Jump to content

Wikipedia:Reference desk/Archives/Science/2012 February 13

From Wikipedia, the free encyclopedia
Science desk
< February 12 << Jan | February | Mar >> February 14 >
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.


February 13

[edit]

Analytical mechanics

[edit]

In analytical mechanics, why are and treated as independent variables? 74.15.139.132 (talk) 03:59, 13 February 2012 (UTC)[reply]

If you think about a general object, it can have any location (x,y,z) and any velocity at that location . In the completely general case, the position and velocity can be thought of as completely independent. It is often useful to think about the positions and velocities to be, in principle, independent objects and then add in dynamical equations that describe their mutual evolution. Treating positions and velocities as independent entities in a larger phase space governed by mutually-interacting constraint equations is simply the conceptual framework used to build analytical mechanics. I realize it can seem a little weird to treat and as independent for the purposes of a calculation, but ultimately the underlying constraints of analytical mechanics will ensure that the end results are connected to each other in exactly the way that you would expect. Dragons flight (talk) 11:15, 13 February 2012 (UTC)[reply]
In fact, expecting that the value of a variable should be related to its derivative is assuming the existence of some differential equation. By treating and separately, we make no assumption whatsoever about the basic laws of the universe. It is soon realized that the minimization of action is a universally valid principle (in other words, the solution to the Euler–Lagrange equation represents the real world). From this statement alone, we can derive conservation of energy, conservation of momentum, the principle of inertia, the relation between force and energy (rather, the definition of force in terms of energy); and so on. The underpinnings of the use of generalized coordinates are covered in Marion & Thornton's Classical Dynamics, if I recall... in Chapter 4. It all comes down to this: in classical physics, we try very hard to begin our analysis with no assumptions whatsoever about the universe. Using the minimum number of experimental facts, we try to deduce and quantify relations that describe the behaviors of physical systems. Nimur (talk) 20:47, 13 February 2012 (UTC)[reply]

Thanks everyone. 74.15.139.132 (talk) 20:44, 14 February 2012 (UTC)[reply]

Is there a name for Banak (Arabic word)

[edit]

[File:http://www.248am.com/images/banak.jpg]

Hi I was searching for the name in English about Banak. They look like seeds but i don't know from where these seed come. Some say it's on tree and other say it's a sea weed. It is available in Kuwait, Bahrain, Iran and Turkey. Is there any Scientific name for it.

Can you Please help me to know what is it and from where does it come and what's the name of it in English. Is it harmful for health.

Best Regards, Thanks Heart2Soul (talk) 07:58, 13 February 2012 (UTC)[reply]

Heh, looks like this isn't gonna be so easy. The Kuwaitis have no idea. Nothing on www.pfaf.org, nothing on PubMed... got one saying Pistacia acuminata.[1] Searching "pistacia acuminata" 'banak' gets me 8 Google hits, largely derived from this one reference. Can someone understand this Arabic source: [2]? Apparently Pistacia acuminata is an invalid species name;[3] at least sometimes it can be Pistacia mutica.[4] But that seems to be rare; taking a step back, the original page that started this[5] contains a reply saying it's a "wild pistachio" of some type, at least; it cites a "Slider Station" which seems to be [6] which is another discussion of this. But again the seaweed claim is there too so I can't so readily rule it out yet - though I'm as skeptical to believe that seaweed could form such a hard structure as I am ready to believe a pistachio can. Ah, just about got it! [7] : Wild pistachios do indeed grow plentifully in Iran; they're Pistacia atlantica Desf. which is another name for Pistacia mutica. That says the wild seed is known as Baneh in Iran, and I'm having a hard time finding banak in searches for the species - not sure if that's a dialect or usage difference. Some pictures of its seeds are here. It grows in the western part of Iran, about the right place for some of the references. Wnt (talk) 10:52, 13 February 2012 (UTC)[reply]

What Is The Use Of Risperdone?

[edit]

I want to know the actual meaning of autism? what is its real meaning — Preceding unsigned comment added by Mohammadbari (talkcontribs) 10:47, 13 February 2012 (UTC)[reply]

Does the Autism article not provide you with the information you require? Roger (talk) 12:23, 13 February 2012 (UTC)[reply]
Also risperidone (note the spelling). Dragons flight (talk) 12:29, 13 February 2012 (UTC)[reply]

This sounds like the OP is asking for advice on a medical isue, and as such an answer apart from the cited article should not be provided. Keit121.221.99.129 (talk) 12:28, 13 February 2012 (UTC)[reply]

Let's not jump to conclusions. So far the OP has only asked for information, not advice. Gandalf61 (talk) 12:51, 13 February 2012 (UTC)[reply]

What is done with the mice babies?

[edit]

I am reading this page on breeding mice. It states "The mice can be left together until the pups are ready to be weaned if the cage doesn't get too crowded. If you need mice of predominantly of one sex, you can remove the unwanted sex a few days after birth". Am I to assume from this that the removed mice babies are killed? If so, by what method? Is there an established way to kill mice in these sorts of breeding circles or do they just step on them or something? 95.88.168.1 (talk) 15:50, 13 February 2012 (UTC)[reply]

Not sure how they do it, but when I catch a mouse in my house, I drown it. It only takes about 12 seconds, and there's no mess and no danger from injecting poison or using poison gas. Putting them in a cage and submerging the cage would work. StuRat (talk) 16:02, 13 February 2012 (UTC)[reply]
If you read on in the same page, you will encounter two sections detailing acceptable methods of euthanasia for adult and newborn mice.-- OBSIDIANSOUL 16:10, 13 February 2012 (UTC)[reply]
Mice are most commonly killed (euthanized) by filling their tank with CO2 (causing asphyxiation). Research scientists also use cervical dislocation (often performed by hand by the biologist or the technician), if certain biological characteristics must be kept intact for post-mortem analysis. Learning to perform these techniques used to be requisite in a lot of university biology curricula. (Nowadays, this is not commonly taught in regular biology classes, because it's unpleasant and unnecessary knowledge for most students). Here's the National Institute of Health's Guidelines for the Euthanasia of Rodent Feti and Neonates - essentially, exactly what you're looking for: the established and accepted way to destroy newborn mice and rodents in scientific laboratories. Nimur (talk) 17:43, 13 February 2012 (UTC)[reply]
See [8] here for the specifics. Fetal/baby mice are incredibly resistant to carbon dioxide - they have a tendency to wake up and surprise you no matter how long you expose them. The actual death is by poison, decapitation, or freezing. And no matter what that guideline says, decapitation can surprise you - you take it out of the CO2, you chop its head off, you put the head aside, and three minutes later you see it trying to scream at you. I wouldn't settle for anything less than thoroughly crushing the "dead" head. Wnt (talk) 19:26, 13 February 2012 (UTC)[reply]
Do you have a source for that last part? I find it hard to believe that a decapitated mouse head would start screaming 3 minutes after being severed. 89.0.151.74 (talk) 01:37, 14 February 2012 (UTC)[reply]
I assumed that was the frozen case, and the three minutes was the time for the head to thaw. StuRat (talk) 02:28, 14 February 2012 (UTC)[reply]
It still seems highly unlikely. And a completely frozen mouse head thaws in 3 minutes? Unless it was being heated up for some reason that seems even more unlikely 188.96.254.242 (talk) 11:55, 14 February 2012 (UTC)[reply]
To my reading, Wnt seems to be suggesting the screaming after chopping off the head happened when the mouse was asphyxiated with CO2 (I guess to try and reduce pain and stop struggling). I suspect it wasn't frozen as well. I agree it seems a fairly strange. Nil Einne (talk) 14:20, 14 February 2012 (UTC)[reply]
This was with newborn pups being at 4 C before the decapitation, but allowing them to warm (and in room air) afterward thinking they were dead. No, as far as I know truly frozen heads don't come back. ;) Wnt (talk) 16:57, 14 February 2012 (UTC)[reply]
How is using CO2 to cause death supposed to reduce pain or stuggling when the buildup of CO2 in the blood is the reason strangulation/hypoxia/drowning feels so bad? Merely replacing the oxygen in air with an inert gas (like nitrogen) Nitrogen asphyxiation would be an easy way to cause suffering-free death. And what, they freeze mice to death? How is that humane? Sagittarian Milky Way (talk) 22:35, 14 February 2012 (UTC)[reply]
It's a lot more humane than having a mouse broken and twitching in a standard mousetrap, or slowly dying of starvation in a glue trap. Researchers are browbeaten into following ethical guidelines handed down from on high; there's no real perception that they're supposed to make sense. Wnt (talk) 05:09, 15 February 2012 (UTC)[reply]
That said, the sensation from breathing carbon dioxide isn't the same as a true suffocation - the body seems to have a strong presumption built in that breathing will solve the problem. Accidental suffocation by dry ice does happen. Fooling with (obviously) less-than-lethal amounts of dry ice to check the effect for myself, my opinion was that carbon dioxide vapor misty with water could cause a burning sensation like snorting soda up into the nose, but otherwise the sensation is no different from what is felt during hard exercise. Both putting them on ice and carbon dioxide are viewed as "anaesthesia" for mouse pups. Wnt (talk) 05:19, 15 February 2012 (UTC)[reply]
(Note that this is partially OT since I'm replying to the point raised by SMW so it primarily concerns adult rodents.) Many sources suggest CO2 is or was recommended as a humane method of euthanasia for rodents [9] [10] [11] [12].
Notably the first source from 2006 suggests while there appears to be some concern with CO2 usage as it appears to cause some discomfort and adversion which is of serious concern, and there's no clear ideal protocl; there is even less evidence on the use of gases causing hypoxia including nitrogen and so they are not recommended. It notes that while there is some evidence of hypoxia working well in other species, there is insufficient research for rodents and you can't generalise (which makes sense to me). Similarly there is insufficient research on the use of volatile anaesthic agents in rodents. Note that the participants there considered a relatively pain and discomfort free death more important then quick loss of consciousness and were operating under that assumption.
Wnt noted that sometimes rodents seem to regain consciousness after exposure to CO2, accepting that this is true it may make sense to use CO2 to knock out the rodents and then use cervical dislocation to ensure death. If it causes true unconciousness but not death with a minimal of discomfort which admittedly both seem to be a big if, then using CO2 combined with cervical dislocation may make sense. The NIH reference actually mentions that regaining conciousness after exposure is documented and suggests thoracotomy after CO2 exposure as one possible method to ensure death (although getting the protocol right is probably equally important).
BTW, the last ref suggests the use a prefill which causes a quick death, and while the first ref agrees on this, it suggests that this may actually be worse at it appears to cause pain for ~15 seconds (whereas rising concentration appears to cause adversion and discomfort at least partially due to dyspnoea but possibly no pain). The last ref also suggests that the are concerns over cervical dislocation, as brain activity, appears to continue possibly for up to 13 seconds after the dislocation. I should mention while looking at something for my earlier reply I came across one source which recommended various anaesthia agents but which didn't mention CO2.
I think the 'take home' message is that the ideal euthanasia method is complicated and relying too much on information from fairly distant species (like humans) is probably not a good idea. Similar using random pieces of anecdotal evidence like the time it takes for death is not ideal either. What we need is more research which as of 2006 appears to be lacking. (Perhaps new tools like fMRI will help although to given the level of research with existing tools like EEG that may be expecting too much.)
Nil Einne (talk) 20:04, 15 February 2012 (UTC)[reply]
Good grief, they should just make a mouse or head crushing device and be done with it. With a no-look disposal. (maybe sewage or incineration or some kind of canister) Maybe each room in a university biology building gets it's own "mouse memory hole" to a single mouse-crushing device if justifiable financially.
And oh, I must've read too quickly. The freezing apparently is to ensure they "stay dead", so they're not just dropping them on dry ice awake. Sagittarian Milky Way (talk) 22:41, 15 February 2012 (UTC)[reply]
  • I did a very brief stint in an Animal Ethics covered industry in a legal jurisdiction that had a very comprehensive and modern Animal Ethics procedure as a result of actions that were perceived to be unconscionably unethical by parliamentarians and practitioners. While Wnt may be right that animal ethics law, regulations and conduct practices may not relate to the experience of animals; and, that they may be a bureaucratic dance; given what I have heard about breaches of Animal Ethics it is better to dance a dance in public, than allow walled gardens of private research or practice to emerge that are not subject to public scrutiny. Many of the reasons are the same as the reasons that we ought to demand that researchers and practitioners expose their methods and data. Fifelfoo (talk) 22:54, 15 February 2012 (UTC)[reply]

Why does my tervis tumbler attract cat lint?

[edit]

I'm sick,and I was drinking ginger ale from a tervis tumbler and my cat got near. When my cat got near, the hair from my cat stuck to the cup. What is causing it to attract the cat hair?

Static electricity. Perhaps the fizzing of the ginger ale causes the cup to acquire a charge ? StuRat (talk) 16:05, 13 February 2012 (UTC)[reply]
In my experience cat fur will stick to any surface... --TammyMoet (talk) 17:37, 13 February 2012 (UTC)[reply]
Except, apparently, cats. :-) StuRat (talk) 02:24, 14 February 2012 (UTC) [reply]
Cat hair was historically one of the first items discovered to easily transfer static charge - I think cat fur and amber are specifically mentioned by Aristotle during discussion of what we now understand as static electricity. (I'm having hard time tracking the reference, but our article has discussion of the ancient history of electricity, and mentions Thales of Miletus - perhaps I am mistakenly attributing this to Aristotle). Hair is particularly notable for its very low mass, and therefore its very high charge to mass ratio - so, it is visibly affected by static electricity. Nimur (talk) 17:48, 13 February 2012 (UTC)[reply]

Effect of exercise on metabolism, continued...

[edit]

It would seem like common sense that, yes, one's metablism would increase while exercising. However, I noticed something very interesting when I was a teenager in a sports camp (15½ years old, to be precise): I, and several other members had a sharp decrease in hunger after we'd gotten in good shape. We filled up our plates, and would leave half of the food sitting there.

However, as an adult, I've noticed the exact opposite: my hunger increases when I am in good shape.

What could explain the difference? Is it a coincidence or does it have something to do with the fact that the teenage body metabolizes differently than the adult body?

Magog the Ogre (talk) 20:38, 13 February 2012 (UTC)[reply]

Exercise has a complex effect on hunger, influencing hormones that control appetite (ref) - it's much more complicated than "more calories have been burned so increase hunger level to compensate". There are also complex reactions to hunger management hormones - see Leptin#Obesity and leptin resistance. Other effects pertain too; swimming appears to increase hunger, perhaps due to effects related to body temperature (ref). -- Finlay McWalterTalk 20:48, 13 February 2012 (UTC)[reply]

occurence of man made plutonium

[edit]

I would like some info on occurence of man-made plutonium,ie how much exists ,its distribution in the environment etc — Preceding unsigned comment added by Sebastian barnes (talkcontribs) 21:28, 13 February 2012 (UTC)[reply]

Have you read our article on plutonium? We note that, as of 2007, the world plutonium stockpile was estimated at 500 tons, with roughly half each in military and civilian stockpiles, though other estimates have been two to three times that amount. We additionally have an article on plutonium in the environment. — Lomn 21:37, 13 February 2012 (UTC)[reply]
Note that the 500 ton figure is the world stockpile of separated plutonium. The estimates can vary quite a bit especially if you count unseparated (e.g. in spent fuel or in some sort of waste solution or just as particles in the environment following nuclear testing/accidents/etc.). --Mr.98 (talk) 21:47, 13 February 2012 (UTC)[reply]

Black hair

[edit]

Hello! My question is: How common is / What is the distribution of true black hair (not dark brown) among persons of white European descent? Your article black hair says that it is "sometimes found" but dark brown is more commonly mistaken for black; I am interested in true black. It seems to me that only persons with at least some non-European or minority bloodline (for example, Jewish or Wanderer; this excludes the Basque nation) have truly black hair, is this correct? Thanks. 24.92.85.35 (talk) 23:00, 13 February 2012 (UTC)[reply]

There's going to be a problem with defining "European descent", as Europeans have had interaction with Africans and Asians from ancient times. The Moors, Turks, and Mongols are just a few examples. Do you classify anyone containing any DNA from those groups to be "non-European" ? StuRat (talk) 23:11, 13 February 2012 (UTC)[reply]
Also, there is considerable variance within Europe. Italians and Norwegians are both whites of European descent, but their hair pigmentation differs drastically.
"True black" is the ancestral color, all other hair colors are relatively recent European mutations. Even in Europe there are people who still have ancestral hair color genes. The further you move from the Baltic and Scandinavia, the more common they become.--Itinerant1 (talk) 00:43, 14 February 2012 (UTC)[reply]
Since every single human that exists has "at least some non-European or minority bloodline" (see Recent African origin of modern humans), your hypothesis that only such people have the kind of hair that's considered "truly black" is correct. There does not exist a single person who has 100% European bloodlines, and has truly black hair (or any other kind of hair, for that matter, or a lack of hair). Red Act (talk) 19:01, 14 February 2012 (UTC)[reply]
Lots of Irish people have jet black hair. Alansplodge (talk) 22:39, 14 February 2012 (UTC)[reply]
Your statement is completely consistent with mine. Red Act (talk) 03:52, 15 February 2012 (UTC)[reply]
As a matter of fact, if you want to go all the way back to our roots, every single one of us is 100% African. Whoop whoop pull up Bitching Betty | Averted crashes 23:48, 14 February 2012 (UTC)[reply]

Uniform current distribution

[edit]

I have a project wher I wish to achieve a uniform current distribution in a conducting rectangular sheet. The current is to be applied along the longer edge (say 200mm) with the opposite edge being the ground return. How do I ensure uniformity in the sheet when the source of current is the collector lead of a transistor?--92.28.88.241 (talk) 23:05, 13 February 2012 (UTC)[reply]

What is your sheet made of, what are it's dimensions, and what level of uniformity do you need really? Dragons flight (talk) 01:26, 14 February 2012 (UTC)[reply]
Copper. 200x100 mm. Say +/-10%?--92.28.88.241 (talk) 01:29, 14 February 2012 (UTC)[reply]
Oh and about 1.4 mils thick.--92.28.88.241 (talk) 01:51, 14 February 2012 (UTC)[reply]
You need to take into account the thickness of the sheet to determine what to feed it with. You need to do 2 things: (1) ensure that the grounding is in a sheet of resistance per square less than 5% (half you permitted variation) of that of your sheet, i.e., use as a ground a sheet at least 20 times the thickness of the sheet, or the same as the test sheet but 20x as long, and (2) feed the sheet via a thin, current distributing, sheet between the sheet in question and the transistor collector. This sheet can be the same as the sheet in question, but 20x as long. Depending on the thickness of the test sheet, you could reduce the size of the feed sheet by using an assymetric interpose sheet. You can make an assymetric sheet out of Teledeltos paper (paper treated chemically to give electrical conductivity with a definite resistance per square) with regaularly spaced longitudial copper strips - make sure the strips approach but do not touch either the copper test sheet and and the feed point. These measures make the ground and feed edges of the test sheet approx equipotential. This will work well at DC and low frequencies, but if you want radio frequencies you will have problems with magnetic fields and capacitance to the ground conductor and feed conductor/sheet. Keit121.215.39.119 (talk) 02:07, 14 February 2012 (UTC)[reply]

My earlier post was made before you gave the thickness. You will have problems with skin effect at higher frequencies. Keit121.215.39.119 (talk) 02:12, 14 February 2012 (UTC)[reply]

Is there an article or link with diagrams you can point me to?--92.25.98.82 (talk) 10:20, 14 February 2012 (UTC)[reply]
Not that I can think of, right now. I recall some general books of some relevance that were in the Uni library when I was there years ago, but they were writen by Ph.d's for Ph.d's in the 1930's, so you may find them difficult anyway. What is the purpose of the uniform current sheet in your project? If you can tell us that, I MAY be able to give you a good reference. If my explanation above is not clear, tell me, maybe I can clarify or make a diagram for you. I've since thought of a couple more ways of providing a uniform current sheet, they are simpler but have limitations that may not suit - without knowing what yor pupose is, I can't tell. Keit58.170.153.169 (talk) 13:15, 14 February 2012 (UTC)[reply]
I need to uniformly magnetise a thin sheet of ferrite/--92.25.98.82 (talk) 14:21, 14 February 2012 (UTC)[reply]
Then you are going about it the hard or even wrong way, needing an excessive current, especially if you want to permanently magnetise it. What is the dimensions of the piece of ferrite you want to magetise? What type of ferrite? If you want to permanently magnetise a hard ferrite, all you need to do is insert it in a uniformly wound reasonable close fitting coil of ordinary wire, and inject a current well above the saturation value. Make the coil close-wound and big enough to completely enclose the ferrite. Because the ferrite will be driven well beyond saturation, it will end up uniformly magnetised even if the field generated by the coil is not that uniform. Keit58.170.153.169 (talk) 15:27, 14 February 2012 (UTC)[reply]
I need to temporarily drive the thin soft ferrite sheet, 200mm x 100 mm x 0.4 mmm, into saturation for about 50ns then bring it out of satn again.--92.25.98.82 (talk) 16:14, 14 February 2012 (UTC)[reply]
Hers the ferrite specs:

 EPCOS AG 2006. FPC film

Physical properties (material values defined on 0.2 mm thick film) 1) T = 25 °C to IEC 51 (CO) 282 2) T = 23 °C and 50% r.h. 3) UL 94, flame class V0 (listed E 140 693) Material Symbol Unit C350 C351 3) Initial permeability 1) f = 1 MHz μi 9 ±20% 9 ±20% Flux density (near saturation) 1) H = 25 kA/m f = 10 kHz BS mT 255 255 Remanent flux density 1) H = 25 kA/m f = 10 kHz Br mT 9 9 Coercive field strength 1) H = 25 kA/m f = 10 kHz HC A/m 600 600 Relative loss factor 1) f = 10 MHz f = 1 GHz tanδ/μi < 0.005 < 0.400 < 0.005 < 0.400 Hysteresis material constant ηB 10 –3/mT < 2 < 2 Temperature coefficient1) α = Δμ/μΔT 1/K < 5 · 10 –5 < 5 · 10 –5 Density kg/m3 2930 2930 Resistivity 1) f = 1 kHz f = 10 MHz ρ Ωm 500 100 500 100 Dielectric constant 1) f = 1 kHz f = 10 MHz εr 700 21 700 21 Dielectric strength kV/mm 1 0.8 Max. operating temperature Tmax °C 120 200 Tensile strength 2) σZ N/mm 2 1.5 2.5 Tearing resistance 2) % 25 25 Compressibility 2) κ N/mm 2 70 70 C350, C351 B68450 … B68452 FPC film 5 09/06 --92.25.98.82 (talk) 16:20, 14 February 2012 (UTC)[reply]

The data was rendered almost meaningless by the copy-paste of tabular data into Wikipedia, however I was able to download the 3 page datasheet for Epcos FPC film with no difficulty. Unfortunately this has got a little outside my competence. However I can say that neither of the two solutions I gave is going to work very well (they being suited for DC & low frequencies), but I still think a close fitting coil surounding the ferrite is the way to go. This will minimise the Li^2 value, and make it at least feasible vis-a-vis finding a transistor to do the job. It will be better to use copper tape, say 10mm wide, rather than round wire. Your solution of a current sheet adjacent to the ferrite could work, if the sheet is SIGNIFICANTLY LARGER in area than the ferrite (say 600 x 300 mm for a 200 x 100 mm ferrite), but will require a truely excessive current. There are 2 issues you may need to address: 1) with a soft ferrite of an intial permability of only 9, you may need to prove analytically that magnetisation will be sufficiently even, throughout the magnetisation/demagnetisation cycle. 2) The requirement to go in and out of saturation within 50 nS will be difficult to meet. Either the current, or the back-emf voltage, or both, (depending on the coil turns) will be such as to require a vey substantial transistor, and such transistors are not fast. A way around this is to use multiple coils and multiple transistors - each coil covering part of the ferrite. Don't parallel transistors directly to one coil - at that speed you will have great difficulty establishing that each transistor is contributing equally. The Wikipedia Science desk is moderately good for thowing up ideas when you are stuck, and veey good for bring to light wierd things that make us think, but not very good for questions like yours where you cannot give us the full picture, where you may be unable to post the problem accurately, and what requires significant mathematical analysis, which we Wikipedians may be too lazy to do for you. I recommend you seek out one or more engineers or university lab staff competent in magnetic cuircuits, and engage them in brainstorming and analysis. Keit124.182.183.68 (talk) 04:17, 15 February 2012 (UTC)[reply]
Sorry about the format of the data. I just had time to copy/paste before having to rush out.
Im glad you think my solution of (a) low inductance coil(s) to saturate the ferrite maybe has a chance. As you rightly say,the current requirements for driving a single sheet of copper maybe excessive for one transistor and again there is the problem of current uniformity. Im interested to know why you think such a current sheet would need to be much larger than the ferritte sheet. Is this because of end effects?
Initially I had thought of using a tapered 'matching section' from the physicaly small transistor collector to the long edge of the copper sheet to try to ensure greater uniformity of current density. However it may be better to follow your idea of parallel conductors each driven by their own transistor. Would these conductors need to overlap the ferrite substantially as well to achieve uniform magnetisation and synchronicity of saturation in your opinion?
Thank you for your interest and help on this rather unususal problem. I am of course, doing my own web based research on this but without a great deal of success; but your answers have been very useful so far!92.25.98.82 (talk) 09:20, 15 February 2012 (UTC)[reply]
I think that the magnetising sheet needs to be significantly larger than the ferrite sheet for these reasons: a) without it the edges of the ferrite will not get the same "dose" of flux as the rest of the ferrite; b) nearby ground and supply conductors, if not in the same plane and direction as the magnetising sheet, will give a magnetic field that will distort the field provided by the magnetising sheet, and c) capacitance between the ground and feed conductors and the magnetising sheet will carry current and thus distort the magnetic field. With a pulse length of only 50 nS this will be a significant issue. Don't forget the path thru the transistor wires and the transitors will set up a magnetic field.
If you go with the current sheet idea, and make the current sheet larger than the ferrite sheet, plus drive it with multiple transistors, I suggest that you interpose a resistive sheet (eg teledeltos paper, or if you can't get teledeltos with low enough resistivity, a THIN copper open mesh will do) between the magnetising sheet and the wires to the transistors. With (say) 10 transistors, you won't need the assymetric conductors I mentioned originally, and you'll need only a short interpose sheet - say 30 mm in the direction of current. The transistor wires will necessarily be quite long, but that's ok (sort of). Take care to install them with a simple graceful curve, solid wire, all identical.
If by "multiple conductors" you mean my idea of having a separate coil and transistor each functioning for part of the ferrite, then install all the turns equally spaced and covering all (just all) the ferrite area, as though it is all just one big coil. Each sub coil needs to be at least 10 turns, preferably 20, or you'll be in trouble with the magetic field from the lead out wires distorting the intended field. Minimise the problem by routing the lead in and lead out wires next to each other - OR - use a multitude of single turn coils each with its lead in and lead out wires next to each other. Either way, use copper tape rather than round wire as I said, so that the ferite is COMPLETELY covered by current carrying conductor.
I have thought of a refinement of your current sheet idea, based on PN avalanche breakdown and a double groundplane, which would work much better at high speed, without requiring high performance transistors, and not requiring long transistor leads - best explained by a diagram. However the pulse shape may not be nice and square - the turn-on will be slower than turn off. How square or otherwise do you need the pulse?
Yes, the internet is fantastic, not not good for research this type of problem, nor is Wikipedia Science Desk. Have you tried engaging suitable Engineers or University lab staff? Keit124.182.183.68 (talk) 13:20, 15 February 2012 (UTC)[reply]
I am not now too keen on the single copper sheet appproach if it means having to use multiple transistors and somehow share the current beween them (which I assume is the pupose of your Teledeltos paper idea). I would however not have multi turn coils because, as we agreed earlir, these would act like transmission lines and necessarily delay build up af the field whilst the current was travelling along the coil. I would rather go for a number of separate flat conductors(each say 2cm wide) and each driven by its own transistor. In this way, the delay of magnetising current from one side of the ferrite to the other side would be minimised. These conductors in conjunction with the associated ground plane would still act as transmission lines, but short ones thus giving quicker magnetisation.
I am intrigued by your idea of (a charged line and?) an avalanche device to generate short pulses. My pulses dont need to be that square at all as long as thers an up period of about 50 ns and a similar down period: in fact the current waveform could even be a soft clipped sine wave!.--92.28.90.165 (talk) 14:28, 16 February 2012 (UTC)[reply]
The pupose of the teledeltos sheet is not to force the transistors to equally share the current - for 50nS, the wires to the transistors will do that adequately well. It won't matter if (say) one transistor takes 20% or 30% more current than another. The purpose of the teledeltos sheet (or copper mesh sheet) is to even out the current distribution in your current sheet conductor, to exceed your stipulation of less than 10% variation. If you use one transistor, you'll need a lot of teledeltos or mesh - if you use multiple transistors, the job is part done, so you'll need a lot less teledeltos or mesh. That's why I specified less teledeltos the second time.
I did not realise that you are the same person who was asking about the 500 turn coil. You asked about hot-to-ground propagation delay, so I gave you an answer on how to estimate the delay. Wether or not it was relavent to a real application was your problem, because you did not state the real application. You may recall I also said that the effect of the delay will generally be swamped out in practice by inductance-limited rate of change. You need to keep things in perspective. For this question I'm talking about coils of 20 turns or less with an axial length of around 200mm and a core of permability 9 times that of air as distinct from you first question about a 500 turn coil of axial length 200mm air cored. Hot-to-ground propagation delay will not be a problem in the coils I have proposed for magnetising your ferrite. Don't forget that the propagation in this case will not be just a simple case of propagation from the hot end anyway. The pulse will progagate out frm the (transistor) switch via both the hot conductor and the ground conductor towards the magentising coil/sheet.
My avalanche & double groundplane/charged line idea requires a diagram to explain. I'm happy to make up a diagram for you, but this question is about to be archived. If you are still interested, please create a new question entitled "Even magetisation of ferrite for 50nS" or similar, and then I can respond. Please include in the text your other requirements (eg type of ferrite, dimensions, etc). Also state in the new question that it is a follow up of this question to allow continued debate - otherwise some other volunteer may consider you a nuisance repeater - we get them from time to time. If you create a new, concise question, it may prompt someone else to make suggestions too. This one is rather long and I suspect other may have lost interest.
We are happy to help you when you are stuck - that's what we are here for - but you need to work too. I've recommended twice that you seek help from a university lab or qualified electronics engineer - because this forum isn't an efficient way to get a solution to this type of problem. Have you tried engaging such persons? If not, why not?
By the type of questions you ask, and the way you ask them, I sense that you are a very intelligent person, but are out of your depth on this, and you are not an electronic engineer. Am I right? We could go on for weeks ping-ponging like this, but without practical input from someone in direct contact with you, an appropriate solution that works may still not emerge.
Having said all that, I'm still happy to cook up a diagram of the avalanche/double ground plane/charged line idea. But you have indicated you would like 50nS "on" time with a similar down time. My avalanche idea is dependent on a relatively long relaxation period - say 50nS on and 1000nS off, as otherwise the avalanche device will have excessive dissipation. Is this acceptable?
Keit60.230.221.190 (talk) 09:57, 17 February 2012 (UTC)[reply]
I have not sought help from others in the fieid because I am a lone worker working from home. I have had experience in electronics engineering and Im sorry that did not come across in my posts. Perhaps i was being too vague. I agree that I am out of my depth, but arent all researchers until they have found the answer? i am following your suggestion and posting a new question defining the problem more explicitly.--92.25.101.91 (talk) 13:45, 17 February 2012 (UTC)[reply]