I have a 100W GE mercury vapor lamp that has a strange feature. The arc tube has a silver-colored coating on the outside of either end of the arc tube. I have searched the gallery for other clear mercury vapor lamps and haven't seen other lamps that look like this. I only see the white coating on metal halide lamps and no coating on mercury vapor lamps.

I have attached images of the etch, the lamp innards, and a close up of just the one end of the arc tube.



What is the purpose of this? Has anyone seen this before? My guess is that it just keeps the electrodes warmer so they last longer, but IDK.

Thanks!

 

I have a GE lamp made in Mexico that has a similar silver coating. I haven't seen this on Sylvania lamps so I wonder if this is exclusive to GE lamps.

 Apparantly it has something to do with thermal losses . See here : https://www.lighting-gallery.net/gallery/displayimage.php?album=5200&pos=150&pid=158914

The big seal flanges are sucking a lot of heat away from the region behind the electrodes, forming there a cold spot. The reflective coating is supposed to prevent that (it reduces the emission coefficient of the arctube in that area, so reduces thermal radiation from there).
It is practically on all 50W MV's, but maybe GE decided it is worth here as well...

The thing is, the MV arc is the most efficient when the arc is fed by 50W/cm of its length or more, lower loading yields drop in efficacy. That means the lower rated power the lamp is, the shorter the arc needs to be.
But at the same time the arc needs to have sufficient voltage drop, so the 10..15V cathode losses do not form that much percentage, practically all MV's ended up with about 100V arc.
So lower power lamps need to have the same voltage over shorter arc, so the voltage drop per unit length must be higher.
And higher voltage drop per unit length means you need higher operating pressure.
And to get higher pressure means the gas vs liquid transition happens at higher temperature, so you need to make sure the coldest spot on the arctube (the area behind the electrodes) runs hotter on lower power lamps. That may lead to use vacuum outer and on mainly the lowest power lamps it is done that way.
But at he same time you have to make sure the hottest part (which uses to be the center section of the arctube, mainly the part above the arc when burned horizontally) does not exceed the limit the quartz can handle, so you need to ensure sufficient cooling there. And the only cooling available without also blocking the light is gas flow around the arctube, so gas fill in the outer.
But the last two aspects are exactly in their opposite, so you have to find some measures to squeeze between. On top of that you are pretty limited in outer pressure selection: When gas filled, you need sufficient pressure to prevent discharges there. Or you need to use really vacuum. You can not go "somewhere between", as that would create tendencies for arcovers in the outer.
All these aspects you may to some extend control by the way how the lamp assembly is designed, but evrything has its downsides (less material left in the seal pinches mean better heat retention but structurally weaker assembly, careful inner wiring and armature arrangement means you may afford lower outer fill pressure and still do not suffer from arcing, larger arctube diameter leads to colder running arctube walls but more mercury amount so longer warmup snd even more complex tapered shape to prevent the end sections from running too cold), the reflective paint is one of them (allows the seals to be bulkier so stronger).

On really the lowest power lamps all such measures, include the coating are practically a must have, along with vacuum outer and so on. The higher you go with the power, the more freedom you have, even up to the (oldest MV) 400W sufficing with hard glass instead of quartz and yet losing only about 20% of light output compare to the modern high pressure designs.

I thought only 40-50W MV lamps have heat reflective coatings at the ends of the arctube.

Interesting why a somewhat more expensive noble metal coating is preferred in mercury lamps instead of that ubiquitous white flame sprayed zirconia we are all used to see in MH lamps?

Just came to my mind: Is it really more expensive? Yes, gold is more expensive material per mass, but isn't there so little amount of it that the process of applying it becomes actually more expensive than the used material itself. So isn't it possible that in the scope of MV manufacturing, the metal coating application may become actually cheaper? The MH may need to go for the Zirconia, as an easy to deposit metals won't srvive the higher temperatures of MH.

But this above is just entertaining a hypothesis which at the end could be wrong, I'm not expert in these lamp making steps...


What I know, for decades gold has been and used to for connection wires for semiconductor devices because it for long term was the cheapest way, even when using one of the most expensive metal for this. So I just tried to apply the same reasoning...

Well that sounds reasonable, as the mass of precious metal in one lamp may be in milligram or even fraction of milligram range. Still may be some hidden costs of dealing with precious metals at the plant like accounting for them and making sure no one steals them. I an pretty sure no one interested in zirconia personally though

One of the main reasons for the metallic heat reflectors on mercury arc tubes is to cause faster run-up, and it has a significant effect on extending life.

The most stressful period for a mercury lamp is during run-up while the arc voltage is low, current is therefore high, and the electrodes run much hotter than normal.  That causes accelerated consumption of the double oxide/tungstate emitter coating, and when that is all gone the lamp reaches end of life.  Slowing the rate of consumption extends life and improves lumen maintenance.

On the low wattage lamps it is helpful because their arc tubes are small and lose a lot of heat to the surroundings, and are slow to run up.  On the high wattage lamps it is also beneficial because the great thickness of the quartz acts as a heat sink and also slows run up.  However from about 80-400W the costing offers little benefit. 

It is not so much a sign quality if one lamp has this coating and another not, because different techniques can be applied to achieve a similar result.  For instance, other manufacturers fill the outer bulb with a gas filling having lower thermal conductivity (ie higher atomic weight and/or lower gas pressure), along with different arc tube dimensions.

Metallic coatings can only be used on mercury lamps, not metal halide types that contain sodium.  The usual materials are gold (Philips and many smaller brands) or platinum (most of the rest).  These are cheap to apply despite the high material cost.  Usually the noble metal is dissolved in thick lavender oil and sold as bright paint by companies like Johnson Matthey and Hereaus.  It can be cheaply applied by paint brush or more usually a pad-printing method.  The coating is reduced to pure metal by heating to around 800C, which is achieved at no extra cost on the ageing wheels that burn in and grade the voltage of mercury arc tubes before outer jacketing.

Metal coatings on a metal halide lamp can give problems of photoemission which gives them an electrical charge which pulls sodium and lithium ions into the quartz and shortens lamp life.  Instead, a non-conductive ceramic coating is usually applied - typically alumina in the Americas and Japan, Titania in Europe, and Chromium oxide in USSR.  Despite the low raw material cost such coatings are far more expensive to apply due to their poor adhesion.  Special binders may have to be added, and applied via dip coating and furnacing, or flame-spray methods.

Good to know.

Also, gold coating is used on that bare burner style UVA-UVB suntanning metal halides. Sure, no photoemission in the air, but I suspect a part of the reason is also that expensive golden look of the lamps