Can you help answer some questions about electrical resistance?

[goc9000]

Hi all !

I have a few questions, mostly about electrical resistance. I'm afraid my pre-college physics training can't help much with any of them, but I was hoping you folks can :)

1) AFAIK, transistors can be used to break/restore the flow of current at will (i.e. act as an extremely high resistance in the first case). My question is: how quickly can *inexpensive* transistors available to the average Joe break/restore the flow ? I'm talking about temporarily interrupting and then restoring current in a circuit about 1000 times per second...

2) With the obvious approximations, what's the maximum voltage that parts of the human body can be subjected to without any obvious effects (tingling, shock, etc.) and without any damage to the tissues even in case of long-term exposure (i.e. when your skin is exposed to the voltage for a few hours, every day)...

3) What's an approximate range of the resistance of the human skin (I know it depends on humidity, etc...), assuming it is reasonably dry (i.e. hasn't just been dipped in water :) )

4) The hardest of all... Suppose we have a rectangular, homogenous, sheet of metal. Point A is somewhere on the surface of the sheet, and point B is in a corner. Am I right to assume that, if a potential difference is created between the points, current will *not* flow in a straight line from A to B (or viceversa), but all along the infinity of paths that join A and B, with varying intensities ? Also, am I right to say that the resistance between A and B isn't a linear function of the length of the straight line between A and B ?

Thanks in advance for any help :)

 

[Locrian]

I'll answer 2 and 3 real quickly if I can,

You can apply huge amounts of voltage without a person making much note, so long as the current is very very small. Those static-electricy generating machines can output tens of thousands of volts, and you barely feel it. However, allow even 100 miliamps to travel through the person, and their heart goes into fibrilations. They'll probably die.

I'm not sure about the human skin... like, if you remove it? I can say that if you measure a human being's resistance from touching their skin, it can be anywhere from 1000 (at the very very low end) to 1 million ohms or more. This is why electricians are always very aware of what they are wearing, how wet they are, and what they are touching; because it all affects their resistance dramatically.

Hope that helped

 

[chroot]

1) Most cheap transistors have settling times on the order of nanoseconds. You can use them to switch in at least the megahertz range. No transistor is going to be bothered by 1 kHz.

2) After much study, the automotive industry decided that anything less than 50V is not generally harmful to humans. I wouldn't want any voltage continually applied to me, though.

3) It varies hugely. The human body model used as an electrostatic discharge test for semiconductor devices is a 100 pF capacitor in series with a 1.5 kilohm resistor. 1 kilohm or so is probably a good guess.

4) Yes, current takes all available paths, even paths that are not straight. The majority of the current, however, flows through the straight path, since it necessarily has the lowest resistance. I believe you are right about the linearity, but I would have to sit down and do a calculation to be totally sure.

- Warren

 

[pervect]

Hi all !

I have a few questions, mostly about electrical resistance. I'm afraid my pre-college physics training can't help much with any of them, but I was hoping you folks can :)

1) AFAIK, transistors can be used to break/restore the flow of current at will (i.e. act as an extremely high resistance in the first case). My question is: how quickly can *inexpensive* transistors available to the average Joe break/restore the flow ? I'm talking about temporarily interrupting and then restoring current in a circuit about 1000 times per second...

That sounds easy. If you use an inexpensive bipolar junction transistor, you'll find that it takes a little while for one to turn off after you saturate it (turn it on very hard). But a mere 1000 hz should not be a big problem.

MOS transistors don't saturate, but they don't conduct as well as bipolars.

2) With the obvious approximations, what's the maximum voltage that parts of the human body can be subjected to without any obvious effects (tingling, shock, etc.) and without any damage to the tissues even in case of long-term exposure (i.e. when your skin is exposed to the voltage for a few hours, every day)...

9v and 12v batteries are fairly safe and won't cause any significant current flow, as long as you don't do something weird like applying the voltage to piercings. The skin is a pretty good barrier which maintains a high resistance to the flow of electric current - if you pierce the skin, the resistance goes down, and you can affect nerves and internal organs a lot more easily.

For long term effects, the electrodes might be more important than the batteries. People who wear "tens" devices for back pain, for instance, should use the right sort of electrode, or the ions might migrate into their skin. I'm not quite sure what the physiological effects of this ion migration are.

I'm not sure what the threshold voltage for feeling a shock is - it's a lot higher than 12 volts, I doubt one could feel 24 volts. One can definitely feel 120 volts, though :-).

As far as safety goes, a milliampere through the heart could be fatal, IIRC.
[/quote]

3) What's an approximate range of the resistance of the human skin (I know it depends on humidity, etc...), assuming it is reasonably dry (i.e. hasn't just been dipped in water :) )

4) The hardest of all... Suppose we have a rectangular, homogenous, sheet of metal. Point A is somewhere on the surface of the sheet, and point B is in a corner. Am I right to assume that, if a potential difference is created between the points, current will *not* flow in a straight line from A to B (or viceversa), but all along the infinity of paths that join A and B, with varying intensities ? Also, am I right to say that the resistance between A and B isn't a linear function of the length of the straight line between A and B ?

Thanks in advance for any help :)

I'm not sure of the last problem, I think that the resistance will be dominated by "contact resistance" terms, which are a bit hard to model. In general, sheets of conductors have a constant "sheet resistance", ohms/square, so if you made wide contacts on both ends of the sheet, one square of the material would have the same reisistance regardless of how big the square was.