I know I will probably never get the opportunity to make HID lamps (or any types of lamps at all), but I certainly do love coming up with ways to do it. Obviously if you are going to make a MH or MV lamp you are gonna need some quartz, and it is gonna have to be manipilated while hot to shape it into the desired shape. I have the following questions:

---

1) Flame Fuels:
I hear that in industrial settings oxy-hydrogen is used for working with quartz. But, I have seen Youtube videos of people making quartz high temperature test tubes with a simple oxy-acetylene torch. ChatGPT says that a hydrocarbon flame has to be oxidizing in order for the quartz to not absorb the excess carbon, but is this really a problem in non-high-precision applications? I know this is a stretch, but oxy-propane has a similar flame temp to oxy-hydrogen (less BTUs, but whatever), would that work? Obviously acetylene is not as accessible as propane.

2) Pinch Seal Details:
I have heard @James say something about mechanical "hammers" for pinch sealing quartz. Maybe that is just the type terminology that is used in the industry, but do you have to be really forceful with the quartz when it is hot? I assumed that the molds for the pinch seals were graphite, but apparently they can also be alumina or other ceramics. Is there a more accessible option? Again I know this is a real stretch, but how about water soaked wood molds (like the ones used for decorative glassblowing)? Obviously graphite isn't that hard to get, but wood is even easier to get (and machine).

4) Shrink Seal Details:
So doing a pinch seal is obviously not super simple. But would doing a shrink seal be easier? As far as I am aware all you have to do it hook it up to vacuum (plug the ends) and then heat it evenly until it collapses on the foil. Am I wrong? This just seems like an easier thing to do than pinch seals. An before you mention it, I definitely do not want to attempt a graded seal. Those intermediate glasses are almost certainly unobtainable.

4) Annealing:
Obviously quartz has a very low CTE, so you would think that annealing wouldn't be super duper necessary, but you have to get it really really hot to work with it so maybe that cancels out(?). Any info about annealing fused quartz? While watching Youtube videos about quartz I didn't see any annealing being done, but maybe that is the wrong thing to do.

---

Thanks!

1) ChatGPT seems absolutely not trustworthy on such topics!  You can successfully work quartz with oxygen and most hydrogenated gases.  In the old days simple oxy-gas was used.  It has a higher flame temperature than oxy-hydrogen, 2776C vs about 2400C, but the latter has far greater heat capacity.  The flow rate of pumping methane into a gas burner is limited by its flame speed of about 2.2 feet per second : if you try to push in a greater flow of gas to get more heat, the flame will blow itself out if the gas speed exceeds the flame speed.  Hydrogen actually has a lower calorific value than methane (fewer bonds being broken and releasing their energy), but it does have a much higher flame speed of about 18 ft/sec.  So you can pump more of it through the burner and because of that the flame contains more heat despite the fact that its temperature is lower.    For that reason, hydrogen soon came to be preferred among lampmakers for working quartz.  It significantly reduces the heating time, and allows smaller burners than can be more precisely aimed.  Moreover due to the greater rate of gas glow it exerts more force on the quartz.  Glass is normally pulled approximately into the desired shape by gravity, but hot quartz doesn’t move unless you push it.  High velocity flames are very useful to help move it before the pinch hammers do the final job,

However, we always think of hydrogen as being a pretty rough and harsh fuel.  It packs in so much heat so fast that it can be a bit tough to control precisely.  The flames also make a hell of a noise, especially when you have large machines with dozens or hundreds of hydrogen fires.  In recent years many of the best lampmakers migrated to oxy-propane on the machines for their highest value and newest quartz lamps.  For instance all of the Sylvania BriteArc lamps developed by my factory since about 2010 ran through oxy-propane burners.  The flames are so much easier to control precisely, even though the heating time may take a little longer.  That gas may be even easier for you to obtain and will help you pump in the heat more rapidly than oxy-gas.

2) We tried dozens of different materials for pinch hammers over the years.  Carbon, graphite etc do not work because those materials are too soft and brittle.  The hammer is just that : a hammer.  You have to whack quartz really hard to make a pinch seal, and also incredibly fast, because it begins to cool and solidify the instant you stop the flames. For many years Naval Brass was the preferred material.  Iron and steel must be avoided because they stick too much.  Copper is too soft and wears out after making a few lamps.  For about the past 20 years the preference has been Elmadur X, a high copper alloy.  Osram discovered that this has about the best properties for reducing sticking and wear and it’s also quite easy to machine, and within a year or two of them sharing that knowledge most lampmakers were using that or some close equivalent.  Graphite was however maintained for some processes that use rollers to gradually push the quartz into shape, for instance on bulb forming or to roll down the diameter of big tubes before then making a narrower pinch-seal.  It does evaporate over time though.  Well actually it sublimes, it reacts with oxygen in the air and then disappears as carbon dioxide.  You can use graphite pencils if you try shrink-seals, that is what I usually use.  But it has to be extremely high density.  Almost everyone bought from Le Carbone in France, an ancient factory that used to be the old national works for making carbon filaments and arc lamp rods but now produces all kinds of high-tech carbon products.

3. Shrink seals are dead simple!  Your assumptions are all correct.  Much easier to make neat looking seals if you have a lathe.

4. Annealing is not so critical but we always did it anyway for high performance lamps.  After sealing in the electrodes but before dosing, they would go into a vacuum furnace at usually about 1800C for somewhere between 6 hours and 24 hours.  Most factories do not bother though because the vac furnaces for such high temperatures are hellish expensive.  I don’t know how much, but remember that every couple of years we had to replace their molybdenum heating elements and those cost something in the region of €100k each.  For your purposes, it would not be necessary.  Also in the factory our main reason for vac furnacing was not so much for annealing as to reduce the concentration of hydroxyl ions in the quartz.  That brought a dramatic improvement in the life of all metal halide types.  With pure mercury arc tubes we had a faster in-line process based on flaming the whole arc tube in soft bushy hydrogen flames while blowing argon or hydrogen into the lamp bodies.

@James
Thank you so much for your very detailed reply! I am very surprised to hear that oxy-propane was used in the industry, i would assume they would use natural gas or acetylene, but I guess not. Oxy-propane is very easily accessible, which is good.

I have never heard of Elmadur X, but naval bronze seems to be pretty cheap on Ebay if I ever find the need to make a pinch seal.

What do you mean by graphite pencils? Do you mean like literal writing/drawing pencil graphite? Or is this some industry terminology?

Nice to know about annealing, I guess I probably won't need to do that.

Thanks again for another excellent quality reply!

I mis-typed the alloy : it is actually Elmedur X.

Graphite pencils, Are basically just thick rods of high density graphite.  Like a pencil but without the wood, and no clay or binder in the graphite.  They are useful in diameters from about 5-10mm.

@James
Oh okay, so just sharpened graphite rods. I knew tools like this were used in scientific glassblowing and the like but I didn't know they had a name.

Thanks again