Slightly OT. Heat and a Bench Light ...

[Jim Adney]

So, if we believe the rating stamped on both the bulb itself, and its box,
and you are prepared to believe me when I tell you that with this bulb
fitted, the temperature of the shade was a whole heap hotter, then
somewhere, there must be another explanation than the two that you believe
are the only possibilities.

I believe everything you've told us about the bulbs. I have no reason
not to, but 60 Watts is still just 60 Watts.

There's no way to get more heat out of it than that. So either the old
60 was actually less than 60 or the new 60 is more than 60 (or both.)
Or you're mistaken about the new ones running hotter, which I agree is
the less likely option.

One other possibility occurred to me, and that is that your socket
chose this moment to become resistive, which could have raised the
total power above 60 Watts and concentrated that extra power
dissipation in the socket - consistent with your observations. I'm not
familiar with the socket you describe, so maybe you can judge the
likelyhood of this. I suspect that if this had happened, the light
output would have been affected and you would have noticed.

-

 

[N_Cook]

On Fri, 4 Jul 2008 01:23:49 +0100 "Arfa Daily"



Separate from this is heat conduction thru the base of the bulb. If
the bulb is shorter and hotter, then it will likely conduct more heat
into the socket. Halogen sockets are generally ceramic or some rather
special high temp plastic in order to deal with this, so putting a
halogen bulb in a standard socket will always result in a destroyed
socket.

-

I'd not appreciated that added effect with Ro80 "Apollo lander" shape bulbs.
The filament is perhaps twice the distance from the socket as the same
wattage but globular lamp. As well as reflecting forwards most of the
rearward-going heat and light.

 

[Arfa Daily]

I believe everything you've told us about the bulbs. I have no reason
not to, but 60 Watts is still just 60 Watts.

There's no way to get more heat out of it than that. So either the old
60 was actually less than 60 or the new 60 is more than 60 (or both.)
Or you're mistaken about the new ones running hotter, which I agree is
the less likely option.

Presumably though, temperature will rise, both at the glass surface and the
surrounding air, if that heat is not being radiated away as quickly as with
the older design ?

One other possibility occurred to me, and that is that your socket
chose this moment to become resistive, which could have raised the
total power above 60 Watts and concentrated that extra power
dissipation in the socket - consistent with your observations. I'm not
familiar with the socket you describe, so maybe you can judge the
likelyhood of this. I suspect that if this had happened, the light
output would have been affected and you would have noticed.

-

Plus I did say earlier that the socket was replaced fairly recently, and is
working just fine again now without one of those bulbs in it.

Here is a picture of the actual item so you will now know for ever, what a
UK bayonet cap looks like !

http://www.comparestoreprices.co.uk...yonet-cap-lightbulb-glass-pearl-pack-of-4.jpg

and here's one on a CFL. which shows the connection pads very clearly.

http://upload.wikimedia.org/wikipedia/commons/b/bc/Lightbulb-Bayonet-Cap.jpg

The B&Q bulb got hot enough to fry that black resin stuff, and completely
destroy one of the connector pads which are made from (lead-free now, I
guess) solder.

Arfa

 

[Arfa Daily]

I've been following this thread rather loosely for the past several days,
so please pardon me if I repeat previously addressed thoughts.

Arfa, I understand that you've stated that you're using lamps labeled as
60-watt in your fixture. Have you actually measured the power consumed
(for instance, using a Kill-a-Watt or similar meter) by the lamps? It's a
possibility that the lamps have been mislabeled, and you actually have 75-
or 90-watt lamps instead of 60-watt lamps.

Since an operating incandescent lamp is essentially a resistive element,
the power consumed should be RMS power, which is the heating value of the
power. Sixty watts RMS into a resistive element should produce the same
amount of heat, whether it's in a large glass bulb or a small glass bulb.
The temperature of the bulb might be higher in a small lamp because it has
a smaller radiating surface. Power that is concentrated in a small area
will have a higher temperature than the same power that has been spread
over a larger area.

Is the radiating surface area of the new lamps small enough to cause the
increase in temperature that you're experiencing?? I suppose the math
could be done to derive the rise in temperature of your new lamps... would
be interesting to see the results.

My ending thought on the issue is that, IMHO, the telling story would be
in the reading of a power meter in the circuit. If you don't have a power
meter such a the Kill-a-Watt, then you could measure it directly with a
voltmeter and ammeter. V * I = P no need to measure phase since it's
essentially a resistor.

Hi Dave
I haven't measured the actual power consumption of the new bulbs. A few
years ago, I would have been surprised if the figure quoted on a lamp was
not at least within say 5% accuracy, but these days, with the reduction in
integrity of just about everything, and the fact that much manufacturing is
now done in 'emerging' industrial countries, I would be less sure of that.
As to whether the new bulb's envelope is small enough to reduce the
dissipation of the heat by the amount noted, I'm not really a good enough
theoretical physicist to make a call of any real value, but given that this
appears to be the only mechanism by which this could be happening, I would
have to say that is probably the answer.

I don't dispute what has been said about 60 watts being 60 watts, or that
heat and temperature are not the same thing. However, radiation efficiency
is key to the relationship between the two, as we both know. A 60 watt amp
with a small silver heatsink on its output transistors will not get rid of
the unwanted heat anything like as well as if that heatsink is black, and
force air cooled.

For all I know, my bench lamp may be a 'special case', and the fitting of
the smaller bulb might just screw up the dynamics of the air circulation
within the shade. As anyone who knows the Terry's Anglepoise will agree, the
shade is not particularly well ventilated. Perhaps it would benefit from
having a ring of 5mm holes drilled around its top ... ??

Arfa

 

[Arfa Daily]

Does it have one of these labels?
<http://www.energysavingtrust.org.uk/energy_saving_products/other_energy_labels/the_eu_energy_label/>
<http://www.thelightbulbshop.co.uk/EnergyLabels.htm>
If so, what are the numbers? We don't have anything similar in the US
so it would helpful if you could supply the numbers for an "ordinary"
light 60w bulb. Note that halogen are in class "D" while ordinary
bulbs are "E" or "F".


Could I trouble you to measure the approximate diameters? I want to
calculate the relative surface areas.

The things I do for science !! I just had to rootle thru' the workshop bin
to locate the bulb which caused all the trouble in the first place. It is
clearly marked 60 watts, and has a CE approval also. I can't tell you what
the energy rating letter is, as I don't have any more on the shelf. I have
two other types from the same stable, and one of the boxes has an energy
rating on it, and one doesn't, so I'll have to look next time I am in the
store.

OK. On the diameters. Consider them for all practical purposes, to be a
sphere. The old was 60mm almost exactly, and the new, 50mm almost exactly,
measured with an electronic calliper, so reasonably accurate figures.

Arfa

 

[Arfa Daily]

I'd not appreciated that added effect with Ro80 "Apollo lander" shape
bulbs.
The filament is perhaps twice the distance from the socket as the same
wattage but globular lamp. As well as reflecting forwards most of the
rearward-going heat and light.

The R80 sounds the way to go from all angles. I'll pick one up next time I'm
in B&Q

Arfa

 

[Mr. Land]

Ah, you see, what you really need is something that'll automatically
sense when the lamp is on, and turn on a cooling fan. I have just the
circuit... %^)

 

[Arfa Daily]

Cease complaining. It's your question that I'm trying to answer.
Never mind that I don't know anything about light bulbs.


Methinks there will be a difference in efficiency rating. If true,
that implies that the newer smaller bulb is really a Halogen filled
bulb, even though it's not marked as such.


Surface area of a sphere is 4Pi*r^2
Old = 4 * 3.14 * 3.0^2 = 113.0 sq cm
New = 4 * 3.14 * 2.5^2 = 78.5 sq cm
The new bulb has about 70% of the surface area as the old bulb.
However, I gotta dust off the college thermodyamics texts before I can
figure out the expected temperature rise given identical power
dissipations. I also gotta lookup the IR transmissivity of glass. The
things I do for science...

Quit complainin' !! I'm helping you figure all about light bulbs with my
question ....

Arfa

 

[Arfa Daily]

I'd be inclined to think a bigger difference could be the relative heat
dissipation via conduction through the base vs. convection through the
glass envelope. If the newer bulb has significantly heftier and/or
shorter leads for the filament, the metal base might be getting
significantly hotter and conducting more heat to the socket.

Incandescent bulbs are inefficient enough, regardless of construction,
that any moderate efficiency gains would have no readily observable
impact on heat output--or, put another way, any improvements that would
reduce the heat output by a nontrivial amount would also produce a bulb
that is obviously much brighter. According to Wikipedia, something like
2% of the electrical energy consumed is converted to visible light.

Peter, Andrew. Thanks for your input. Both good thoughts.

Arfa

 

[Arfa Daily]

Cease complaining. It's your question that I'm trying to answer.
Never mind that I don't know anything about light bulbs.


Methinks there will be a difference in efficiency rating. If true,
that implies that the newer smaller bulb is really a Halogen filled
bulb, even though it's not marked as such.


Surface area of a sphere is 4Pi*r^2
Old = 4 * 3.14 * 3.0^2 = 113.0 sq cm
New = 4 * 3.14 * 2.5^2 = 78.5 sq cm
The new bulb has about 70% of the surface area as the old bulb.
However, I gotta dust off the college thermodyamics texts before I can
figure out the expected temperature rise given identical power
dissipations. I also gotta lookup the IR transmissivity of glass. The
things I do for science...

I was in the store today, and had a look at the boxes these bulbs are in.
There is no mention at all of them being a halogen type, but all of the ones
that are genuinely halogens, do say so. The energy rating on the box was "E"

Arfa