Is it inevitable that bulbs turn black before running out?

My Streamlight TL-3 bulb. I recently felt it not as brightly as before. Then after checking, I found bulb became grey. It should be due to the tungsten vapor attached to the glass.
I do not expect this because I ever saw similar small flashlight bulb as transparent as new one when running out.
So I want to hear your comments about it. And two questions,
1. For my TL-3 bulb, was it because I used 2 16500 lithium-ion instead of 3 CR123A that led to the black?
2. Does halogen bulbs normally remain transparent as new when running out? I assume TL-3 bulb is not halogen, at least not full halogen filled.

the "typical" high powered bulb will when driven with good power
(burn through the wire and that "typically" at the very 1st powerup that day --> cold wire killed from current spike)

Your bulb - underdriven - lasted longer but got the smokes inside instead

The halogen cycle that makes halogen bulbs so bright relies on a high temp inside the glass envelope to keep the vapor from the filament from depositing on the glass envelope, and re-depositing on the filament instead. When running under voltage the temperature doesn't get high enough to keep the cycle going. This allows the metallic vapor to deposit on the glass envelope instead of returning to the filament, resulting in the light not being as bright, and the filament slowly eroding.
Given the cells involved, I have to assume you were just under the optimum voltage to keep the halogen cycle going. What kind of life did you get out of the lamp assembly? Was it close to what you were expecting?

A vacuum or gas filled bulb will normally darken before failure. The tungsten evaporates from the filament and is deposited on the glass bulb, thereby darkening it. If the bulb failed due to overvoltage, mechanical shock, or defective manufacture, without significant use before failure, then the bulb will be clear since little tungsten was evaporated.
The degree of darkening will vary to an extent depending on the wattage of the bulb, presuming the same dimensions. An 8.4 volt 0.75 amp PR base bulb will tend to darken more than a 2.4 volt 0.3 amp bulb in the same size.

A halogen bulb will often remain clear at end of life. The tungsten still evaporates, but it is re-deposited on the filament and therefore does not darken the bulb.

I believe this is a prime example of, "sputtering out".

tbx8192 said:

1. For my TL-3 bulb, was it because I used 2 16500 lithium-ion instead of 3 CR123A that led to the black?

Click to expand...

Pair of lithium Cobalt cells in series won't cause overdrive or blacken TL-3 Xenon bulb. Pair of IMR will.

tbx8192 said:

2. Does halogen bulbs normally remain transparent as new when running out? I assume TL-3 bulb is not halogen, at least not full halogen filled.

Click to expand...

Low power bulbs such as TL-2, TL-3, P60 and P90 are not halogen.

Timothybil said:

The halogen cycle that makes halogen bulbs so bright relies on a high temp inside the glass envelope to keep the vapor from the filament from depositing on the glass envelope, and re-depositing on the filament instead. When running under voltage the temperature doesn't get high enough to keep the cycle going. This allows the metallic vapor to deposit on the glass envelope instead of returning to the filament, resulting in the light not being as bright, and the filament slowly eroding.
Given the cells involved, I have to assume you were just under the optimum voltage to keep the halogen cycle going. What kind of life did you get out of the lamp assembly? Was it close to what you were expecting?

Click to expand...

It is the first TL-3 bulb I used. It has worked for about 10-15 hours. I don't know how long it will last.

fivemega said:

Pair of lithium Cobalt cells in series won't cause overdrive or blacken TL-3 Xenon bulb. Pair of IMR will.

Click to expand...

I used ICR16500 rated at 1100mAH. I saw its 1.2A discharge curve. The on load voltage started from 4.0V. While for the Surefire 123A curve, its peak was at 2.5V. So 2x16500 has 8.0V peak vs 3x123A's 7.5V peak (average voltage both at about 7.2, very close). That's my original thought that two lithium leads to some degree overdrive than 123A.
Why IMR will overdrive the bulb while ICR won't? IMR cell's rated voltage is even lower than ICR (3.6 vs 3.7)?

fivemega said:

Low power bulbs such as TL-2, TL-3, P60 and P90 are not halogen.

Click to expand...

So using 123A also blacken TL-3 bulb? I ever saw similar bi-pin bulb which looked as clear as new but already used for quite long time.

tbx8192 said:

Why IMR will overdrive the bulb while ICR won't?

Click to expand...

IMRs won't drop the voltage as much as ICRs under same load.
For example, pair of IMRs will run the bulb at 8 volt but pair of ICR will do at 7.6 volt.

tbx8192 said:

So using 123A also blacken TL-3 bulb?

Click to expand...

Chance of blackening the Xenon bulb at spec is much less than overdriving.

tbx8192 said:

That's my original thought that two lithium leads to some degree overdrive than 123A.

Click to expand...

Three primary 123A will run TL-3 bulb at normal standard spec.
Two lithium cobalts will slightly overdrive TL-3 bulb but not too much.
Two IMRs will severely overdrive TL-3 bulb.

Please note that all above statements are in general and may not apply on every application.
Also remember that no battery in the world will direct drive any bulb at same voltage throughout the run time.
Let's say average TL-3 bulb life at spec is 30 hours which means 30 sets of batteries.
If you run a new TL-3 bulb only 15 minutes per new set batteries, do you think it will last 120 sets of battery or 30 hours?
Calculated, estimated 30 hours bulb life is average at 7.5 volts.
New sets of primary 123A may run at 7.8 volt for few minutes then will drop to 7.7, 7.6, 7.5 and will end at 6 volts.
This variation is less when using lithium cobalt and much less when using IMRs.