Wikipedia:Reference desk/Archives/Science/2017 May 22
Science desk | ||
---|---|---|
< May 21 | << Apr | May | Jun >> | May 23 > |
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 22
[edit]Brush for painting walls
[edit]What kind of brush was used before the invention of the Paint roller and what was it's name? In Romania, the paint roller was not used before the fall of communism (1990). The painters used a big and thick brush called bidinea - the word comes from the Turkish badana. It looked pretty much like an American Wall-paper brush. I wonder if Wikipedias in other languages have an article for this item. — Ark25 (talk) 01:05, 22 May 2017 (UTC)
- Paintbrush#Decorators' brushes has some information, and a picture. Rojomoke (talk) 04:01, 22 May 2017 (UTC)
- I still use them (for better coverage on rough surfaces), but just call them large paintbrushes. Dbfirs 08:52, 22 May 2017 (UTC)
- They're still used today, made much the same as they always were. A wallpaper paste brush is easier to find and very close in style (if of the traditional style).
- Masonry walls are alkaline, which will damage many paints. So "wall paint" has historically been alkaline itself, either "whitewash" or "distemper" (a very thin lime wash). This was applied with a distemper brush, a large brush itself resistant to alkali, so it was tied together mechanically, rather than the bristles glued in. Owing to their size (often 6" wide) they were usually a hollow core, i.e. the bristles were placed on two sides of a wooden centre, with two wood sides nailed or riveted on (rivets stay tight longer in wood that's being continually wetted). As these crude washes will never give a particularly smooth finish, they never needed a very high quality brush.
- The manufacture of paint brushes isn't a simple "one style makes all", and some large or specialised brushes get very expensive. Japanese paper-making brushes can be a few hundred $ each, mostly because they're so rare and are handmade. French oval varnish brushes became unavailable a few years ago, when the last machine that made the ferrules was sold. Andy Dingley (talk) 10:43, 22 May 2017 (UTC)
- Here's one type of distemper brush, a WWII ex-RAF one: https://hatchfive.wordpress.com/2015/06/
- This one is (as described above) mechanically bound, by the bristles being string-tied bundles. As such bundles are always round (or slightly oval), a wide brush is made by putting pairs of bundles onto the handle. Andy Dingley (talk) 21:46, 23 May 2017 (UTC)
- Thanks for all the answers. — Ark25 (talk) 15:18, 24 May 2017 (UTC)
- See here for some examples of various brushes. Iapetus (talk) 17:07, 24 May 2017 (UTC)
Eliminating infectious diseases by bioengineering large numbers of brainless humans with weakened immune systems?
[edit]"How do the principles of virulence help to explain the enigma of invertebrate immunity? One answer is that invertebrates don’t need an acquired immune system because they never had it. The parasitic agents of invertebrates have not coevolved with acquired immunity so their virulence is calibrated to the coevolved innate immune system. The proposal here is that contrary to widely held views of practicing immunologists, the immune system is not evolutionarily selected to prevent infection in an absolute sense. Rather, it is selected to make one individual slightly more resistant or at least different than others of the same or related species. The adversary of any individual is not really the world of parasites, they are truly undefeatable, it is his or her neighbor. A zebra doesn’t have to outrun the lion, just the slowest member of the herd."
This then suggest that we could benefit from creating vast numbers of biological systems that can be infected by the same microbes that infect us, but with weaker immune systems. On the long run this would cause the microbes to evolve to infect these systems instead of us. Or you could say that since the relevant competition is between us, we put ourselves collectively in the winning position by creating a large number of losers. This will then work because we're then going to keep the losers alive artificially. For ethical reasons one would then be led to creating brainless humans, or just large volumes of human tissues that can be kept artificially alive in some way. Count Iblis (talk) 01:30, 22 May 2017 (UTC)
- No, this logic is flawed. It would only work if regular humans no longer interacted with each other, but only with the brainless easy prey-humans, in which case you don't even need them, just keep the regular humans isolated from each other. Anywhere a large pool of regular humans existed, microbes would still evolve to take advantage of them. Those microbes have no way of "knowing" that easier prey exist elsewhere, and no way to get there, in any case, so they will continue to evolve to take advantage of their current hosts. StuRat (talk) 01:36, 22 May 2017 (UTC)
- I don't think StuRat's argument is valid, so long as the prey-humans have enough contact with real humans to transmit disease to them. The prey-humans would fulfill a role broadly similar to that of plants in crop refuges, which someone should probably blueify. See eg [1] and [2]. Adrian J. Hunter(talk•contribs) 10:58, 22 May 2017 (UTC)
- As some day it may happen that a victim must be found, I've got a little list... The OP's biological application of Chamberlain's Appeasement method against microbes should ensure "Health For Our Time", what could possibly go wrong? But it really doesn't matter whom you put upon the list, For they'd none of 'em be missed. SdrawkcaB99 (talk) 11:06, 22 May 2017 (UTC)
I can think of a more practical context for a related idea in the context of antiviral drugs for influenza. Drugs like oseltamivir can result in resistance, the prevalence of which is limited by the larger number of people who don't have immediate resort to a doctor to wave the magic pen unlocking their access to the drug, who presumably serve as a sort of susceptible reserve population like the GMO-free band around a Roundup Ready cornfield. Stockpiling antiviral medications for pandemic influenza implies such decisions with the limited stockpiles. But I haven't actually looked into what has been worked out about all this scientifically... Wnt (talk) 14:01, 22 May 2017 (UTC)
- Thinking a bit more about this.... According to the article one would explain the fact that young people rarely get diseases like pneumonia while it's not unusual for old people to get it, not by saying that young people we have such good immune systems but because old people bring down the average that the microbes have adapted to. If immunity in old people were to decline less, then infectious diseases would occur at similar rates as they do now but it would be more evenly divided between old and younger people. One may then postulate that the aging process could have been an outcome of natural selection. Not only would biology not have come up with a fix for the aging processes, it may actually have fine tuned it in such a way as to steer away diseases from the young in an optimal way. Count Iblis (talk) 01:02, 23 May 2017 (UTC)
Why epinephrine is discriminated from nor-epinephrine and dopamine by the definition of hormone?
[edit]Why epinephrine is discriminated from nor-epinephrine and dopamine by the definition of hormone? I've read the talk on the article "list of hormones" and I saw that the same answerer there, made two different comments... in one he said: "The catecholamines such as epinephrine, etc. are classified as neurotransmitters, not hormones." and later after someone comment on his answer he said "Epinephrine has significant distant effects from the adrenal glands where it is synthesized and therefore it is legitimately classified an hormone... The rest of the catecholamines such norepinephrine appear to act primarily as neurotransmitters and therefore have not been added to the table.". and I really didn't understand his discrimination between epinephrine, norepinephrine and dopamine. I would like to understand it. 93.126.88.30 (talk) 12:44, 22 May 2017 (UTC)
- Neurons send neurotransmitters at the synapse between the axon terminals and dendrites. Hormones may use the circulatory system or the extracellular fluids to deliver signals, near and far. Sometimes, the boundary between a hormone and neurotransmitter is not very clear. 140.254.70.33 (talk) 13:24, 22 May 2017 (UTC)
- Epinephrine and the other catecholamines are all made by neuroendocrine cells. The difference in whether a specific catecholamine is described as a neurotransmitter or a hormone is its primary function.
- Quoting from our article neuroendocrine cell,
"Neuroendocrine cells are cells that receive neuronal input (neurotransmitters released by nerve cells or neurosecretory cells) and, as a consequence of this input, release message molecules (hormones) to the blood."
- Neurotransmitters provoke the production of hormones - they are the body's signal to produce hormones and drive organs and other groups of cells into action. However, both general classes of biomolecules can act centrally on the brain and also in the neuroendocrine system distributed throughout the body. The extent to which this happens wasn't always understood when molecules were being classified as one or the other. "Neurotransmitters" tend to have a much shorter duration of action in the body than "hormones", which is probably what the author of that article was getting at. loupgarous (talk) 23:30, 23 May 2017 (UTC)
Histamine is considered as hormone or not?
[edit]Histamine is considered as hormone or not? In our article here (histamine) as well as in the list of hormones- no mention about its definition as hormone but on other places it's considered as an hormone as well. Wikipedia normally doesn't take a side but just share with the readers the different opinions objectively. For example in this encyclopedia it's considered as an hormone.93.126.88.30 (talk) 12:43, 22 May 2017 (UTC)
- If you look up hormone, there are two problematic aspects to the definition the article gives: hormones are secreted by glands, and they are transported by the circulatory system. This is a typical definition but in biology definitions tend to creep. Are individual immune cells secreting histamine from wherever part of a "gland"? Well, is a gland defined by a visible morphology, or is anything that secretes a hormone a gland, making the definition tautological? Then there is the circulatory system aspect. Localized inflammatory reaction does involve the circulatory system. But I think usually when people think of a "real hormone", they mean something that you can take out a vial of blood from whatever site is most convenient to measure it, even if the nurse has to pull out a junkie's penis hunting for a suitable vein. But histamine's effect is, by design, local, and the overall blood level, if there is one, likely doesn't represent the intensity of its effect where it matters. There are some relevant concepts that come up if you look at endocrine, exocrine, paracrine, apocrine etc. Wnt (talk) 14:11, 22 May 2017 (UTC)
- Thank you for your answer, but unfortunately I really don't understand why acting of substance on cells near to secreting cell (this is the definition of paracrine hormone) or acting on the secreting cell it self (this is the definition of aoutocrine hormone) avoid this substance to be called an hormone.
- Regarding to the question: "Are individual immune cells secreting histamine from wherever part of a "gland"? Well, is a gland defined by a visible morphology, or is anything that secretes a hormone a gland, making the definition tautological?" before we discuss about this question we have to agree firstly what hormone is and to know what the definition of gland is. And as we know glands can be exocrine and endocrine while according to what I've been told today, just those who secreted by the endocrine are considered hormones. Then not any gland secretes hormones... 93.126.88.30 (talk) 17:20, 25 May 2017 (UTC)
- Based on encyclopedia.com's ridiculous article about homeopathy I wouldn't trust it for anything.
- I agree with Wnt that a hormone, as I understand the term, is a molecule that works by endocrine signalling, meaning it's transported throughout the body rather than acting locally (paracrine signalling). Histamine's best-known effects are local. But histamine has many effects, some of them apparently systemic (eg), so perhaps it acts as a true hormone in addition to its better-known effects. Also, even peer-reviewed articles sometimes speak of "paracrine hormones" (pubmed), so there's inconsistency (or at least sloppiness) in how people use the term "hormone". Adrian J. Hunter(talk•contribs) 14:25, 22 May 2017 (UTC)
How can gunpowder propel something over Mach 1?
[edit]If it doesn't burn faster than the speed of sound? Sagittarian Milky Way (talk) 15:32, 22 May 2017 (UTC)
- I think the question is misguided. Gunpowder does indeed burn "slowly", i.e. it makes no detonation, but that defines how fast the combustion front can move, which has no direct relation to the speed of a projectile. The latter is propelled, not by the combustion front itself, but by the heated gases trapped (e.g. in a gun's rifling), hence Jayron32's taciturn answer.
- However, there is still a nontrivial question: as the projectile accelerates and reaches the speed of sound, the heated gases are no longer at pressure equilibrium. Said otherwise, the projectile will "outrun" the pressure that is trying to accelerate it, and hence the acceleration should stop at the speed of sound (actually, a bit slower). But then it is the speed of sound in the heated gases, higher than at room temperature (the projectile will move in the air). TigraanClick here to contact me 15:45, 22 May 2017 (UTC)
- Oh right. Duh. Sagittarian Milky Way (talk) 16:05, 22 May 2017 (UTC)
- The important thing is that there is no law of conservation of velocity, but there is a law of conservation of momentum... --Jayron32 16:19, 22 May 2017 (UTC)
- There is no law on how many percent of the momentum goes to the bullet, though. If a rifle's illegally sawed off to a 2 millimeter barrel the bullet will not leave as fast.[citation needed] Sagittarian Milky Way (talk) 16:55, 22 May 2017 (UTC)
- Again, I have no idea how to respond when you write such confusing statements and non-sequiturs. It makes it difficult to help you find answers to your questions. --Jayron32 17:01, 22 May 2017 (UTC)
- What he wrote is pretty straightforward - if we suppose that the expanding gas were to stop at a certain velocity, then the bullet would carry only a fraction of the momentum and the remainder would stay with the gas as it leaves the gun later on. Really, F=ma seems more of a non sequitur since the whole issue is whether the gas can keep up with the bullet to apply any force to it past a certain point.
- Again, I have no idea how to respond when you write such confusing statements and non-sequiturs. It makes it difficult to help you find answers to your questions. --Jayron32 17:01, 22 May 2017 (UTC)
- There is no law on how many percent of the momentum goes to the bullet, though. If a rifle's illegally sawed off to a 2 millimeter barrel the bullet will not leave as fast.[citation needed] Sagittarian Milky Way (talk) 16:55, 22 May 2017 (UTC)
- The important thing is that there is no law of conservation of velocity, but there is a law of conservation of momentum... --Jayron32 16:19, 22 May 2017 (UTC)
- Oh right. Duh. Sagittarian Milky Way (talk) 16:05, 22 May 2017 (UTC)
- On second thought, I find myself wondering as I think about that is whether "speed of sound" is relevant to the expanding gasses in the gun barrel. I mean yes, it can limit the speed of disturbances in air, but those gasses aren't moving through air, they are the medium. If the gun barrel were perfectly smooth, then we can think of the gas right behind the bullet as stationary in its own frame of reference! Maybe what we need to be looking at is how a bullet or any other kind of fast-moving piston can move down a bore faster than the speed of sound in the air in that bore. How does the air get out of the way faster than the speed of sound? (It's time for me to look up shock wave and start from scratch, because I don't know this!) Wnt (talk) 18:18, 22 May 2017 (UTC)
- Air is highly compressible, so can just compress until the end of the barrel is reached, then move to the sides and decompress. Of course, this will create a bang. StuRat (talk) 19:19, 22 May 2017 (UTC)
- Which bang is known as the bullet crack. It is also called a bullet bow shockwave; but in article one thing I do not get is when and where a witness will not be able to hear the cracking sound produced by the bullet. --Askedonty (talk) 19:38, 22 May 2017 (UTC)
- Air is highly compressible, so can just compress until the end of the barrel is reached, then move to the sides and decompress. Of course, this will create a bang. StuRat (talk) 19:19, 22 May 2017 (UTC)
- Infact is seems Mach 6-9 is typical for explosions. See article Detonation velocity. --Kharon (talk) 17:02, 22 May 2017 (UTC)
- @Kharon: Not all explosions are detonations. See also high explosive, which gunpowder is not. TigraanClick here to contact me 09:35, 23 May 2017 (UTC)
- Infact is seems Mach 6-9 is typical for explosions. See article Detonation velocity. --Kharon (talk) 17:02, 22 May 2017 (UTC)
- The essential item in this discussion is the law of conservation of momentum. Confined in the gun barrel, the pressure behind the bullet as the gunpowder (smokeless powder, for high-velocity rounds, actually) charge imparts momentum to the bullet from the expanding gases, which at the point the bullet leaves the bore of the gun barrel is what propels the bullet forward. A mathematical summation of the forces acting on that bullet would show the cumulative force on that bullet over the amount of time it takes for the gases to push the bullet out of the barrel, plus the wavefront of the escaping gases pushing the bullet afterward.
- The difficulty the OP had was imagining the bullet being pushed by an unconfined wavefront of expanding gases, outside the barrel, when most of the velocity and momentum imparted to the bullet are transferred during a time interval when the bullet is being pushed through the barrel, some of the gases's energy is dissipated as heat from friction between the barrel bore, the lands of the rifling, and the bullet, heat from the powder's combustion inside the barrel but mostly the linear transfer of energy from the expanding gases against the barrel and the bullet.
- As the bullet "gives" and the barrel doesn't, most of that pressure moves and accelerates the bullet for that fraction of a second it's in the barrel to make it travel faster than sound. The momentum imparted to the bullet's mass is where most of the force of the exploding powder and its violently expanding gases goes while the bullet is in the barrel.
- Since it is imparted over a period of time, more of that momentum's pushing the bullet away from the explosion than it would be if the bullet were just placed at the edge of a charge of gunpowder detonated outside a gun barrel - the barrel "focuses" the momentum mostly toward the bullet, so it moves faster than the molecules of the expanding gas. loupgarous (talk) 00:13, 24 May 2017 (UTC)
- Even more impressive is that there are various mixes of smokeless powder that have different burn rates. The amount of powder a case can hold, the size and weight of the bullet, along with the pressure rating and length of barrel are engineered to provide the best ballistics possible. A gurn rate that quickly reaches maximum pressure and a burn rate that sustains that pressure as the bullet allows the volume to expand provides the best return. --DHeyward (talk) 08:31, 24 May 2017 (UTC)
What kind of voltage regulator is this?
[edit]The voltage regulator article discusses many different kinds of voltage regulator. What [kind of regulator is this] so that I can read about that one specifically? Does it turn all the unwanted energy to heat? So 1 A @ 5 V from a 12 V supply would give 7 W heat? Thanks! ----Seans Potato Business 22:25, 22 May 2017 (UTC)
- The the specs say it includes "thermal shut-down control" does imply that it's converting excess energy into heat. (I find that they rarely say so outright, as this is obviously an inefficient and non-ideal way to do things.) StuRat (talk) 23:06, 22 May 2017 (UTC)
- I think the way to approach this question is to begin with simple conservation of energy, as we do when first considering power transformers, gear trains etc. If the output is 1 A at 5 V, that is a power of 5 W. Assuming simple conservation of electrical energy, the input would be 12 V and 5/12 A (or 0.417 A). Power transformers and gear trains operate with a certain amount of the input energy leaving the system as unwanted heat, and it would be the same with this voltage regulator. Assuming 20% of the input energy is lost to heating of the surroundings leads to the conclusion that the input current will be 1.2 times 0.417 A (or 0.5 A). Input power will be 6 W and output power will be 5 W, so energy will be lost to heating of the surroundings at the rate of 1 W. Dolphin (t) 23:35, 22 May 2017 (UTC)
- (edit conflict)::See 78xx. User:StuRat is correct in that the 78xx series dissipates excess power as heat, but his reasoning is a bit off. For example, switching regulators (which do not work by dissipating excess power as heat) often include thermal shutdown capability. Note any voltage regulator that is less than 100% efficient will dissipate some energy as heat and therefore may be susceptible to thermal overload. Physics being what it is, all regulators are less than 100% efficient. Shock Brigade Harvester Boris (talk) 00:55, 23 May 2017 (UTC)
- I'm rather surprised that overheating is a concern for those. Of course, some heat will be generated by other methods, but it seems far less likely to be more than the device can handle, not being on the same order of that from a device where 100% of the unused energy goes to heat. StuRat (talk) 01:06, 23 May 2017 (UTC)
- Clicking on the datasheet option and looking at the block diagram, it's apparent (if you're used to looking at this sort of thing) that the series pass element is Q17 and R11 and most of the waste heat will be dissipated in these two elements. So for output of 5 V and 1 A, about 7 W will be dissipated in Q17 & R11. A small amount of power will be dissipated in the rest of the regulator. Jc3s5h (talk) 00:26, 23 May 2017 (UTC)
- The device is a linear series regulator and is mentioned here. Jc3s5h has answered correctly; the small amount of power additional to the 7W dissipated in Q17 & R11 is (Id x 12) W where Id is the quiescent current to ground, about .004 to .006 A according to the data sheet. Dolphin's answer incorrectly supposes conservation of energy but the device is only 41.5% efficient and wastes 58.6% of input energy as heat. Thermal protection is provided in case someone fails to provide the Heat sink that the device needs. Blooteuth (talk) 01:18, 23 May 2017 (UTC)
- You are looking for a 7805 linear voltage regulator in TO-220 package. The 1.5 amperes may be the shortcut peak current. The 78xx series have some special variants like 78L05 or 78S05. Theres a 100 mA variant in TO-92 package. There are 2 and 3 ampere variants avail. Note: Never clamp linear regulators in parallel. Such circuits cause output voltages on the input voltage level and oscillating output similar to a noise amplifier or function generator. Indeed the heat is the power drop in the regulator device 7 W = (12 - 5 V) * 1 A. This are 7 watts heat spread over a necessary heat sink. This series usually require 3 volts more input than output to generate a proper output voltage level. A socalled low drop regulator needs less volts to operate properly. The maximum input voltage is 37 volts. When the heat drop exceeds its maximum, have another circuit before to generate 8 volts for the 5 volts regulator. Depending on the application, a stepdown converter, also called buck converter which is a switching mode voltage regulator. It might be more energy efficient, but for some applications too noisy on the output. For this issue, have the linear regulator behind the buck converter. Switching mode voltage regulators can reach an energy efficiency up to 97 %. Another example are Class-D amplifiers, using an audio input as reference voltage. German Wikipedia has an overview of switching mode voltage regulators, see de:Schaltregler. Most linear regulators have 3 pins. A switching mode regulator needs an extra input pin to sense the output. Depending on the type a voltage divider can be installed there to make the circuit generate another output voltage. --Hans Haase (有问题吗) 17:39, 23 May 2017 (UTC)
- Dolphin's answer is completely wrong, and StuRat's is not helpful. Series regulators of the LM-7812 and LM-7805 type (which is the type the OP asked about) regulate the output voltage by wasting the excess. If you connect an LM-7805 to a 12 volt DC source, it will provide a regulated 5V output. Very little energy will be dissipated in the device when no load is placed on the output, since the device consumes only a few milliamps itself. When connected to a 12 volt supply and 1 Amp is being drawn by the load, the device is dropping 7 volts @ 1 Amp, which is indeed 7 Watts, answering the OP's question. The data sheet in the link does not mention the device's power dissipation, but these devices are typically rated at 5W, and at 7W it should therefore be mounted on a suitable heatsink. The alternative is to drop the input supply to 10V, when 5W will be dissipated. I don't recommend going lower, as these devices have a threshold of about 2.5 volts above their stated output, and operation becomes erratic when the supply drops below that threshold. Akld guy (talk) 20:40, 23 May 2017 (UTC)
- The words of the datasheet "If adequate heat sinking is provided..." require the designer to do a proper thermal calculation. For the TO-220 package, Table 2 Thermal Data indicates we need a heatsink with thermal resistance no more than Rthj-amb - Rthj-case = 50 - 5 = 45 °C/W. If the ambient temperature is 25° then we must not dissipate more than (150 - 25)/50 = 2.5 W because 150° is the absolute maximum junction temperature. Our need to dissipate 7W therefore demands a bigger heatsink, certainly one with no higher thermal resistance than 12.86 °C/W. However a responsible engineer would not settle for a design on the edge of self destruction and would lay in a healthy safety margin against eventualities such as increased ambient temperature and poor ventilation. His options include choosing a bigger heatsink with lower thermal resistance, choosing the 2-bolt TO-3 package for better heatsinking and reducing the device dissipation with an external dropping resistor as shown in the datasheet Fig. 32. (If the load happens to be a 5 ohm resistance and the supply is exactly 12V then a 7 ohm Resistor could substitute for the whole regulator.) Blooteuth (talk) 01:06, 24 May 2017 (UTC)
- Using a 7 ohms resistor does not support the full load of the whole circuit. As some linear regulators need 3 extra volts for the internal circuit, it is possible to use a 4 ohms resistor when the 12 volts are stable. The other idea is using a 7809 or 7808 to supply the 5 volts regulator, sharing the thermal load. --Hans Haase (有问题吗) 11:03, 24 May 2017 (UTC)
- @Hans Haase I don't think you have noticed that I only mentioned the 7 ohm resistor as a theoretical substitute (replacement) for the whole regulator. Of course it would not regulate the output voltage. Blooteuth (talk) 21:49, 24 May 2017 (UTC)
- Yes, thanks! Indeed, 1 ampere output at 12 volts input, the 7805 temporary behaves like 7 ohms resistor. The right way for thermal calculations assuming a normal operation, but just bear in mind other factors of real use like stable output and shortcut scenario, I pointed on. --Hans Haase (有问题吗) 09:44, 25 May 2017 (UTC)
- @Hans Haase I don't think you have noticed that I only mentioned the 7 ohm resistor as a theoretical substitute (replacement) for the whole regulator. Of course it would not regulate the output voltage. Blooteuth (talk) 21:49, 24 May 2017 (UTC)
- Using a 7 ohms resistor does not support the full load of the whole circuit. As some linear regulators need 3 extra volts for the internal circuit, it is possible to use a 4 ohms resistor when the 12 volts are stable. The other idea is using a 7809 or 7808 to supply the 5 volts regulator, sharing the thermal load. --Hans Haase (有问题吗) 11:03, 24 May 2017 (UTC)
- The words of the datasheet "If adequate heat sinking is provided..." require the designer to do a proper thermal calculation. For the TO-220 package, Table 2 Thermal Data indicates we need a heatsink with thermal resistance no more than Rthj-amb - Rthj-case = 50 - 5 = 45 °C/W. If the ambient temperature is 25° then we must not dissipate more than (150 - 25)/50 = 2.5 W because 150° is the absolute maximum junction temperature. Our need to dissipate 7W therefore demands a bigger heatsink, certainly one with no higher thermal resistance than 12.86 °C/W. However a responsible engineer would not settle for a design on the edge of self destruction and would lay in a healthy safety margin against eventualities such as increased ambient temperature and poor ventilation. His options include choosing a bigger heatsink with lower thermal resistance, choosing the 2-bolt TO-3 package for better heatsinking and reducing the device dissipation with an external dropping resistor as shown in the datasheet Fig. 32. (If the load happens to be a 5 ohm resistance and the supply is exactly 12V then a 7 ohm Resistor could substitute for the whole regulator.) Blooteuth (talk) 01:06, 24 May 2017 (UTC)
- In 1976, I built my IMSAI 8080, which was sold as a kit. It had a big unregulated linear power supply (built from a transformer that weighed about 10 pounds, a 50 amp full wave bridge, and two electrolytic capacitors about the size of coke cans) that supplied about 6 VDC to the backplane. Each board had linear regulators (7805s, as I recall). One of my initial boards (the SRAM board I think) was a real power hog. There was a 5 Watt resistor in parallel with the regulator, because the board's circuitry was known to demand enough 5 VDC to pull the voltage down that ceramic resistor was hot enough to burn your finger. The 7805 then provided enough additional current to maintain the voltage when the unregulated power supply sagged: it had a big heat sink. Those were the days.-Arch dude (talk) 23:27, 26 May 2017 (UTC)