Panel & dial illumination bulbs - specs

[julie_m]

Yes; but if you are giving unrectified 50Hz AC to the lamp, it is getting 50 crests and 50 troughs = 100 pulses every second.

[G8HQP Dave]

Yes, 50Hz AC gives 100Hz heating (plus a DC component).

[emeritus]

I can't claim any particular expertise in lamp design, but I did acquire a number of specialist books relating to incandescent lamp technology following a library clear-out after GEC had sold its Osram lamp business. These span the period from 1912 to the 1980's.

The only reference I have found to Tungsten lamps behaving differently on AC and DC is in a 1936 book on Tungsten. This describes a problem that affected early designs, and was soon solved.

It seems that, originally, pure Tungsten was employed for filaments, but they often failed due to the formation of crystals in the filament which gave it an irregular step-like structure. The consequential "necking" at the step transitions gave rise to localised heating and early failure, and the phenomenon was worse with AC than DC. However, by the early 1920's the problem had been solved by adding impurities to the Tungsten. (The only other mention of lamps with different lives with AC and DC was with the long obsolete Tantalum filament lamps, which only gave satisfactory service with DC.)

Strangely enough, awareness of this failure mechanism of Tungsten filaments, and how to fix it, seems to have become lost. Moving on some 50 years, the "International Lamps Ltd." catalogue of 1986 refers to a "filament notching" failure mechanism that had been observed in "recent years" (see "Total Life") that sounds suspiciously like the one exhibited by the early pure Tungsten filaments!

[stevehertz]

Originally Posted by vidjoman

It’s quite common for 7 or 8 volt bulbs in Japanese amps running on 6.5 volt rails in valve amps back in the 60’s. Gave enough brightness and a good long life.

Saying that, I have an instance I'm working on now where they are running a 6v bulb off a 7.5v supply - a 25% overrun. The reason is, in a lot of these 70s hifi components, especially something like a large receiver, there were lots of bulbs used to indicate the various and many functions and also for dial illumination purposes. It seems these would be chosen for their format and size, and not so much conformity of their supply requirements. The result being a need for multiple supplies of different voltages to power all these different lamps. That would increase circuit complexity and cost, so instead it appears they winged it and connected bulbs of different voltages to the same supply. I mean, in the case of overrun bulbs, the simple addition of a series dropper resistor would do the trick, but often they didn't do it.

[broadgage]

SOME types of 6 volt lamp will survive well on the 7 volts.
If the lamps were intended for vehicle use, then by convention they are called "6 volt" but are in fact designed for about 7 volts, that being the average voltage when the engine is running and the battery is on charge.

Vehicle lamps are also designed to withstand the vibration from the engine and from poor roads, and will last longer in the more benign conditions of audio equipment.

In industrial control panels with 6 volt pilot lamps, I always preferred "Lucas" branded lamps as I knew these to be designed for 7 volts and therefore to be longer lasting on 6 volts.

[TrevorG3VLF]

Originally Posted by ColinTheAmpMan1

Are you sure about that? I was under the impression that normal AC in the UK was 50Hz. If this is half-wave rectified, you still get 50Hz I thought. Full-wave rectification results in 100Hz, surely?
Colin.

The heat is generated on the positve and negative half cycle so 100 pulses on 50Hz. When rectified you only get 50 pulses with a delay between them.