Ohm's Law is a great tool for noobies, and it still works for the intellectually masochistic, right after they know the exceptions, which is their nature.
Today, we have a noobie.
Perhaps we should have a Thread in Off Topic for the Math Majors, Physics Majors, History Buffs, and those with Obsessive Compulsive Disorder to discuss this.
Then we could just provide a link to that Thread so the noobies can ignore it.

 

Neither is an ohmic device.

@wayneh is confused as to what constitutes Ohm's Law.

I concede the point, sort of. Ohms law requires R to be independent of I. My bad. I say "sort of" because in a filament, I would argue that R is indeed independent of current, if you could control temperature (for instance by immersion of the filament in a liquid bath). Of course in a lightbulb, temperature is a function of current and thus resistance is also.

Apologies to @#12 and the noobie.

 

Apologies to @#12 and the noobie.

This argument has erupted since before I signed up on any website, and it continues.
I think I described it fairly well several years ago when I said, "It is not my purpose to write the definitive, all inclusive, perfect description of electricity. This is nothing more than a practical method of starting to learn how to manipulate electricity."

It is clear that many people have an interest in the, "correct" interpretation of (maybe it's not) Ohm's Law.
I think the right thing to do is give them a place to argue their beliefs from, "whodunit" to, "What does it mean?"
That might reduce the number of noobs that get drenched in theory, beliefs, and sometimes Calculus that they can't comprehend and which won't make their LEDs work properly.

After taking time to steam some broccoli, I thought, "This epic and historically significant examination could use a "pinned" Thread so it has the celebrity status it deserves." We can give the noobies a link to where it resides regally beside, "Decoupling and Bypass Capacitors".

Now I'm going to report my own post so the Mods can consider whether this is a real problem and whether they can do anything to help.

 

@#12 In many contexts you make a very valid point, but I think in this particular case it doesn't apply because the heart of the issue that this "noob" is having is making the very common mistake made by many noobs that Ohm's Law always applies to everything. Thus pointing out that it doesn't, and why, is directly addressing their specific problem.

 

Understanding the limitations of the instrumentation is yet another problem.

Test leads can add about 0.5 ohms of resistance to what your measuring. The same goes for measuring current.
There's a voltage burden associated with the meter. So, if you measure the current and the voltage with the same meter, it's not the same circuit. When measuring current, the current is right, but the voltage you got before you added the current meter isn't identical.

You can simultaneously measure the voltage across a device and the current through the device and get the right answer UNTIL the resistance of your measuring instrument (Voltmeter typically 10 M ohms) starts to dominate.

 

Understanding the limitations of the instrumentation is yet another problem.

I remember when learning the limitations of my meter was a months long process. I probably could have run the math and learned most of it in a day, but I was busy working on fairly crude circuits and the meter errors didn't seem important right then. You can get practical measurements on your first day, but you have to learn the interaction of the meter and leads before you know how well you can trust your meter and which measurements are most prone to error. If you don't have auto-zero, the low ohms scale is the worst. Trying to get exact correlation with two measurements and Ohm's Law was described by KISS in post #13. You must develop some skills before you can get that method to work pretty well to three significant figures, and you can't even get that far without a 1% guaranteed meter.

If you're working with old vacuum tube TVs that had 20% resistors (like I did) or just checking to see if a motor has most of 240 VAC, almost any meter will do. If you're trying to nail down the junction of theory and practice, you have to consider all the variables, and one of them is your meter.

 

Ohms law does indeed apply, that's why it's called a law. Ignorance of what R is does not negate Ohms law. It just make it harder to apply.

Hi,

I read both of your posts (yours and the member you replied to) and sorry to say i dont think you are looking at this correctly.
Something can be resistive and not follow Ohm's Law.

Ohm's Law is a very specific statement about resistance that narrows down the cases to those where the resistance is constant. The term "ohmic" is a very unfortunate choice of words because resistance is measured in Ohms yet not every resistance is ohmic. So when we say, "ohmic" it sounds like we are saying, "it has Ohms", when really we are saying, "It follows Ohm's Law". Unfortunately sometimes when we say 'ohmic' we really mean 'resistive' too, and then we have to know the context of the discussion.

Strictly stated mathematically we would have:
resistance: v=i*r (v,i,r can all vary)
Ohmic: v=i*R (v and i can vary, but R can not vary)

We also have an assumed range of operating conditions that we take to be true. For example, when we say that a copper wire is ohmic we usually mean that when the temperature is fairly constant it is ohmic. We hold these kind of assumptions because if we didnt, we'd have to explain every single operating case that could possibly happen, and that would mean nothing under the sun is "ohmic". For example, if we dont assume the wire is at nearly constant temperature, then what about when it gets up to over 1000 degrees and starts to melt, and then later melts so bad that the liquid metal starts to drip from the wire causing a break in the wire. Then the resistance goes up so high it's getting near infinite. So we start with maybe 1 ohm and end up with an infinite resistance, which does not help much in most applications. So we have an assumed operating range, and investigate that only when we need to.

But the most important aspect is being able to distinguish between what is resistive and what is ohmic. Something ohmic is also resistive, but something resistive is not always ohmic.

A light bulb is resistive, but it is not ohmic because it's resistance varies greatly over the normal operating range. In fact, the light bulb has an exponential response whereas Ohm's Law is a perfectly linear response.
A silicon diode is also resistive, but is not ohmic. We can "linearize" the diode curve over a small operating range and then see something that resembles an ohmic device, but we can not classify the ENTIRE device as ohmic because over the normal range of operation it is not ohmic, and that normal range is much more than a tiny change in current.

BTW the difference between the wire and the diode is that the wire more closely follows Ohm's Law over the normal operating conditions, while the diode rarely follows Ohm's Law in an application. So before we class anything as ohmic or non ohmic we have to know the overall character of the device, not just a small operating region, unless that appears in the typical applications. Thus a wire that is rated for 10 amps in a given application may change resistance slightly over the full operating range of 0 to 10 amps and still be thought of as ohmic, while a diode that runs over the full range of 0 to 10 amps will change resistance by a huge amount, possibly 0.1 Ohms to 100 megohms. So to this end we usually allow the resistance R to vary slightly over the full intended operating range and still call it ohmic:
v=i*R with R approximately constant over the full operating range.

Note also that every device has some inductance simply because it has length and when current travels over a distance it always encounters inductance, yet we dont consider that for many devices unless we really have to. So resistance is considered non reactive, and an ohmic device is considered non reactive also, until such time as it becomes significant in the application.

So in short we have:
Ohmic: v=i*R which requires R to be nearly constant and thus is a straight line, and anything else is non ohmic:

 

Ohm's Law is a great tool for noobies, and it still works for the intellectually masochistic, right after they know the exceptions, which is their nature.
Today, we have a noobie.
Perhaps we should have a Thread in Off Topic for the Math Majors, Physics Majors, History Buffs, and those with Obsessive Compulsive Disorder to discuss this.
Then we could just provide a link to that Thread so the noobies can ignore it.

Resistance is futile, that's the law.

 

A light bulb is resistive, but it is not ohmic because it's resistance varies greatly over the normal operating range.

Not true, if you control temperature. The resistivity of a tungsten filament is a function of temperature, not current, and therefore it is indeed constant (at a given temperature) and therefore ohmic.

But I've already conceded the point as pedantic versus practical. Nobody controls the temperature of a lightbulb filament, and so resistance is effectively a function of current because of the effect of current on temperature. A lightbulb behaves as a non-ohmic device, as does a motor.

The mile-high point to the TS is that some things behave differently when operating than they do when cold on the lab bench.

 

Nobody controls the temperature of a lightbulb filament, and so resistance

Call me "Nobody" then, That was my best way of testing a PID temperature controller. A thermocouple on top of a 500 Watt incandescent lamp.

Reminds me when I was with some friends on a service call (occasionally I tagged along).

I said "Nobody's home"

He looked at me funny. Then I pointed to the vanity tag on the car parked in the driveway "Nobody".

 

Call me "Nobody" then, That was my best way of testing a PID temperature controller. A thermocouple on top of a 500 Watt incandescent lamp.

Different critter. wayneh is talking about controlling the temperature of a light bulb filament separate from the current flowing through it so that you can change the current without changing the temperature, and hence maintain the same filament resistance regardless of current.

 

Conceivably you could maintain the filament temperature fairly constant with a change in voltage by immersing the filament in stirred oil or water (up to some limit).

 

Conceivably you could maintain the filament temperature fairly constant with a change in voltage by immersing the filament in stirred oil or water (up to some limit).

It would be EXTREMELY challenging. Consider the volume of a filament. Now consider that you've got to get dozens (or hundreds) of watts of heat out of that volume without letting the temperature raise by any appreciable amount -- perhaps not even a degree or so (Celsius or Fahrenheit, take your pick). You'd probably need a pretty fast flowing cooling fluid that had a lot of volume in a big heat exchanger elsewhere in the circuit. But, yeah, if you REALLY needed to do it, it could be done.

 

Well I was thinking of a low power light bulb of perhaps a few watts, and not necessarily operating up to the the rated voltage, to keep the heat removal requirements reasonable.
Just as a demo.

 

Well I was thinking of a low power light bulb of perhaps a few watts, and not necessarily operating up to the the rated voltage, to keep the heat removal requirements reasonable.
Just as a demo.

I worked in the lightbulb industry for several years, and like Bahn's just said, it would be unbelievably complicated... the filament would have to be cooled directly by whatever fluid you chose to use. Imagine the challenge of pumping the fluid at the perfect rate needed to keep the filament's temperature stable!

 

Imagine the challenge of pumping the fluid at the perfect rate needed to keep the filament's temperature stable!

Why does it have to be perfect?
The flow just has to be fast enough to keep the filament from appreciably rising in temperature above the liquid temperature.
I'm not talking about running the bulb at full current but just enough to demonstrate that the rise in resistance is not due to current but to temperature.

 

Why does it have to be perfect?
The flow just has to be fast enough to keep the filament from appreciably rising in temperature above the liquid temperature.
I'm not talking about running the bulb at full current but just enough to demonstrate that the rise in resistance is not due to current but to temperature.

Guess I fell for the perfectionist trap... I was thinking about setup in which a wide range of current could be ran through the thing, while trying to maintain the temp at an arbitrary value within ±1°C

 

Not true, if you control temperature. The resistivity of a tungsten filament is a function of temperature, not current, and therefore it is indeed constant (at a given temperature) and therefore ohmic.

But I've already conceded the point as pedantic versus practical. Nobody controls the temperature of a lightbulb filament, and so resistance is effectively a function of current because of the effect of current on temperature. A lightbulb behaves as a non-ohmic device, as does a motor.

The mile-high point to the TS is that some things behave differently when operating than they do when cold on the lab bench.

Hi,

So i guess if the filament breaks then we just have someone come in and hopefully weld it back together so it keeps being "ohmic" ? Then we can say it is still ohmic even if the filament breaks.

I see you are still not looking at this correctly. You're trying to take something that in normal operation is non ohmic and force it to be ohmic, and you have not even suggested how this might be done. "IF" is a very big word. I think it is a good exercise in thought though, but you may misled others into thinking the forced version is the same as the non forced version, and it looks like this may have already happened. Still, it's an interesting sub point and so i believe it is good that you brought this up.

Let me reiterate my previous statement:
MrAL:"A light bulb is resistive, but it is not ohmic because it's resistance varies greatly over the normal operating range."

Now let me condense and paraphrase your statements:
wayne: "A tungsten wire can be made ohmic by controlling it's temperature."

Now my questions are:
What is the temperature of the filament of a 120 volt 100 watt light bulb running at 100 volts when the surface of the glass envelope is cooled to within 1K of absolute zero?
What does the temperature of the filament change to when the voltage is increased to 120 volts (maintaining 1K glass envelope temperature)?
And i ask again, is the light bulb ohmic or non ohmic?

So you see the light bulb is non ohmic because that's the way it is usually used, but also we find that it's not possible to control the temperature of the wire inside the light bulb because there may be operating conditions that do not allow this to work. A tungsten wire is not a light bulb. We may or may not be able to control the temperature of a wire of any metal, but that does not change the fact that the light bulb (normal incandescent type) is non ohmic and you cant force it to be ohmic without extreme modification, in which case it is no longer a light bulb so a statement about light bulbs does not have to apply. A tungsten wire is an extreme modification so it is no longer a light bulb.

 

So by that definition a copper wire is non-ohmic since its resistance changes with temperature and a current through the wire will change its resistance.

 

So by that definition a copper wire is non-ohmic since its resistance changes with temperature and a current through the wire will change its resistance.

AFAIK the conductivity of all materials is affected by temperature ... I guess that means that there's no ideal "ohmic" material in the real world