I'm confused about ohms law

[MrAl]

I cannot see why, when talking about the resistance of semiconductors materials
its deemed necessary to explain in terms of semiconductor devices.

A piece of semiconductor material is passive, a semiconductor device ie a transistor/diode is formed
from the junction of two dissimilar semiconductor materials.

Conductors have a temperature/resistance coefficient and their resistance changes in a linear way with respect to temperature,
either in a negative or positive sense.
Manufactured alloys can be made that have close to zero temperature coefficients.

Manufactured compounds can be made that have a non linear resistance change with change in temperature,
again with a negative or positive sense.

Which ever type of material you choose, conductor or semi-conductor, it will obey Ohms Law at any given temperature.
The same rule applies to thermistor compounds.

You can say that a thermistor obeys Ohm's Law at a given temperature
the same way you can say a potentiometer obeys Ohm's Law at a
given rotation, but describing it that way misses the whole point
of having that given element in the first place.

For example, if a thermistor obeys Ohm's Law then try replacing
all the resistors in a TV set with thermistors and see what happens
once the set heats up a little.

A diode is a good example of a semiconductor (yes, with a junction)
because it is often the first thing looked at when talking about
the issues that come up with devices like this.

 

[MikeMl]

...
Strictly speaking, semiconductors (like si diodes) do NOT obey Ohm's Law. This is
because the current through the element is not proportional to the voltage across it.
It's as simple as that. The voltage across a regular diode is around 0.7 volts and
as the current varies widely the voltage across it does not change that much.
It doesnt matter how small of a range you choose (a change of 1 amp, a change
of 0.1 amp, a change of 0.000001 amp) it still does NOT obey Ohm's Law.

An element that obeys Ohm's Law follows this equation exactly:
V=I*R, or similar.

A semiconductor diode does not follow this Law, therefore it does not obey
Ohm's Law...

I said this about 15 posts ago.

 

[Sceadwian]

EVERYTHING follows ohms law, just because the voltage/current/temperature of a device produce effects that change the materials effective resistance doesn't mean it breaks the law, it just means it's an active device. Every device is in actuality (meaning the real world) an active device.

Simply put, at any given instant the values for ohms law are absolute, they can NOT be violated without breaking the first law of thermodynamics.

Semi conductor junctions follow this same exact law. The materials effective resistance however is dynamically controlled by the existing electro magnetic field. This does in NO way affirm that it breaks ohms law, only that that effective V/I or R are dynamically related to one another.

 

[MikeMl]

EVERYTHING follows ohms law...

Look at this equation in the attachment which describes the behavior of an ideal diode:

Do you see any reference to "Ohm's law" or even "resistance"?

All of Ohm's original work had to do with what we call linear resistors.

 

[Sceadwian]

Mike, read my post again. I said effective resistance. The math doesn't prove the real world wrong. At any given instant period of time the equivalent V I R HAS to work out.

The materials effective resistance however is dynamically controlled

The dynamically controlled bit, at least as far as an ideal diode goes is determined by the equation you listed. Which can be turned into VIR curves that will EXACTLY match ohms law.

 

[MrAl]

Mike, read my post again. I said effective resistance. The math doesn't prove the real world wrong. At any given instant period of time the equivalent V I R HAS to work out.


The dynamically controlled bit, at least as far as an ideal diode goes is determined by the equation you listed. Which can be turned into VIR curves that will EXACTLY match ohms law.

Hi again,


Sorry to disagree here, but it does not matter how you describe the
resistance ('effective resistance' or the more common 'dynamic resistance')
it still does not obey Ohm's Law. I believe you are thinking about the
idea that when you bias a diode at a given point, you can APPROXIMATE
the resistance at points close to that region using the same equation
that is used to describe Ohm's Law. That still doesnt mean that it
obeys Ohm's Law however. You might say that it approximates Ohm's
Law at the given point, but it is not Ohm's Law.

Ohm's Law really requires proportionality, and you can not have that
without a straight line characteristic curve, and that curve must pass
through the origin (point 0,0).

Saying that a diode pn junction "obeys Ohm's Law" is almost like
saying nothing at all, because it does not help determine anything
else except for the resistance at that one specific point. Also,
it is almost like saying, "The slope on Mount Everest is 25 degrees
at an elevation of 500 feet", which doesnt say anything about what
happens when you take one step down or one step up...does it
drop off quickly just below that point, or rise up suddenly just above
that point? Who knows, because we havent said very much about
how that slope changes. The only time this would help is when
we climbed up to that one point (and hopefully made it if there wasnt
any drastic change in slope that we couldnt cope with).
On the other hand, if the slope obeyed a law of proportionality, we
would know the constant (the slope 'm', or the electrical 'R')
and we could simply multiply that m by the distance from say the edge
and we would know the slope at that point too. We would then know
the slope anywhere on the mountain side.

Also, if you look up Ohm's Law on the web you will see the word
'proportional' and that word intrinsically implies more than one value
not just a single point. This is what sets the definition apart from a
single point definition of some sort...that one word nails it.

Another way to describe the idea of some resistance at some point is
to say that the curve has been piecewise linearized, but again that
is not that the diode obeys Ohm's Law, it's just that we can approximate
it within some given range that way.

 

[MikeMl]

Thank you, Mr Al. You very eloquently stated what I was trying to say.

 

[Tesla23]

EVERYTHING follows ohms law, just because the voltage/current/temperature of a device produce effects that change the materials effective resistance doesn't mean it breaks the law, it just means it's an active device. Every device is in actuality (meaning the real world) an active device.

Simply put, at any given instant the values for ohms law are absolute, they can NOT be violated without breaking the first law of thermodynamics.

Semi conductor junctions follow this same exact law. The materials effective resistance however is dynamically controlled by the existing electro magnetic field. This does in NO way affirm that it breaks ohms law, only that that effective V/I or R are dynamically related to one another.

Now this is getting silly. Ohms Law is a useful approximation that over a limited range of voltages and currents an ohmic device shows V = IR. Similarly, for materials, over a limited range of electric field E and current density J, j = σE.

It has absolutely nothing to do with conservation of energy (the first law).

It is a macroscopic relationship, it is never true on a microscopic level when you consider individual charges. When you apply an electric field to a material that contains free charges then they start to move. Individual charges accelerate and generally impact on other parts of the material and are scattered (e.g. in a solid electrons are scattered by the lattice and by impurities), this creates a slowing effect. The result is that on average, under a wide range of types of scattering, the current density is proportional to the applied field. Many things can break this relationship, for example as the field gets larger the charges tend to move faster and the collisions may release more charges (breakdown), the number of charges may be limited (saturation), increased current density may cause magnetic fields that affect the current flow etc...

Simply google on 'non-ohmic' for examples.

Mr Al has covered the behaviour of devices quite well. Non-ohmic is an established term for a device with a non-linear V/I relationship.

 

[Mikebits]

Ohm's Law really requires proportionality, and you can not have that
without a straight line characteristic curve, and that curve must pass
through the origin (point 0,0).

Ohms law does not state linearity. In its most basic form one might presume this. Ohms law is not violated if it R is raised to a degree of power. The basic law is still followed, only the result is different.

V will always be I x R regardless if R has a degree of non linearity. The calculations become much more complicated to be sure, but if you break away all the coefficients and other variables, you end up with V=I x R.

At any given current, the diode will have a resistance, and I am sure this resistance will equal R= E/I. R will be a function of temp and I and as I increases Junction temp increases which accounts for non linearity, but at the end of the day ohms law is not violated.

A PIN diode is a good example of this. It is often considered a variable resistor.

https://www.microsemi.com/micnotes/701.pdf

I support Sceadwians stance on this topic.

 

[eng1]

Just my point of view:
When you apply the formula V = R*I to a diode (or other non-linear devices) you are doing small-signal modelling at a given bias point, a linearization of the I/V curve around the bias point.
The Ohm's law - as I learned it - states that if you apply a voltage between two points of a given conductor and measure the current, the ratio V/I is constant for different values of V. On a I/V chart, if you choose ANY pair of points (V,I), the slope of the resulting segment is always the same.
I second what MrAl and MikeMl said.

 

[Mikebits]

Ohms law still applies on a tangent of a line. Ohms law is not confined to Y=mx+b

 

[Tesla23]

Ohms law does not state linearity. In its most basic form one might presume this. Ohms law is not violated if it R is raised to a degree of power. The basic law is still followed, only the result is different.

I'm sorry but ohm's law does state linearity. The PIN diode that you mention is clearly a non-ohmic device. For many references and supporting material simply google 'non-ohmic device' and you will find many universities teaching that diodes are non-ohmic, and that an ohmic device is one with V=IR. Passing through the origin is fairly important for a passive physical device. Here is the first link google gives you **broken link removed**

You seem to be confusing the dynamic resistance, or small signal resistance of a device as indication of ohmic behaviour. All the small signal resistance does is allow us to remove the DC bias conditions of the device and replace it, for the purposes of an AC analysis, by an ohmic resistor. As long as the ac voltage is small then this may be a reasonable approximation to the curved VI characteristic of the original device.

The PIN diode is remarkably non-ohmic. Not only does the small signal resistance vary with bias current, but it also varies with frequency due to carrier lifetime effects.

 

[MrAl]

Hello again,


[see drawing that follows]

We say that a device follows Ohm's Law when its general characteristic
follows V=I*R, not just because a single point v1=i1*r1 can be calculated.
If we did allow that then we would have to say that everything under the
sun follows Ohm's Law and then it wouldnt be any good to us anymore.
When a devices general characteristic does not follow that law of
proportionality we say that it does not obey Ohm's Law.

I've included a drawing of two unknown curves. The device in Fig. 1
is known to follow Ohm's Law and the point shown is at I=2 amps and
V=1 volt. The question is, what is the voltage when the current
changes to only I=1 amp? The answer is rather simple, V=0.5 amps.
The device in Fig. 2 is known to NOT follow Ohm's Law, and the point
shown is again at I=2 amps and V=1 volt because its dynamic resistance
is the same at that one point as the device in Fig 1. The question now
is, what is the voltage when the current changes to only I=1 amp for
the device in Fig. 2 ?

In both cases we were given these things:
1. The current at one point.
2. The voltage at that same point.
3. Whether or not it follows Ohm's Law or not.

In the case where the device followed Ohm's Law we could calculate
the voltage at every point on the curve knowing just the current.
In the case where the device did not follow Ohm's Law we could
only calculate the voltage at the previously known point.

What is the big difference?

We knew the voltage AND current for one point for both curves, yet
we were only able to calculate other points for the device that
followed Ohm's Law. The big difference then is that one device
followed Ohm's Law and the other device didnt. Thus, the information
provided by knowing that one device follows Ohm's Law tells us
something more about that device that can not be explained knowing
only one point alone.
Now if we were to say that the device in figure 2 also followed
Ohm's Law even though v=i*R for that one point alone (R a constant)
we would *still* not be able to calculate other points for that
curve. Thus, saying that a device who's characteristic does not
follow the law of proportionality follows Ohm's Law doesnt tell
us anything more about the device than we already knew knowing
the original two points...we need a new law or at least an
equation that links I to V for this new curve. In other words,
it wont do any good to say that it follows Ohm's Law.


Perhaps this quote from the University of Montana says it the best:

START QUOTE (University of Montana)
III. Large deviations from Ohm's Law: Some devices are useful because
they deviate so drastically from Ohm's Law. One of these is a diode.
END QUOTE

 

[Mikebits]

Perhaps this quote from the University of Montana says it the best:

START QUOTE (University of Montana)
III. Large deviations from Ohm's Law: Some devices are useful because
they deviate so drastically from Ohm's Law. One of these is a diode.
END QUOTE

But perhaps, if you included more of the quote you cited, we can make a different interpretation. More of your quoted source.

Some devices are useful because they deviate so drastically from Ohm's
Law. One of these is a diode. It has the property that it will only allow current to ow in one direction, and
only if the voltage across it exceeds about 0.6 V.

Pretty sloppy that the professor does not check his own writings. ow vs. flow. At any rate, I think his meaning is that the diode does not follow ohm's law due to the fact that it does not conduct until .6volt, and only conducts in one direction.

To me non ohmic indicates something that does not act like a resistor. And to this I agree, a diode does not. But a diode behavior aside from it's differences of a resistor does however follow ohms law.

 

[dougy83]

My POV:

for V/I = R to hold true, dV/dI = R should also hold true, no?

That said, effective resistance may be calculated for any given voltage and current; ohms is defined as volts/ampere. e.g. V = I * R, where denotes a subscript 'i'. For ohmic devices we may add the constraint, V = I * R.

 

[Leftyretro]

I'm sure glad we've cleared this up for the OP

Lefty

 

[ericgibbs]

I'm sure glad we've cleared this up for the OP

Lefty

hi,
Your 'motto' sums it up quite nicely.

As soon as we attempt to measure resistance with a bench testmeter, we apply a voltage, from the meters internal battery, across the semiconductor device and it becomes active.

So the 'measured resistance' [actually measured current thru the device] becomes a function of the applied meter voltage.

 

[Miles Prower]

After seeing contradictory posts on many forums , and getting differing opinions from my teachers , i'm completely lost with the ohms law....

I have a number of questions. Can someone please answer them for me?

It isn't as complicated as it's being made out. (See attached) The most general form of Ohm's Law takes care of everything. In this, the "backwards 6s" symbol means a partial derivative. In this case, the rate of change of voltage with respect to current. Both voltage and current can change with respect to other things, such as time for AC.

1) Do semiconductors obey Ohms law? and why do they obey/not obey?
if they obey , then why is the drop across a ideal diode constant even when the current in it changes?

Yes, they do obey Ohm's Law. If you have something that produces a constant voltage regardless of current (other examples: Zener diodes, gas discharge tubes, active regulators) then the voltage isn't changing. Thus, R= 0 / delta I= 0R. Ideally, it has zero resistance. Of course, nothing in the real world is ideal, so there is always some resistance, though it'll be pretty small.

By the same token, something that produces a constant current has infinite resistance: R= delta V / 0= INF. Again, no such thing, but a good CCS will have a very high resistance.

2) I heard someone say in another thread "semiconductors obey ohms law , but semiconductor junctions dont". why is it so?

That is wrong

3) A thread in another forum said that liquids dont obey ohms law as ions are the current carriers... true or false?

Wrong again. Doesn't matter what the charge carries are: electrons, positrons, mesons, tau particles, ions of all sorts. You have current, Ohm's Law applies.

4) Is ohm's law a special case of V=IR or is V=IR a special case of ohms law?

See above

5) Is Ohms law a law at all?

Not really

6) Is there any theory in physics which explains why ohms law works the way it does?

This gets into Quantum Physics, and it's not easily explained.

 

[Sceadwian]

I think Mike, Eric, Dougy, and Miles explain what I was trying to say pretty well. Ohms law R=V/I or any of it's derivatives do not require linearity, they are equations based on instantaneous values which at any given point have an equivalent value that WILL work out mathematically.

 

[MrAl]

Hello again,


In short, Ohm's Law is always a ratio, but a ratio is not always Ohm's Law.


Some of you guys are just taking Ohms Law out of context and making up your
own definition, why i cant understand.

Before Ohm came along anyone could calculate at an operating point r=v/i
just because we can calculate a ratio v/i doesnt mean it is 'following' Ohms Law.
Ohms Law is a general statement that describes the general behavior of an
element, just just at one point.
You can calculate a voltage and current ratio for anything under the sun, but
that doesnt mean it follows Ohm's Law. In other words, just because a relationship
exists between voltage and current at one single point that doesnt mean that
same relationship follows at every point.

If i were to accept Ohm's Law for any device then...
Inside my keyboard i am typing on right now there is a voltage potential of about
5vdc powering the chip inside. Outside my house my car has a little voltmeter
that stays on all the time to measure battery voltage, and draws about 1ma of
current. Since i have 5v and 1ma, i guess my co-joined keyboard and car voltmeter
follow Ohm's Law.

Ohm's Law allows us to understand certain types of elements without much
difficulty. Ohm's Law doesnt help us to understand the basic operation of the diode
however because the diode does not follow Ohm's Law.

Just to note, it doesnt help to bring attention to something that doesnt
change the fact that the diode does not follow Ohm's Law. This quote
does not change that:

START QUOTE
Some devices are useful because they deviate so drastically from Ohm's
Law. One of these is a diode. It has the property that it will only allow current to ow in one direction, and only if the voltage across it exceeds about 0.6 V.
END QUOTE

In fact, there is a better 'argument' that would help your case and that is what
is sometimes called the 'ohmic' region of the diode. Funny thing is, this is not
actually part of the ideal diode anyway and it does not perfectly follow Ohm's
Law either. Also, that's only one part of the entire characteristic anyway so
there is no way we can say that the diode follows Ohm's Law.

Again, when we calculate r=v/i we are simply calculating a 'ratio', not Ohm's Law.
It's only when we can calculate R=v/i for a constant R that we can say that it
follows Ohm's Law.

Also note that when we calculate values for a resistor of say 10 ohms we can always
get the voltage knowing only I because V=I*R, and we always know R ahead of time
because it is constant. When we try to do the same with a diode, we can not calculate
V=I*R (Ohm's Law) because we dont know what the heck R is ! Thus, saying a diode
follows Ohm's Law just doesnt do any good at all. If you dont agree, show me one case
where knowing Ohm's Law helps us calculate something we didnt know before for an
element that is non linear (a diode is such a good example).

In short, Ohm's Law is a ratio, but a ratio is not always Ohm's Law.