4 ms·

Isn't that complete nonsense? Although the gate gets more charge, that's going to be balanced out by less charge elsewhere. The USB drive as a whole isn't gaini

by kens 7y ago

Isn't that complete nonsense? Although the gate gets more charge, that's going to be balanced out by less charge elsewhere. The USB drive as a whole isn't gaining electrons, or else it would have a negative static charge, which would be dissipated as soon as it is connected to ground.

monocasa 7y ago
It's not a closed system, USB devices need to be plugged in to change state like that. Also, GND isn't neutral charge.
- kens 7y ago
  Yes, that's the point. The USB device will lose as many electrons as it gains.
  - monocasa 7y ago
    It doesn't though, that's not how flash works. The floating gate traps electrons in a pool of insulation (that they only got in there in the first place due to quantum tunneling).
    - segfaultbuserr 7y ago
      The question can be reduced to the following: Where does the electric charge that jumps through the insulator to the floating gate come from? Does it come from somewhere inside the chip, or does it come from the external power source? I'm not sure, so I'll just make a guess here: I fail to see the power source we used here has any special property, or how does the chip affect the power source in an unusual way, so I assume our power source obeys KVL and KCL as usual, which means its charge is conserved from the perspective of the power source, and there's an equal number of charge leaving and arriving at both electrodes of the power supply. Thus, the electric charge received by the floating gate is of internal origin, and no charge is gained or lost if we count the chip as a whole. Correct me if I'm wrong.
      - monocasa 7y ago
        It's not really a closed loop system anymore where KVL/KCL strictly apply, it's more like an electron gun.
        segfaultbuserr 7y ago
        I'm still not convinced that the electric charge on the floating gate comes from the power supply. I believe the energy comes from the power supply, but the electron comes from the conductive path in the chip. Consider a simple case of an indirectly-heated vacuum diode, there's a filament, a cathode and an anode. The current in the filament heats up the cathode and allows an electron on the cathode to gain energy and breaks free, flying towards the anode. This electron returns back to the cathode via the external circuit. But whether the electron gets "recycled" or not (whether it's a closed loop or not), note that the electron comes from the metal that forms the cathode, it does not come from the power supply. So the anode receives as many electrons as the cathode loses. The power supply only heats the cathode indirectly to transfer the energy to the free electrons that already exist on the cathode from the beginning. Same situation occurs in an electron gun. The electrons come from the metal of the cathode itself, it doesn't come from the power supply. Although I don't know anything about physics, but I believe it's a similar situation in Flash chips. A potential difference applied between the conductive path between the drain and source near the floating gate accelerates the electrons in the path, allowing some to jump to the floating gate. The electron is not supplied by the power source. > it's more like an electron gun. More likely than not that is a return path for the electron to go back to its power supply in many electron guns, otherwise the object at the end of the beam eventually becomes positively charged and repeals the electrons. In a CRT, a conductive coating [0] is connected to the power supply to collect the electrons after they hit the screen. But again, I think it's not the point, the point is that the electrons in the floating gate comes from somewhere else in the chip, not from an external source. BTW: I'm not trying to "win" an argument, I just think it's an interesting question. I did not downvote your post. [0] https://en.wikipedia.org/wiki/Aquadag https://en.wikipedia.org/wiki/Aquadag
        segfaultbuserr 7y ago
bluesign 7y ago
Yeah actually from my understanding, even in the transistor, state (location) of electrons should change only, not the number of electrons.
- nomel 7y ago
  Isn’t the concept that the location difference causes a potential/energy difference, which has mass?
  - therein 7y ago
    But the two different locations in question are both on the device itself.
  - viklove 7y ago
    It's all changing within the system (USB drive), so to an outside observer there is no change in mass.
    - nomel 7y ago
      A system with more potential energy will have more mass [1]. For example, a compressed spring has more mass than a spring at rest, and will take more energy, from an external force, to get to the same velocity. Same with a spinning disk. So unless there's a nice balancing act, with no potential energy change from moving electrons around within the drive, the mass of the drive will change, negligibly, from the potential energy change of the gates. So, the real mass loss of a USB drive is electrons out in addition to potential energy loss from the floating gates. So it's, the world smallest double whammy! [1] https://physics.stackexchange.com/questions/274329/does-potential-energy-of-an-object-increases-its-relativistic-mass/274680 https://physics.stackexchange.com/questions/274329/does-pote...

by kens 7y ago

monocasa 7y ago

It's not a closed system, USB devices need to be plugged in to change state like that. Also, GND isn't neutral charge.

kens 7y ago
Yes, that's the point. The USB device will lose as many electrons as it gains.
- monocasa 7y ago
  It doesn't though, that's not how flash works. The floating gate traps electrons in a pool of insulation (that they only got in there in the first place due to quantum tunneling).
  - segfaultbuserr 7y ago
    The question can be reduced to the following: Where does the electric charge that jumps through the insulator to the floating gate come from? Does it come from somewhere inside the chip, or does it come from the external power source? I'm not sure, so I'll just make a guess here: I fail to see the power source we used here has any special property, or how does the chip affect the power source in an unusual way, so I assume our power source obeys KVL and KCL as usual, which means its charge is conserved from the perspective of the power source, and there's an equal number of charge leaving and arriving at both electrodes of the power supply. Thus, the electric charge received by the floating gate is of internal origin, and no charge is gained or lost if we count the chip as a whole. Correct me if I'm wrong.
    - monocasa 7y ago
      It's not really a closed loop system anymore where KVL/KCL strictly apply, it's more like an electron gun.
      - segfaultbuserr 7y ago
        I'm still not convinced that the electric charge on the floating gate comes from the power supply. I believe the energy comes from the power supply, but the electron comes from the conductive path in the chip. Consider a simple case of an indirectly-heated vacuum diode, there's a filament, a cathode and an anode. The current in the filament heats up the cathode and allows an electron on the cathode to gain energy and breaks free, flying towards the anode. This electron returns back to the cathode via the external circuit. But whether the electron gets "recycled" or not (whether it's a closed loop or not), note that the electron comes from the metal that forms the cathode, it does not come from the power supply. So the anode receives as many electrons as the cathode loses. The power supply only heats the cathode indirectly to transfer the energy to the free electrons that already exist on the cathode from the beginning. Same situation occurs in an electron gun. The electrons come from the metal of the cathode itself, it doesn't come from the power supply. Although I don't know anything about physics, but I believe it's a similar situation in Flash chips. A potential difference applied between the conductive path between the drain and source near the floating gate accelerates the electrons in the path, allowing some to jump to the floating gate. The electron is not supplied by the power source. > it's more like an electron gun. More likely than not that is a return path for the electron to go back to its power supply in many electron guns, otherwise the object at the end of the beam eventually becomes positively charged and repeals the electrons. In a CRT, a conductive coating [0] is connected to the power supply to collect the electrons after they hit the screen. But again, I think it's not the point, the point is that the electrons in the floating gate comes from somewhere else in the chip, not from an external source. BTW: I'm not trying to "win" an argument, I just think it's an interesting question. I did not downvote your post. [0] https://en.wikipedia.org/wiki/Aquadag https://en.wikipedia.org/wiki/Aquadag
        segfaultbuserr 7y ago

bluesign 7y ago

Yeah actually from my understanding, even in the transistor, state (location) of electrons should change only, not the number of electrons.

nomel 7y ago
Isn’t the concept that the location difference causes a potential/energy difference, which has mass?
- therein 7y ago
  But the two different locations in question are both on the device itself.
- viklove 7y ago
  It's all changing within the system (USB drive), so to an outside observer there is no change in mass.
  - nomel 7y ago
    A system with more potential energy will have more mass [1]. For example, a compressed spring has more mass than a spring at rest, and will take more energy, from an external force, to get to the same velocity. Same with a spinning disk. So unless there's a nice balancing act, with no potential energy change from moving electrons around within the drive, the mass of the drive will change, negligibly, from the potential energy change of the gates. So, the real mass loss of a USB drive is electrons out in addition to potential energy loss from the floating gates. So it's, the world smallest double whammy! [1] https://physics.stackexchange.com/questions/274329/does-potential-energy-of-an-object-increases-its-relativistic-mass/274680 https://physics.stackexchange.com/questions/274329/does-pote...