Curious incandescent resistance measurement
TimR
to match the wattage, as there are three others in the same large
room.
There were no markings on the lamp, whether watts, manufacturer, etc.
So I measured the ohms on a similar lamp with a digital multimeter,
and compared it to other incandescents until I found a match. That
may have been a dumb idea, but I'd never had a bulb like this one
before.
Consistently, this lamp read 10.2 ohms in one direction, and 11.2 in
the other direction when I switched the leads. Other shaped lamps
didn't do that. It can't be a fluke, 'cause I bought it at Radio
Shack! Seriously, is this just an artifact of an inexpensive
multimeter, or is there a reason for this behavior?
Victor Roberts
The most likely reason is that your probes were not making
good contact to one part of the base. While a tiny
fraction of incandescent lamps may include a diode to reduce
the RMS voltage, modern ohmmeters are designed to use a
sense voltage that is too low to turn on diodes, unless the
meter is set to the diode test mode.
--
Vic Roberts
http://www.RobertsResearchInc.com
http://www.cflfacts.com
To reply via e-mail:
replace xxx with vdr in the Reply to: address
or use e-mail address listed at the Web site.
This information is provided for educational purposes only.
It may not be used in any publication or posted on any Web
site without written permission.
Douglas G. Cummins
Victor Roberts wrote:
> On Thu, 31 Dec 2009 05:29:42 -0800 (PST), TimR
> <timot...@aol.com> wrote:
>
>> One of the large globe incandescents in my house burned out. I wanted
>> to match the wattage, as there are three others in the same large
>> room.
>>
>> There were no markings on the lamp, whether watts, manufacturer, etc.
>>
>> So I measured the ohms on a similar lamp with a digital multimeter,
>> and compared it to other incandescents until I found a match. That
>> may have been a dumb idea, but I'd never had a bulb like this one
>> before.
>>
>> Consistently, this lamp read 10.2 ohms in one direction, and 11.2 in
>> the other direction when I switched the leads. Other shaped lamps
>> didn't do that. It can't be a fluke, 'cause I bought it at Radio
>> Shack! Seriously, is this just an artifact of an inexpensive
>> multimeter, or is there a reason for this behavior?
>
> The most likely reason is that your probes were not making
> good contact to one part of the base. While a tiny
> fraction of incandescent lamps may include a diode to reduce
> the RMS voltage, modern ohmmeters are designed to use a
> sense voltage that is too low to turn on diodes, unless the
> meter is set to the diode test mode.
Also, the cold resistance (bulb off) is going to be different from the
warm resistance (bulb on). You'll get a better idea of the bulb's
wattage by using your DMM to measure the current of a similar lit bulb.
Obviously, take care in setting up the test circuit - you probably
will need different leads then the ones you got standard with your DMM.
Failing that, just go to the store and buy several different wattages
and try them all, returning the ones that don't match your existing.
--
Douglas Cummins
Calcoast - ITL
TimR
> Also, the cold resistance (bulb off) is going to be different from the
> warm resistance (bulb on). You'll get a better idea of the bulb's
> wattage by using your DMM to measure the current of a similar lit bulb.
That makes sense, but I did not intend to calculate the wattage by
knowing the resistance and voltage.
I just measured several wattages I had on hand, and went with the
closest. All measurements were cold.
TKM
"TimR" <timot...@aol.com> wrote in message
news:c01f8ede-0a66-442c...@e37g2000yqn.googlegroups.com...
It's a very interesting question -- thanks for bringing it up.
I would think that the wattage of a lamp could be determined from the
resistance of its cold filament. Maybe so, but I couldn't find tables of
such information in any of the historical lighting sources including early
editions of the Illuminating Engineering Society Handbook. What is
published is the ratio of hot to cold resistance values, so you would have
to work backwards knowing the voltage and current rating of specific lamps.
Maybe that's because there are several filament variations for the same
wattage lamp that involve different filament diameters and alloys.
With regard to the different meter readings you saw when you reversed the
leads, I thought that the aging of the tungsten filament might create a
diode via migration of the tungsten onto the lead wires; but there's no
mention of that in the reference literature. So, I don't have any other
explanation for the different readings other than what Vic has suggested.
Terry McGowan
TimR
> "TimR" <timothy...@aol.com> wrote in message
I will note that on some bulbs I had great difficulty getting
consistent results. The probes have sharp points, and while on a few
bulbs location didn't matter, on others it changed the resistance
reading by large amounts - in the neighborhood of 50 ohms difference.
I considered getting some fine sandpaper and polishing both bulb and
probe, but in the end didn't bother as real accuracy here wasn't
critical.
I'm intrigued by the comment that the resistance may not be directly
proportional to the wattage of the lamp.
What, then, determines whether a lamp is a 60 W or a 100 W? I assumed
that the filament material was the same, and the only real difference
was additional resistance to limit the current on the 60 W to less
than that on the 100 W. Is that not true?
Victor Roberts
The hot resistance of the lamp is related to the power as:
Rh = (V^2)/P
Tungsten has a rather large change in resistance between
room temperature and its normal operating temperature; the
resistivity of pure tungsten increases by a factor of 14.34
between 300K and 2700K. If you know the operating
temperature, you can theoretically measure Rc, use that to
calculate Rh, and then use that to determine the power:
P = (V^2)/Rh
One issue that complicates this calculation is the fact that
different wattage lamps, and even lamps of the same wattage
that have different designs, operate at different filament
temperatures.
Another issue that complicates the calculation is that part
of the cold resistance is due to the resistance of the leads
inside the lamp, the resistance of the clamp connections
between the support leads and the filament itself, and the
contact resistance of the probes used on the meter, as you
have found. (Four-point Kelvin probes are usually used for
this measurement to remove the contact resistance problem.)
Finally, the 14.43 factor is for pure tungsten, and the
tungsten used for lamp filaments is often doped with other
elements, though I don't know if these other materials are
present in high enough quantities to effect the tungsten
resistance change with temperature.
Determining the cold resistance of the filament with high
accuracy is often rather difficult, which can make the final
power calculation inaccurate.
TimR
I can see that my assumption that resistance determines watts was too
simplistic.
The other thing I tried was to compare several replacement bulbs
visually. I didn't have much luck with that, because these globes
are about five inches in diameter. As the replacements I had handy
were all small, the size effects seemed to confuse my eyes. Low
wattage lamps that are physically smaller seemed brighter, but didn't
seem to light objects nearby as well.
Don Klipstein
TimR wrote:
I would blame low-level diode phenomena between dissimilar materials,
and thermocouple effects between dissimilar materials if the temperature
is not the same at every junction. (Such as warmer than room temperature
when you place ohmmeter leads against shell and tip of the lamp - while
shell and tip are likely made of different metals.)
I do suspect some junctions, such as merely crimped ones, are unreliable
at being "clean connections" at millivolts to maybe ballpark volt range
and can have some diode effect. However, such junctions have an
impressive track record at showing themselves to be good once applied
voltage is high enough to force through at least a few 10's of milliamps.
(By any chance do you know what the voltage across the lamp is when it
is reading 10.2-11.2 ohms?)
- Don Klipstein (d...@misty.com)