Jump to content

Danijel Gorupec

Senior Members
  • Content Count

  • Joined

  • Last visited

Community Reputation

76 Excellent

About Danijel Gorupec

  • Rank

Contact Methods

  • Website URL

Recent Profile Visitors

The recent visitors block is disabled and is not being shown to other users.

  1. I heard some amateur astronomers complaining more about bluish light than about reddish light when making observations (I guess, blue disperses more causing more light pollution). I also heard somewhere, but I cannot remember how much was this reliable, that night animals might get disturbed more by bluish light than by reddish light. Anyone heard about this? For me, the ideal street lighting would adjust its intensity during the night - decreasing its power and shifting more red after 23:00h... I guess it might also adjust to foggy conditions somehow... or even to moonlight intensity.
  2. Hmm... but let's compute expectation value for kinetic energy... I am doing this for the same wave-function (ground state particle-in-a-box): ('h' is actually h-bar) The above is the correct result when the integral is computed over the whole box. But even if I integrate over only a part of the box, I still obtain real values - does it mean that such result might have some actual meaning? Not being lazy, I tried to compute kinetic energy distribution (or density) over the box. Moderators can move this to the speculation forum. I am using a sloppy notation as I know no better. This is the kinetic energy density function for a ground-state particle in a 1D box, how I calculated it.... Can this have any meaning? The function shape resembles the position probability density function for the particle. I would be very glad if there is any function that can show kinetic energy density for wave-like particles. I like energy distributed in space (that is, local i like the local conservation of it). However, I never heard anything about it.
  3. So, from various sources this is an time-independent wave-function for a ground state particle in an infinite square (widht 'a'). The momentum calculation gives zero, as expected. But if I only integrate over half of the box width (the last line) then I get an imaginary result. Does this mean that this result has no interpretation (that it, it means nothing)? [sorry, the 'h' should actually be h-bar, but I don't have this symbol]
  4. Hello... Say there is a particle in a 1D box. The particle is in certain state. I know its position-space wave function. I can compute expectation value of its momentum by sandwiching the momentum operator between psi* and psi and integrating over the whole box width (or over whole infinity). Question... If I don't integrate over the whole box, but only over a part of the box, can the result be interpreted as follows: if I simultaneously measure momentum and position of identically prepared setups, and if I only consider those cases where a particle was found within the part of the box for which I computed the integral, then the average momentum of those particles will equal to my integral?
  5. Yes, it makes sense to me too... In fact, I guess that the spin-flipping due to the thermal agitation is the reason why paramagnetic materials stop being paramagnetic when removed from an external field. In the meantime I tried to find more argon data to check if argon energy levels are spaced close enough to be excited by room-temperature collisions... unsuccessfully... But I guess not. Anyway, it is my 500th post, and I am giving +1 to everybody... you lucky devils!
  6. I understand... I abandoned the idea that rotation of atoms can be used as energy storage (to aid heat capacity). I can accept that atom rotation is not a thing. I am investigating now if electron energy transitions in some heavier atoms are small enough to count as 'unfrozen' degrees of freedom for heat storage.
  7. Ok, I see that the 'rotating atom' is a problematic idea. Some back of the envelope calculations, if I did it correctly: Average kinetic energy of a gas particle at 300K is about 0.04eV; The 1420MHz photon has about 6 micro-eV. Therefore, the thermal motion cannot excite an electron in a hydrogen atom from 1S to 2S, but it should be able to flip-flop its spin easily. Is this happening in a real gas? (I guess it might be happening in monatomic hydrogen, but not sure about H2 - probably spin cannot be flipped in H2?) Next, if we are dealing with heavier atoms (say argon) then I expect external electrons have smaller difference between energy levels - possibly comparable to 0.04eV. Should I then expect that those electrons in argon contribute to the heat capacity of argon gas? I guess you are referring to the very high (10.2eV) temperature needed - not that there is some other limitation why collisions are unable to excite quantum states?
  8. Hmm... I guess, one way to know that they are rotating is by measuring heat capacity of the gas (and from this we know that they are NOT rotating). But this is a proof of neither, imo. Because atoms are not solid bodies, I guess that their rotation would mean a net non-zero rotation of all the parts. So, when thinking about your last sentence, I guess that my question should actually be: why in monatomic gases, thermal excitation do not cause change in orbital momentum of atom's external electrons? Is this a better way to ask? (I really am not sure what I am talking about - I am just learning the new stuff here).
  9. in monatomic gases, why isn't atom rotation taken as one possible degree of freedom - that is, why energy cannot be stored in atom rotation? Is this degree of freedom non-existent or just frozen? Google finds many answers, but they seem to differ. Some say that rotation is simply not physical (whatever it means), some say that a single quantum of rotational energy is much larger than available energy at normal temperatures...
  10. I read about Oliver Heaviside - people mention his eccentricity. Among other things, he apparently liked heating his home to levels uncomfortable for most people. Sources don't tell if he preferred smoothies over coffee.
  11. I tried to sketch motor torque curves (at nominal and reduced voltage) and some hypothetical load torque curve (a constant-torque load - red line in the graph). You can see how, with reduced motor voltage, the motor will settle at lower speed (lower rpm). This means that its slip increases (the slip says how much its speed is reduced in comparison to the synchronous speed). The thing is that the current the motor draws (its impedance, if you want) very much depends on the slip. As a result, at the reduced voltage the motor will draw significantly more current (It might go above its nominal current causing overheating and stuff, despite the fact that the delivered torque is still at the motor nominal level). As you can understand, some load curves (like that of a fan) will not cause so much problems with a decreased voltage as some other load curves might (the displayed constant-load curve is a bad one).
  12. In fact... I am not sure electromagnetic interaction is responsible for collisions in a (fermionic) gas.
  13. Hmm... But if they are fermions, they should be, in a way, colliding, no? On the other hand, if they are bosons, they should be lumping toghether, no? I mean, they should be interacting by some mean... unless you believe they are so much different that are beyond wave mechanics?
  14. How you described those 'slingshots' reminds me on evaporative cooling - is that how you imagine it? Or you thinking about slingshots caused by massive ordinary-mater objects? (I like to think about DM as gas... I guess, DM particles may still obey Bolltzman distribution of their energies, unless maybe the gas is so cold that quantum effect could prevail... If I suppose Boltzman distribution, then some particles should obtain escape velocity... I wonder if we know the upper limit for the mass of one DM particle?)
  15. Ha ha... tensors... I would not know what to write about them further in the glossary I am thinking to remove surface integral of scalar field - I don't know if it is used anywhere in magnetism. At the moment I keep it for symmetry
  • Create New...

Important Information

We have placed cookies on your device to help make this website better. You can adjust your cookie settings, otherwise we'll assume you're okay to continue.