aldehyde/ketone groups et all don't predominate. Incidentally the IUPAC name for glucose is 6-(hydroxymethyl)oxane-2,3,4,5-tetrol OR (2R,3R,4S,5R,6R)-6 -(hydroxymethyl)tetrahydro -2H-pyran-2,3,4,5-tetraol. Notice the tetraol there? Nil Einne (talk) 22:22, 30 March 2009 (UTC)

Actually reading about sweetness (the criteria for hydrophobic component and a hydrogen bond donor and a Lewis base 0.3 nm apart) made the light click on. Now what I'm trying to understand is why sugars and imitation sugars (ethylene glycol, artificial sweeteners, or otherwise) don't activate the "superacidic" response from the sour portions of taste buds. Surely the H+ receptors (and those bitter receptors responding to OH groups at the back) must detect those too. Or are the unpleasant sensations associated with OH groups (and the sour sensations associated with COOH groups) suppressed neurobiologically by inhibitive synapses activated by sweetness receptors? John Riemann Soong (talk) 06:06, 2 April 2009 (UTC)

Developing immunity or resistance to a pathogen

This is a general question rather than a request for medical advice. People often get Upper respiratory tract infections manifesting as "colds", sneezing, sore throats, bronchitis, sinus infections, stuffy nose, coughing, fever, or the flu, and after a few days or a couple of weeks make a full recovery, without antibiotics or other medical treatment beyond rest and plenty of liquids. The impression is that the body has "fought off the infection." What mediates the acquired resistance of those who have gotten over the upper respiratory infection? What sorts of pathogens would be capable of reinfecting the person if he/she were re-exposed to the exact same germ/virus after recovery, and what sorts would be very unlikely to cause a re-occurrence due to a new exposure? The section "Clearance and immunity" in Infectious disease does not tell me whether the victims of Typhoid Mary who survived typhoid, for instance, would likely be reinfected if they foolishly hired her again as their cook. If she were "Common Cold Mary," "Streptococcus Mary," "Pneumonia Mary," or "Influenza Mary" and were a persistent carrier, how likely would those who recovered from the exact pathogen she passed to them be to suffer a recurrence from her renewed presence, or say if they ate frozen leftovers she had prepared which were full of the pathogen she carried, after they had recovered from the original infection? Edison (talk) 03:01, 30 March 2009 (UTC)

I'd say the following in somewhat indirect response to your question.

Most infections evoke both a humoral (antibody) and cell-mediated immune response.

The response is not always absolute: evoking either or both immune responses is not a guarantee that one is now protected absolutely from reinfection, but rather that one is less likely to develop severe disease on re-exposure to the organism in question.

The response fades over time; protective antibodies and cells capable of cell-mediated response gradually disappear from the circulation. They may be quicker to appear on re-exposure, but they don't just hang out in the circulation forever waiting for it.

With regard to typhoid, see [1]. It seems that within years such protective immunity as develops fades, so re-infection upon re-exposure is certainly possible - and is possible even before, as protective immunity is not absolute. - Nunh-huh 03:36, 30 March 2009 (UTC)

With respect to the Flu and especially colds, the microorganisms constantly mutate, so an immunity ceases to apply, as the next time you're infected it's likely to be a different strain. StuRat (talk) 05:47, 30 March 2009 (UTC)

Ah, but the question specified "Same strain." If you got infected from food Mary cooked, would the frizen leftovers be able to reinfect after your immune system ramped up and you recovered? Similarly, if Person A got sick, and passed the upper respiratory infection to Person B, then A got well, could A be reinfected from B? Do pathogens mutate so fast that A and B could perpetually pass a cold or URI back and forth? Edison (talk) 18:47, 30 March 2009 (UTC)

I'd say no to both cases, or, if you were reinfected, it would be less severe. StuRat (talk) 04:53, 31 March 2009 (UTC)

I agree, but this reminds me of dengue because partial immunity to one serotype, due to prior infection with another, can result in dengue hemorrhagic fever (for details relevant to the current discussion, see the third paragraph under dengue epidemiology). --Scray (talk) 05:32, 31 March 2009 (UTC)

I don't have time right now for a more complete response, but take a look at Clonal selection theory - that's generally thought to be the primary mechanism of immunity. In addition to this mechanism for immunological memory, there is now pretty good evidence that the innate immune system has some capacity for "memory" in up-regulating innate immune components (e.g. natural killer cells, complement system) at heavily-exposed sites of the body, perhaps making people somewhat more resistant to certain exposures. The important factors for immunity (T versus B cells, innate versus adaptive) are likely (in some cases, are known) to depend in detail on the pathogen and route of exposure - so it's risky to generalize. --Scray (talk) 19:17, 30 March 2009 (UTC)

macrophage apoptosis and fever

Hi,

Im am interested in the connection of macrophages and fever, especially if/how macrophage apoptosis is connected to the onset of fever. If I understood correctly, macrophages produce cytokines that have a role in the development on fever symptoms but is this more the case with active macrophages or do the dying macrophages release cytokines in significant amounts. —Preceding unsigned comment added by Vuori12 (talk • contribs) 07:39, 30 March 2009 (UTC)

Since no-one else yet has replied I'll offer my understanding of the matter, without references to back it up. The production of cytokines by macrophages is an active process, and thus requires the macrophage to be (1) stimulated and (2) alive. Sure, an apoptotic macrophage may release some of the cytokines it contains. Nevertheless, apoptosis, as opposed to necrosis, is a process that causes little disturbance to a cell's environment. Apoptosis is programmed cell death, and not uncontrolled release of a cell's contents. --NorwegianBlue ^talk 19:38, 31 March 2009 (UTC)

Humans on Titans surface.

Saturns moon titan has an atmospheric pressure that's actually slightly more than that of Earths, but it's extremely cold. Would it be possible to survive outside on Titan with little more than an oxygen tank and heavy insulation? If not, what exactly would you need to survive outside on Titan? I'm very curious. 63.245.144.68 (talk) 09:31, 30 March 2009 (UTC)

According to our article on Titan, the surface temperature is −179.45 °C, and that's without wind chill. By comparison, the coldest temperature ever recorded in Antarctica is −89.2 °C, and it takes far less than that to really do serious damage very quickly. In simple terms, any exposed skin is going to freeze solid pretty much on the spot. Proper insulation would help, sure, but we're probably talking full-body coverage here -- in other words, essentially a space suit. I'm not sure if it would have to be completely airtight, but you couldn't just dress really, really warmly by conventional standards and wear a respirator and expect to survive. -- Captain Disdain (talk) 09:51, 30 March 2009 (UTC)

Atmospheric composition aside, I don't think there's any place in the Solar System (apart from Earth) where humans can survive the temperature. But please correct me if I am wrong. A Quest For Knowledge (talk) 12:14, 30 March 2009 (UTC)

Go down deep enough in any of the outer planets and the temperature will rise enough so humans can survive. They mightn't like he pressure though :) More seriously Mercury goes around slow enough to give places where a person could survive the temperature for a while. Dmcq (talk) 12:40, 30 March 2009 (UTC)

Oh, depending on where and when you are, Mars sports some fairly nice temperatures -- it can get up to 20 °C in the summer, which means that you could walk around in shorts and not get chilly. Of course, the very low atmospheric pressure would cause other problems, but in terms of the temperature, it wouldn't be that bad. Still, in nasty winter conditions, it can get as low as −140 °C, so you'd definitely better be sure about the weather before you step outside... -- Captain Disdain (talk) 12:53, 30 March 2009 (UTC)

I suppose that temperature is rather important for manned planetary exploration, but not because of the possibility of walking around without a space suit, as pressure and atmospheric composition will likely make those necessary in any case. Where I think the importance would be is in the energy needed to heat the human habitat, or, in the case of a location that gets too hot, the energy required to cool the habitat plus the added weight of an air conditioning unit (or do they automatically include a temperature control unit that both heats and cools ?). StuRat (talk) 14:00, 30 March 2009 (UTC)

A World War 2 vintage heated flight suit could counteract the low atmospheric temperature, along with electrically heated socks and gloves which are presently commercially available. A heated set of goggles would afford clear vision. This is still more arctic clothing than "space suit." Now we should consider any toxic effects of the chemical composition of the atmosphere. I would say the explorer would need an oxygen rebreather or air tanks, due to the toxic atmosphere. These are also presently commercially available. Edison (talk) 01:51, 31 March 2009 (UTC)

monochrome TV transmitter and reciever

please help me with the block diagram and description of a monochrome TV transmitter and reciever —Preceding unsigned comment added by 117.206.32.67 (talk) 12:35, 30 March 2009 (UTC)

This appears to be a homework question, which Ref Desk policies prevent us from answering directly: we believe that the value of homework lies in doing it yourself. However, we also have an article on how television works, linked helpfully from our television article. If you have specific points you'd like to clarify, you're welcome to return here and ask. — Lomn 13:06, 30 March 2009 (UTC)

The how it works article is very good but boy is it going out of date fast. Soon you won't be able to get anything except digital television, and there'll be no need for those dangerous high voltages because the display is an lcd screen or something similar. And who'd bother with a monochrome lcd screen? Dmcq (talk) 15:34, 30 March 2009 (UTC)

Monochrome would be easier to make... but LCD is so cheap now it'd be more expensive to hire a design team. E-Ink is the only thing I can think of that is still in black and white these days. Anythingapplied (talk) 15:51, 30 March 2009 (UTC)

I actually did make a b/w receiver from plans some years ago. I took a fair bit of tweaking but it worked nicely eventually. I don't think I'd attempt that with digital television, and anyway it's cheaper to buy complete products than the components - a bit of a pity I think. Dmcq (talk) 16:27, 30 March 2009 (UTC)

The OLPC XO-1 while a colour screen has a monochrome mode which is sharper and reflective (no backlight) therefore using less power although I guess it's rather similar to e-ink devices Nil Einne (talk) 22:19, 30 March 2009 (UTC)

Could your monochrome screen be better in terms of contrast etc? If so perhaps you could sell a small number to monochrome movie buffs although it's unlikely to be worth it developing oneNil Einne (talk) 05:01, 31 March 2009 (UTC)

This same question was asked in the computing desk and help desk. Please don't do that. Nil Einne (talk) 04:57, 31 March 2009 (UTC)

Send analog values over isolated grounds

I am trying to transfer an analog value (0-5V) from one microcontroller to another. Both microcontrollers are on different power sources (both grounds are isolated) and need to be kept that way. I can easily transfer digital data with optocouplers but I can't find an easy solution for analog signals. Another constraint is that I have a limited amount of I/O's that I can use.

Here are some crazy ideas that I've had to solve this:

• Convert the analog signal to an AC signal with amplitude modulation. Transfer the AC signal using a 1:1 coil and then convert the AC signal to DC. Seems too complicated
• Convert the analog signal with an ADC (8 bit) and use an optocoupler for each line. Too many I/O's

Any suggestions? --jcmaco (talk) 15:42, 30 March 2009 (UTC)

It seems that this could work. http://search.digikey.com/scripts/DkSearch/dksus.dll?Detail&name=516-1609-5-ND The example circuits are quite helpful. --jcmaco (talk) 15:56, 30 March 2009 (UTC)

I'd probably go for the ADC solution, if you use i2c or something similar you don't have too many wires connecting. I believe there's optical isolators for i2c but I don't know how much they cost. Dmcq (talk) 16:10, 30 March 2009 (UTC)

What type of analog signal. If we are talking music this won't work. Otherwise just find a couple of old fax machines and cannibalize the modulator/demodulator circuits. That will get you from analog to digital. Then you can send it through your optocoupler. (Remember to switch the signals, because fax converted digital to analog to go into the phone line. You need it the other way round. )76.97.245.5 (talk) 16:43, 30 March 2009 (UTC)

Telemetry was sometimes passed by a capacitor with switches (reed switches) rapidly opened and closed to alternately couple it to the sender or receiver. Not sure why you couldn't use optoisolators as an analog link. In the pre-digital days, an analog signal was sometimes changed to a pulse rate for telemetry. The pulses could be sent by optoisolator or transformer. Edison (talk) 18:53, 30 March 2009 (UTC)

Firstly - you can get analog opto-isolators (actually, they are more usually called 'opto-couplers' for some reason). But aside from that, most microcontrollers and A-to-D converters have separate analog ground and digital ground for precisely for this kind of reason. Take for example, the Atmel AtMega series (popularised in the 'Arduino' board family) - it has an 'AREF' pin and separate AGND and AVCC for exactly these kinds of reason. You still have to tie the AGND to the AGND of the other processor - but that doesn't connect their digital grounds together. There are of course limits...if the incoming 'analog ground' is a million volts above the receiving systems' digital ground then something is going to fry! I believe that the AtMega allows a third of a volt between analog ground and digital ground. SteveBaker (talk) 22:45, 30 March 2009 (UTC)

Good point. The OP seemed to be talking about sending digital data as well using optocouplers so my reasoning was he could measure the analogue signal near the source and send everything down a single serial link. I've a horror of multiple wires and it removes any need for ADC at the end point. If i2c is too slow he could go up to USB. Though it sounds a bit like the OP wants to generate an analogue signal at the source and then decode it at the destination, I wonder if that is true? Dmcq (talk) 09:43, 31 March 2009 (UTC)

Convert voltage to current and use an LED and photodiode. --Srleffler (talk) 04:39, 2 April 2009 (UTC)

Doppleganger

What are the percentages/odds that there is someone in this world that looks exactly like you, physically? I ask because some of my friends had mentioned that there is this girl that looks just like me but she lives downtown. They see her walking around with her dog and one of them approached mistaking her for me. But when he looked at her closely, he noticed that it wasn't me. Just someone that looks like my twin but slightly different. (FYI - I don't have a twin at all. Already grilled my parents). --8.4.8.12 (talk) 19:29, 30 March 2009 (UTC)

I'm not sure this can be sensibly calculated from first principles. There are two major factors here—the most obvious is genetics, and I'm not sure we have any real idea of how many genes go into physical appearance and the complications of them being expressed. The second is environment, which for something like appearance (and other things, but appearance is comparatively straightforward to demonstrate) is MASSIVE (hence people of different socio-economic classes are often visibly of said class at just a glance, both in terms of obvious accouterments like clothing, hair style, makeup, but also things like nutrition, exposure to sun, exposure to cigarette smoke, etc.).

So I don't think you can calculate the odds in that sense. You could try something more empirical—takes lots and lots of photos and have lots and lots of people record data on how similar they look. Methodologically it might be tough but I'm sure one could figure out something sensible. Of course, in such a survey cultural factors would still play a huge role—Asians famously look more alike to non-Asians than within their own groups, for example—so you'd have to be very careful in figuring out who your "judging" people were.

All that being said, anecdotally, it's not uncommon to run across a number of people who look vaguely similar from a distance. Exact duplicates seem to only really occur rather rarely (not including twinning), but then again the existence of "impersonator/body double" as an occupation for the very famous seems to imply that out of a large enough pool of people you will have a few people who look very similar. --140.247.249.33 (talk) 19:55, 30 March 2009 (UTC)

Wow, that is crazy! Thanks for the input. --8.4.8.12 (talk) 14:31, 31 March 2009 (UTC)

Scroll back up aways and look at the recent question about fingerprints. It's generally asserted that no two people have the same fingerprints. If we believe that - then the odds of two people looking exactly alike is zero because even if they are identical twins with very similar upbringing - their prints don't match. But I'm guessing you are going to say that you don't care about such minor things. OK - so what are the odds that someone of the same sex as you has the same eye and hair color and wears their hair in the same basic style? Probably something like one in 20 ? Certainly one in one hundred - so out of 7 billion people - there are perhaps 70 million who "look like you" - at least to that level of similarity. But then you're going to tell me that this isn't similar enough - you want them to have "similar" bone structure - a "similar" body shape - a "similar" complexion...etc. So here is the problem...there is no decent scientific definition of "kinda-sorta-similar-enough-but-not-really-utterly-identical" - hence there can be no statistics with which to answer this question. SteveBaker (talk) 22:23, 30 March 2009 (UTC)

I'll try to make some rough order of magnitude guesses. A google gander of "facial recognition false positive rate" (minus quote) netted numbers from 0.1 to 2% false matching. Facial recognition software tends to uses facial features' sizes, positions and shape as a metric. Let's say then out of 100 false positives maybe 1 matches your facial structure rather closely. (Close enough to convince someone it was you if you photoshopped your hairstyle, eye color, skin color, freckles, moles, etc over it without altering the dimensions.) Assuming a f.p.r. of 1% then .01% or 1 out of 10000 people has a really similar facial structure to you. I am pulling this out of thin air, but let's say 1 out of 10000 people are the same gender and have similar skin and hair color, height, weight, age, build, and hairstyle. Then that means that 1 out of 100,000,000 (10000*10000) people naturally look eerily similar to you. If there are 6.8 billion people currently living then there are 68 doppelgangers running around out there. At a distance the number of people who look like you will naturally increase. Sifaka ^talk 01:28, 31 March 2009 (UTC)

I'd also add that some people look "more unique" than others. For example, older people tend to look more distinct from one another due to an accumulation of scars and other identifying feature. In the case of women, there's also the effect of make-up, which can make people look more similar than they are without cosmetics. StuRat (talk) 04:47, 31 March 2009 (UTC)

Surely older people tend to look more alike than more distinct? Most of them have grey hair for example. "Accumulation of scars"? Well, not in my neighbourhood fortunately.--Shantavira|^{feed me} 11:59, 31 March 2009 (UTC)

You must have noticed that older people have more scars, moles, age spots, etc., than younger people. Each of those items can potentially aid in identification. And they don't all have gray hair; some dye it, some have white hair, and some have no hair. They also tend to wear a variety of hats. StuRat (talk) 03:45, 1 April 2009 (UTC)

Hello - you're on Car Talk...

I know lots and lots about cars - but not much about cars with automatic transmissions.

Sadly, my son just inherited my Wife's crappy (but super-reliable) Mazda Protege 5 wagon (poor kid) and after just a couple of weeks in his tender loving care, it's broken.

The symptoms started out as a kind of whiney sound that was mostly noticable at low speeds - I guess belts - but that wasn't it.

Today, he said that as he tried to pull out of a parking space - the car started to 'bunny hop' - just like a stick-shift being driven by a driver who doesn't know where the clutch is! However, once he gets it up to speed, it drives just fine.

I'm guessing that we're looking at a torque converter issue - that's the thing in an automatic that does what the clutch does in a manual - right? If that were locking up instead of spinning freely - that would do it. My naive mental picture of one of these gizmos is of a 'fluid coupling' that somehow 'locks' into a solid coupling at speed...if that's true then it kinda sounds like it's refusing to 'un-couple'.

So....

• What kinds of things can go wrong with Torque Converters?
• Can you repair a torque converter? Do you just replace the torque converter?
• What do they cost to fix?
• Is there something else in the transmission that might cause these symptoms?
• Does anyone want to buy a Mazda Protege 5 wagon?  :-)

SteveBaker (talk) 23:59, 30 March 2009 (UTC)

Just to be sure, you did check the transmission fluid (while the engine's running), right ? And you also made sure the proper fluid was used, I assume. With anything involving transmission repair, the labor costs can outstrip the parts, as it's not easy to disassemble a tranny. StuRat (talk) 04:40, 31 March 2009 (UTC)

Yeah - the labor cost is what bothers me. I haven't checked the transmission fluid. I don't see where it gets topped up - there are a bunch of obvious fluid containers - but I don't see any transmission fluid. But those things aren't always easy to find - right? As I said - I've never worked on automatics, but also, we don't have a copy of the shop manual and my son is 200 miles away - so I'm trying to diagnose this over the phone! Last night I told him to get it towed to the nearest transmission shop (we have AAA) and have them replace fluids, filters and seals and anything else that seems transmission-related and costs under $100! SteveBaker (talk) 12:37, 31 March 2009 (UTC)

The dip-stick for tranny fluid is typically near the firewall, down low, extending from under the engine block on either the right or left side. When checking the fluid also look at the color, which should normally be red or pink. Note that a transmission fluid flush often makes problems worse, so don't fall for that. If the color is bad, change the fluid. If the level is low, but the color is good, just add fluid. StuRat (talk) 03:40, 1 April 2009 (UTC)

This really does not explain the second problem, but your initial whiney sound resembles one which I had recently. Turned out to be the break pads. Here in the UK things like this get checked up on MOT tests, but I'm not sure what it's like for you in the States in regards to cost. Best way is to get a mechanic to do a full check of the systems that could likely cause such an issue (breaks, suspension etc) and if you can't find the problem there, then assume the engine. —Cyclonenim (talk · contribs · email) 12:05, 31 March 2009 (UTC)

No - it's definitely not brake pads - the car makes the noise even when your foot is not on the brake. Also, the car passed it's Texas State Inspection (the Texas equivalent of the MOT) just a couple of days ago - AFTER the whining noise started - and they check the brakes. I got the chance to drive the car when the whining started - and it sounded exactly like a belt slipping - but the belts are both new and correctly tensioned. I guessed that there might be some oil on one of the belts or something...but that wasn't it. As my son later realised, it wasn't making the noise when the car was idling - or after the car had built up speed. Which fits the 'torque converter' theory...but now that it's doing this 'bunny hop' thing, I'm thinking that the 'lockup clutch' has locked up and won't unlock. SteveBaker (talk) 12:37, 31 March 2009 (UTC)

Don't assume that the brakes aren't in use just because the brake pedal isn't depressed (and how could it not be depressed, what with people stepping on it all day ? :-) ). A fairly common problem is for the emergency/parking brake to get stuck at least partially engaged. StuRat (talk) 13:54, 2 April 2009 (UTC)