Wikipedia:Reference desk/Archives/Science/2016 May 26

Science desk
< May 25	<< Apr | May | Jun >>	May 27 >

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.

May 26

Pure gold as a structural material

Say you want a solid gold statue of a dude sitting on a throne. The throne is something many times stronger than gold like stone or steel and is not attached to the gold.

1. How many times life size could you make it assuming enough gold?
2. What body part would deform the easiest?
3. How big of a cube will not flow?
4. How much of a range is there in which gold will settle over time but not immediately deform before your eyes?
5. If all other humans suddenly disappeared and I wanted humanity's legacy to be a 20 meter gold building (because), where should I put it for it to survive the longest?
6. Should I bother to mortar or try to find a way to weld the gold bricks or is it a waste of time?
7. I have to dig to bedrock if I don't want the soil pushed up around it, right?
8. What's the best shape?

Cube:

Pros: No sharp tip to be melted every time lighting hits like pyramid. Cube with melted corners and edges looks better than pyramid with melted tip. Hides the face with dust and bird poop on it from the ground. Flatter faces.

Cons: Won't be as tall as pyramid. Can't incorporate stairway to work site unlike pyramid.

Pyramid with flat top:

Pros: Lightning resistance with most of the benefits of pyramid (easy to build, tall, pounds per square inch, few bricks in the upper half)

Cons: Not willing to spend 50 years building something with parochial overtones. Resembles weird seal of US:

Hemisphere:

Pros: Least likely to attract lighting for the height.

Cons: Hard to build.

1. How many years will it last and will it be melted to a puddle by lightning, flow during earthquakes, be buried by sediment still in good condition..?
2. How hard is it ship gold overseas in a pandemic kills everyone but you-type scenario? If committing grand theft ship is too hard it'd have to be between Arctic Canada and Panama or maybe South America (Darien Gap). The tropics would require finding a lot of vaccines first and above ~the 40th parallel the next ice age would screw up alignment and levelness at minimum.
3. How hard is this with only general nerd knowledge?
4. How big could I build it before I die?

(I probably could never really get to much of the world's gold bricks but this is already fantasy) Or maybe I could? Step 1: Walk 10 miles to nearest Ivy League library Step 2: Find out where the explosives & gold are Step 3: Steal truck.. Sagittarian Milky Way (talk) 01:42, 26 May 2016 (UTC)

Sorry to reformat your comments, but this was taking up a massive amount of space. Nyttend (talk) 05:29, 26 May 2016 (UTC)

Because gold is expensive, most things made from it are hollow, so you should be wanting hollow structures, if your budget is limited. Graeme Bartlett (talk) 09:04, 26 May 2016 (UTC)

3 and 4 are actually pretty interesting. My intuition says there is some size gold cube that would start to sag/flow even at standard temp and pressure, but I don't even know how to pursue this further. All I can think of is the pitch drop experiment, which at least shows an example of something that does flow but can't really be seen. Angle of repose and mass wasting are relevant, but still don't help me understand what would happen with a giant cube of gold...SemanticMantis (talk) 15:14, 26 May 2016 (UTC)

Here's a rough calculation: Materials deform under stress, which is force per area. This can also be expressed as Density*Volume*gravitational acceleration/area, which for a cube of side length "a" reduces to density*a*gravitational acceleration. The Gold article lists a tensile strength of 120 MPa, but we're looking for compressive strength. Some Googling turned up this paper [1] where figure 4a seems to indicate a compressive strength of a bit less than 100MPa. Since it's a rough calculation take the strength to be 100MPa. Setting this equal to the expression for stress and solving for "a" (density is 19300Kg/m^3 and g = 9.8m/s^2) gives a = 528.7m, which according to 86... would require much more gold than all the available gold on earth. (Someone should check my math.)

For the basic concepts of deformation over time, see Creep (deformation).

--Wikimedes (talk) 08:45, 27 May 2016 (UTC)

Arsenic addiction

Is it possible to be addicted to arsenic? The article James Maybrick contains the words "arsenic addiction", but it also says "While there in 1874 he contracted malaria, which was then treated with a medication containing arsenic and the result was that he became addicted to the drug for the rest of his life.". I found some results on Google but no definitive answer from a reliable source. Page 64 of The Maybrick A to Z by Christopher Jones, [2], [3] [4]. The Quixotic Potato (talk) 11:19, 26 May 2016 (UTC)

Note: an IP said to represent a banned user posted a useful reference about development of tolerance to arsenic used for malaria prophylaxis and treatment - http://chestofbooks.com/crafts/scientific-american/sup3/Arsenic-For-Malaria.html . This was deleted by User:GoodDay but since the reference is useful I've preserved it. Wnt (talk) 12:49, 30 May 2016 (UTC)

My assumption is that the original meaning was that the man had chronic malaria, and used this drug to control it for the rest of his life. "Addiction" in the 1800s was a rather vague term, I think. Wnt (talk) 12:36, 26 May 2016 (UTC)

I am not a native speaker and I am not a medical professional, but I would guess that "he became addicted to the drug" means that he became addicted to the medication (of which arsenic may have been an ingredient, but probably not the ingredient that made it addictive. To me something like laudanum would be more likely to be the addictive ingredient). The Quixotic Potato (talk) 15:11, 26 May 2016 (UTC)

Arsenic#Medical use maybe will help in your research? --Jayron32 12:38, 26 May 2016 (UTC)

I've read it, but unfortunately it doesn't anwer the question. I'll go read the articles it links to. The Quixotic Potato (talk) 15:17, 26 May 2016 (UTC)

From that section: "In subtoxic doses, soluble arsenic compounds act as stimulants, and were once popular in small doses as medicine by people in the mid-18th to 19th centuries." --71.110.8.102 (talk) 15:28, 26 May 2016 (UTC)

Good point; I guess any stimulant can be addictive, even if it is incredibly dangerous stuff. The Quixotic Potato (talk) 16:00, 26 May 2016 (UTC)

Well, this isn't much of an argument yet. I mean, you could say that soluble phosphorus compounds tend to be nerve gases, or nucleic acids. From [5] we get some names of antimalarial arsenic compounds: neo-kharsivan (hmmm, what's kharsivan?), novarsenobillon, liquor arsenicalis (with or without strychnine!). Yikes, doctor, are you trying to cure me of the worm or cure the worm of me? Now if we can figure out which drug, and what it is... meanwhile, this source [6] from 1915 describes kharsivan and neo-kharsivan as substitutes for salvarsan and neosalvarsan; also 'novarsenobenzol "Billon"' was described as one; I suppose that's the same as the one above? Apparently this drug could also cause malaria [7] says that bags of the drug were hung above 8 patients, and somehow the parasites were able to transfer themselves - though I can't help but think it must have been carelessness with the needles or tubes rather than them literally swimming up the tube, as funny as a picture that might be. Looking further ... novarsenobillon is definitely a brand name for neoarsphenamine,[8] which we have as equivalent to neosalvarsan, agreeing with [9]. Tentatively I think I'll stencil that in as a redirect. Meanwhile, "liquor arsenicalis" is apparently potassium arsenite ... with a touch of lavender fragrance (not making this up!) [10] Wnt (talk) 17:02, 26 May 2016 (UTC)

Napoleon Bonaparte had plenty of arsenic in his body when he died. --AboutFace 22 (talk) 16:35, 26 May 2016 (UTC)

Is Arsenic an Aphrodisiac?: The Sociochemistry of an Element by William R. Cullen describes arsenic addiction and its symptoms in literature (p. 13) as well as people eating arsenic powder for any number of other reasons. Alansplodge (talk) 18:24, 26 May 2016 (UTC)

I had to reenter the search to get Google to show the page, but it did. However, what it says is that "trademarks of arsenic addiction" included horny calluses and white streaks in the nails. I believe that "arsenic addiction" in this context means the habit of taking arsenic but not an irrational mental urge to keep taking it for pleasure. The rest of the page kind of emphasizes these were being taken as treatments for syphilis and malaria, not for fun. If people thought it might have other health benefits it is no wonder, it being the treatment of choice for several diseases; certainly they had no deep sense of cumulative toxicity or carcinogenesis! Our wikt:addict doesn't really explain it well, but note their examples like "They addict themselves to the civil law." I think the modern sense follows international concern about opium addiction in the early 1900s and doesn't really apply to things written in 1800; "addict" should be read more like 'fan' or 'devotee', I think. Wnt (talk) 21:12, 26 May 2016 (UTC)

As a counterpoint I offer internet addiction and shopping addiction, I believe the technically correct terms now do not include the word "addiction" but, it's basically arguing semantics, there is little "in effect" difference between an "addiction" and a "compulsive behavior disorder". Vespine (talk) 06:05, 27 May 2016 (UTC)

• Arsenic is used to cut cocaine. μηδείς (talk) 06:04, 27 May 2016 (UTC)

Looks like a rare thing, with no obvious point to it unless the dealer wants to kill his customer. Wnt (talk) 11:42, 27 May 2016 (UTC)

Interesting stuff, thanks everyone! Based on what I've read I think it is likely that he started by using something like Fowler's solution, and later he was actually eating arsenic in powdered form. [11] [12] [13] [14]. Knowing what we know now this is completely bizarre, but I guess people back then were unaware (or less aware) of how dangerous arsenic is. The Quixotic Potato (talk) 11:34, 27 May 2016 (UTC)

Well, given that people usually first encountered the drug due to a disease like syphilis or malaria that could be chronic, it is hard to separate the legitimate medicinal effect from beliefs that it increased vigor in some other way. Your third source makes it sound almost as if the man had Munchausen syndrome; it's hard to say what effects were in the mind, especially when people were under the grim shadow of the gruesome effects of syphilis. I can't rule out the possibility of genuinely positive-seeming effects from it though - after all, who is going to take arsenic for a careful modern study of its chronic effects. Homeostasis almost would be expected to adjust some metabolic parameters to work better with a constant level of arsenic over time, but who really knows? Conceptually, arsenic is not evolved as a toxin, and humans have not evolved to be poisoned by it. It is more like a random mutation whose intensity is determined by the dose. In genetics, small mutations are expected to be 50% beneficial - the analogy given is that if you are shooting an arrow and someone moves the target a millimeter, like as not the move will put you closer to the bullseye. Large mutations become increasingly likely to be harmful; nonetheless, things can be pleasant and still harmful. Unless a large population of arsenic fiends turns up for study, I don't think we're going to get a compelling scientific answer about the "addiction", but can only note, per that source, that some people believed it had beneficial effect on them. Wnt (talk) 12:11, 27 May 2016 (UTC)

My other favorite encyclopedia has some interesting things to say about arsenic. Note particularly the bottom panel: "for centuries arsenic was taken by peasants in the Austrian Alps to improve breathing. Their bodies grew to tolerate the poison, which they consumed in normally fatal doses." (I've wondered about this specific James Maybrick question before, too.)  Card Zero  (talk) 23:52, 27 May 2016 (UTC)

@Card Zero: This site also says <bogus!>"Arsenic's popularity as a poison was owed to the difficulty of detecting it in a corpse. Death occurs after the liver has removed most of the poison from the blood and transformed it to naturally occurring chemicals, providing a minimum lethal dose is given."</bogus!> I don't know what kind of a game they're running there but either they suffer from the Philosopher's Gallstone or they're seriously unreliable. Wnt (talk) 10:39, 28 May 2016 (UTC)

That said, this actually holds up, at least that the claim exists. From [15] references "the famed arsenic eaters of Styria", and "arsenic prophylaxis of the fakirs" as a bonus. The former is cited to KH Most, "Arsenic as a poison and charm in German folk medicine with particular respect to Styria." Santa Barbara, CA: Scitran, 1939... Wnt (talk) 10:47, 28 May 2016 (UTC)

Another source [16] "The Arsenic-Easters of Styria", Scientific American, October 2, 1869 describes people consuming large quantities of arsenic (0.56-0.62 grains 2 or 3 times a week) and claims to have observed this and detected arsenic in the urine, but not in the quantity taken. Their conclusion was that much of it was insoluble and excreted in the feces - whether that's the actual pharmacodynamics or something deceptive by the druggists, I'm not ready to conclude, but they seemed convinced. They described it being given to horses to make them fatter and more "courageous", which echoes a statement about feeding it to pigs in your source above. Wnt (talk) 10:59, 28 May 2016 (UTC)

Hmmm, this rabbit hole goes on forever. From arsenic I happened across Scheele's Green, which once was used to color cakes and candies with arsenic. Apparently arsenic was extremely cheap as a mining byproduct - you could buy enough to kill 50 people for the price of a cup of tea according to [17] - and various myths about it abounded, such as it serving as a "Victorian Viagra", as that source puts it. I can't help but wonder whether this was an old problem with correlation and causation, as people using arsenic for a certain infection would doubtless be less worried about what they might catch later on. Hell, for all I know it was a steampunk version of PrEP... anyway, in Greenock the poisoned candy was particularly notorious, so much so that green candy was stigmatized ever after in Scotland. I have no idea, but I suspect, that a modern urban legend about green M&M's might be a manifestation of this? Wnt (talk) 16:15, 28 May 2016 (UTC)

Can anyone identify what this structure is?

Can anyone identify what this structure is? I stumbled upon it whilst running in rural Sussex, deep within woodland.

--AloeTougha (talk) 17:28, 26 May 2016 (UTC)

Maybe a foundation_(engineering) of an old mill? For reference, this [18] is a picture of an old mill foundation. Is it on a creek? It looks like it might be at the top of a small rise/flood plain, which is a pretty good place for a mill, and add further circumstantial evidence to the mill hypothesis. SemanticMantis (talk) 17:42, 26 May 2016 (UTC) (PS I moved your pictures the right so they won't interfere with indentation.)

Seems a bit too small for a water mill, this one is a rather small example, most are this size or even bigger. My thought was perhaps a disused railway embankment and it was the foundations for one of those corrugated iron trackside sheds that you used to see. There seems to be a concrete sill inside the brickwork which would support the iron sheets a bit like this one. Alansplodge (talk) 18:00, 26 May 2016 (UTC)

Doesn't seem very deep in woodland; there's some sort of structure nearby in photo 3. Maybe the foundations of a public toilet, workmen's hut, or possibly a double grave. Akld guy (talk) 19:11, 26 May 2016 (UTC)

A foundation of some kind... Difficult to say precisely. Even difficult to date, but I'd guess 1920–1960. I could easely be wrong though. Unused for at least 20 years. Could be wrong about that too.

Maybe some sort of pumping station? You can find these pretty much everywhere. Or an observation tower/watchtower. Those have been build for entertainment throughout the century and in my country (Netherlands) military watchtowers have been build even in the 1950s. Or it could have been a transmitter tower. Old railway lines in rural areas are generally easy to find on aerial photographs, so that would be easy to check if you have an exact location.

Do I see some kind of observation tower in the background of that final image? Maybe they just moved it.

Maybe you can find some clues on old topographic maps. In some countries these are freely available on the web; I don't know about Britain. Or you can try and contact a local historical society. PiusImpavidus (talk) 19:32, 26 May 2016 (UTC)

Notice there is an internal lower brick course. This doesn’t seem to suggest floor boards as there is no threshold gap in the higher external course. More likely a support for a cover. Also, the lower brickwork has been concrete rendered. Now why would one go to that trouble unless one wanted to (say) avoid something acidic leaking out an dissolving the lime mortar? Therefore, it looks to me like an ordinary common cesspit. Similar to a septic tank but the liquor just drains out the bottom. Do a simple Bray analyses of the soil within. If it show extremely high for phosphorus (nitrogen will be very high as well) then you can be reasonably certain it is a cesspit.--Aspro (talk) 19:54, 26 May 2016 (UTC)

"Cesspit" sounds so terribly undramatic. One needs to bring a handheld videocamera to the site, framing one's face with it at a strange angle in wavering hands, and explain the concrete rendering and acidic leakage in an ominous tone thusly: "We believe this masonry was not laid with the intention of keeping something out. It was designed to keep something in." :) Wnt (talk) 12:15, 27 May 2016 (UTC)

"We walked nearer, and I saw that it must be five feet across, with walls a good foot thick and some six inches above the ground level — solid work of the seventeenth century, or I was much mistaken. That, Pickman said, was the kind of thing he had been talking about — an aperture of the network of tunnels that used to undermine the hill." InedibleHulk (talk) 16:26, 27 May 2016 (UTC)

Try old-maps.co.uk --Shantavira|^{feed me} 20:31, 26 May 2016 (UTC)

If you could give an accurate Ordnance Survey Grid Reference, I can check it against current maps (of which my Office has a complete set, and possibly some older issues (of which I have a number at home), to see what might have been there. My money would be on some old military site – as an Army Brat, I know there are a lot more around than people generally realise. {The poster formerly known as 87.81.2301.95} 185.74.232.130 (talk) 21:44, 26 May 2016 (UTC)

Thanks for your answers everyone. I have looked at a few old OS maps from the early 1900's but there is nothing there. Any help will be appreciated. The co-ordinates are: 51° 6' 27.20" N 0° 18' 49.62" W --AloeTougha (talk) 21:54, 26 May 2016 (UTC)

When in doubt, grab a pickaxe and shovel. Unless you're in doubt about the landowners, of course. Some things are best left buried. Might seem obvious, but you could ask the landowners, if you can find them. I'd guess there's some kind of tube down there, but I'm no engineer. InedibleHulk (talk) 22:25, 26 May 2016 (UTC)

Clickable coords: 51°6′27.20″N 0°18′49.62″W --Tagishsimon (talk) 13:18, 27 May 2016 (UTC)

The suggestion that it's something related to a railway can probably be ruled out. According to the Railway Atlas: Then and Now (2012, ISBN 978-0-7110-3695-6) by Paul Smith and Keith Turner, in 1923 there was no other railway between the two railway lines running north and northeast from Horsham that still exist today. This doesn't prove that there never was one, but if there wasn't one in 1923 it's pretty likely that there never was one. (If there was, the applicable volume of A Regional History of the Railways of Great Britain ought to say.)

As an outright conjecture, I wonder if it could be something to do with Gatwick Airport, like some sort of obsolete navigation beacon. --69.159.60.83 (talk) 05:42, 27 May 2016 (UTC)

• I'll point out that there's no old-growth forest around the structure at all. At least 90% of the trees seem less than half a century old, although that's a wild guess, as growth rates in Britain may be slower than in the American NE. In any case, this structure was most likely in an open field when built. Notice there is also no evidence of rubble or charring. μηδείς (talk) 05:58, 27 May 2016 (UTC)

From those co-ordinates, it's only about 200m north of the edge of a substantial (~1km diameter) clay pit working, and might therefore be a remnant of older structures associated with that site. {The poster formerly known as 87.81.230.195} 185.74.232.130 (talk) 13:36, 27 May 2016 (UTC)

The location is just a stone's throw away from, and possibly on land belonging to, Langshurst House which is a barbed-wire enclosed UK government security centre that develops technology for things such as "Mitigating the Threat of Hostile Drones". It might be a WW2 related structure, such as a CH radar beacon. Note: the RAF's offer of a prize to anyone who can demonstrate a death ray that can kill a sheep at 100 yards is still unclaimed. AllBestFaith (talk) 16:42, 27 May 2016 (UTC)

Excellent. InedibleHulk (talk) 18:19, 27 May 2016 (UTC)

Looking on the satellite photo at max magnification on Google, I see that there is some pattern of discoloration in the forest to the southeast, sort of looks like the word "Tío" to my eyes, reading toward the southeast. Is that the clay working? Do you think there might be other ruins visible there? Wnt (talk) 18:51, 27 May 2016 (UTC)

It still looks like a ordinary cesspit to me. If in doubt, get more data. Even a simple home-made probing rod should find the inlet pipe on the side with the higher ground level. Soil analysis will remove all doubt.--Aspro (talk) 16:40, 29 May 2016 (UTC)

δ34S

What does the delta stand for in δ34S? It must be «δύο» the Greek word for the number two but it's not mentioned.2A02:582:C62:9B00:2095:AF4A:971F:2F9D (talk) 19:25, 26 May 2016 (UTC)

It doesn't stand for anything. It doesn't stand for anything special in this case. It derives from the long tradition of using the character delta to indicate differences and changes in mathematics. 'Delta is the initial letter of the Greek word διαφορά diaphorá, "difference" ', from Delta_(letter)#Lower_case, where it is specifically mentioned as being used for Environmental_isotopes. See also Isotope_geochemistry and e.g. δ13C for analogous use. This probably could be better explained in the article you linked, I may fix it up a little. SemanticMantis (talk) 19:38, 26 May 2016 (UTC)

Ok I added this bit "The delta character is used by convention, derived from use in stable isotope chemistry." SemanticMantis (talk) 19:46, 26 May 2016 (UTC)

how to calculate resistor values in this circuit?

I sorta-kinda "get" how it works (the lack of a direct connection of Q9 and Q10 to Vcc threw me a bit off initially as well, but the circuit makes sense once one realizes that the emitter-base junctions are "just" diodes), but how to calculate R1 and R2, given Vcc, beta etc? Or are they current-limiting resistors whose value is not that critical? Also, what should R3 and R4 ideally be? Asmrulz (talk) 21:26, 26 May 2016 (UTC)

The motor driver circuit is well discussed at the Stack Exchange link you gave. The design for a battery operated toy aims at economy and operation on as low battery state as possible. The transistor pair Q10/Q12 (or Q9/Q11) is a "Darlington pair" of PNP transistors used in an H bridge. (The two Wikipedia articles are relevant though they focus on NPN and MOSFET transistors respectively.) The transistors are driven as saturated switches so their current gains (beta) are low. My quick calculation is that R1 (or R2) limit current to (5 - .7 - .7 -.7)/51 = 57 mA. We don't have the motor resistance (about 12 ohm?) or the input voltages but assuming Q10 has beta>10 we might use R3=R4= 100 ohm. That incurs less than 0.6V voltage drop. It may be possible to omit R3 and R4. The reliability of the simple circuit depends on its switching behaviour where there could be transient high currents if a transistor in the H bridge turns off slower than another turns on, or if the input signals overlap, and on tolerating transient high voltage from the inductive part of the motor load. AllBestFaith (talk) 09:30, 27 May 2016 (UTC)

Great explanation! If I have this right, the two alternate modes of the H bridge (which determine which way the current runs through the motor) are ensured by wiring the relevant switches, i.e. transistors together so that the base currents from the pnp and npn feed into each other. The "mysterious" part here is that it's easy to think of a transistor base voltage/current as an input, but in this case, it is determined by the collector and emitter voltages on each! I still don't understand exactly how that is determined. But it is clear enough that the resistors limit how high it can go, and therefore, how open the switches can be. The transistors to left and right then can be thought of as variable resistors that impose an extra resistance on top of that, depending on the control settings, to close the switches if current is absent... Wnt (talk) 11:33, 27 May 2016 (UTC)

I don't see how Q10/Q12 and Q9/Q11 constitute Darlington pairs.... Even so, I think the purpose of Q10 and Q9 is to simply short the bases of the diagonally opposite transistors together. The circuit works with a small-ish resistor in place of the C-E path of Q10 and Q9 (one at a time, of course), and likewise when Q10/Q9 are NPNs or MOSFETs (2N7000 in my case.) Asmrulz (talk) 17:56, 27 May 2016 (UTC)

A Darlington pair consists of two bipolar transistors connected so the emitter current of the first transistor (Q10) is the base current of the second transistor (Q12). The overall current gain I_{coll Q12} / I_{base Q10} is the product of the betas of the transistors. This is true whether their collectors are connected as in a simple Darlington or have different loads. In the actual circuit, Q10 collector load is Q14 base-emitter and Q12 collector load is the motor. It's also correct to say that Q10 serves to join the base currents of Q12 and Q14. I agree that Q10 and Q9 could be changed to NPN transistors which also reverse the motor direction. Here is the data sheet for your 2N7000 transistor that you should note has a maximum continuous current rating of only 200 mA. AllBestFaith (talk) 20:40, 27 May 2016 (UTC)

I suggest You to redraw the while circuit where the R1, R2 and base pins of the 4 transistors of the H-bridge are in the middle and the only the motor pins A and B are outside the H-bridge. Q9 and Q10 are inside. Then You get understand how it works. Or simple, image R1 and R2 are not on top of the sheet. As the circuit dropping now is a relational map of the voltages levels and we can solve it easily with this better understanding:

R1 limits base current of Q12 and Q14, which is controlled by Q10

R2 limits base current of Q11 and Q13, which is controlled by Q9

R3 limits the base current of Q10 which is taken away from the base current of Q12. As these transistors are used in switching mode for a motor controller, You need to to control the transistors behave like a closed switch, fully opened for current. A transistor as a voltage drop on is base emitter diode. Careful if the transistors are darlington type in a single case. Refer the data sheet. The maximunm current needs to be switched at the minimum and the maximum of Vcc. Q9 and Q10 are behind R1 and R2 to be more independent from Vcc, but if A or B is less than Vcc, Q11 and Q12 are not shootdown. It is on Q13 and Q14 to shootdown. Here might be an error in the circuit design. R1 sould be swapped with R1 and R2 sould be swapped with Q9.

R3 and R4 are same value.

R1 and R2 are same value.

R3 needs to be dimensioned to control Q14, controlled by Q10 to maximum current are lowest amplification of worst sample of Q14 and Q10 each, but still does support Q12. Refer the data sheet for minimum and maximum amplification. Take the voltage drop of the transistors: Now see R1 and Vcc. Subtract the voltage drop of the whole transistors:

U_R1=(Vcc-(Q12_EB+Q10_EC+Q14_BE))

In the circuit of the link:

U_R3=Q10_EC+Q14_BE-Q10_EB

When swapping Q9 and Q10 with their preresistors R1 and R2 each as I suggested:

U_R3=(Vcc-(Q12_EB+Q10_EB))

Note A and B are controlled by a semiconductor, which might be not reach ground level of voltage.

When Vcc drops, the current of the motor also drops. So it is easy to get minimum and maximum values.

Now apply ohms law to calculate the resistors. --Hans Haase (有问题吗) 09:18, 28 May 2016 (UTC)

Redraw like this? Asmrulz (talk) 20:58, 28 May 2016 (UTC)

Yes, exactly, Asmrulz! --Hans Haase (有问题吗) 10:26, 30 May 2016 (UTC)

When Q9 and Q10 would be NPN type transistors, I suggest to keep them at this place, but with it's E and C raplaced. As they are PNP type, they should be replaced with it's preesistor each. As the H-bridge controls a motor which is an inductive load, Q11 to Q14 require to have a blocked diode in parallel to rectify the inductive bursts of the motor coils. A and B also must not idle and required to be tied by additional resistors to close completely when tured off. --Hans Haase (有问题吗) 11:17, 30 May 2016 (UTC)