Glass Torches

[Jerry]

Got my two bench torches working a couple days ago. The little one turned out to be a lot cooler that I though. Move the handles up and it opens the valves for the pinpoint flame, move them down and it opens the valves for the broad flame.



Sargent Welch bench torch by macona, on Flickr



Fischer multi flame burner by macona, on Flickr

[vmike]

What gas are you feeding the torches? I have a couple I've acquired at surplus auctions, but have never worked with glass. It's on my list of things to become proficient with.

mike

[Doug Coulter]

Most people use propane/oxygen for hard glass, and propane/air for soft. You will want to get Kohl's book if you really want to get into glasses, as he gives the data and usage of zillions of types (it's like specialty steels, there are a lot of glass "alloys").

Pyrex is fairly cheap and probably the easiest to work, at least for me. It happens to match the tempco of tungsten close enough that you can make feedthroughs etc with TIG rod up to 1/16" (McMaster has it down to .040, which is better yet -- the match isn't perfect, so the smaller the better). Pyrex is called 7740 by one maker, and Kimble calls theirs number K-33. It's the same stuff, and sometimes the art suppliers have it cheaper than the normal vendors -- glass work is popular for crafts types.

Soda lime glass is harder because it has such a high tempco it's hard to have it not break from internal stresses. Lead glass (as sold by the neon tube guys) is hard to work because when you think you're getting it hot enough, the lead reduces to metal and it becomes opaque -- you work that at pretty low temps while it is still pretty stiff. Quartz is hard, and you probably need oxy-acetylene for it. I've managed some small work with oxy-propane, but it's really not hot enough. It works at white heat, so you need good arc welding goggles for doing that.

Working glass, once you get the basics (expect to mess some up) is not real hard, if you understand the tempco issues and what happens with thick/thin together and cooling rates making stress even in all the same kind of glass -- it's all in what happens around the solidification temperature, and what parts get there first -- making thick to thin kinds of things difficult (just like welding). Kohl gives info on times and temps for annealing your work -- and it's way smart to do that or it may just explode on the bench after it cools. Some people just turn off the oxy and coat the work with soot so it will cool slowly -- better than nothing but not as good as a heat treat oven.

I made a viewer out of 2 crossed polarizer films and a CCFL bulb beneath that. You put your work between the films, and the stress lines show up, so you know if it's going to go "bang" at random sometime later. Once you pick up some experience, you don't need that as bad, but at first it's worth way more than it costs to make up.

Strong's "Procedures in Experimental Physics" has a good beginner tutorial in basic glass working and tools you need. It's not that much, but you do need them. McMaster sells graphite in the various forms you need, not too expensive, but remember it's very heat conductive, so you mount it with some insulation between it and your hand. Various jigs are worth making to guide things together for joining if you do much. I got an old kiln brick (the lightweight foamy kind) and cut some grooves in it with the mill. I can hear Jerry weep, but I used cheapo chinese tools for that, and they are even still sharp -- this stuff is not the firebrick you get at the hardware store -- it has near zero thermal mass and instantly heats white in the torch flame.

I got some "didyium" glasses that just wipe out the sodium D lines the glass emits at working temp so I can see what I'm doing -- very worth it. For quartz, arc welder's goggles do.
The rest is just getting your feet wet -- it gets easy with some practice. Maybe a lot of practice. Depends on how generally coordinated you are, it's kind of a touchy-feely thing.

Oh, BTW, if you have a lathe, it can be real handy for this.

[Jerry]

I am using propane-oxygen. I think the torches were designed for natural gas though. Boy, do they suck oxygen! In a short time went though half a 150 cu ft bottle.

The thing to remember with lead glass is to run a oxidizing flame. With that the lead never comes out of solution.

[Doug Coulter]

Well, I've had that happen even to the inside of a piece of lead glass I was heating the outside of...FWIW. You just can't get that stuff too hot or the PbO becomes Pb again, even in air, it seems.
I'll have to try it with an oxidizing flame, but in general those are just too hot for what you do with that kind of thing -- you can crack this stuff with the heat shock.

Yes, this is a good demo of why liquid hydrocarbon fuels are cool in IC engines. You don't have to carry the oxidizer around! Which in the case of air, is about 15/16 of the total weight.
No battery will ever match that unless we find some new stuff in the periodic table to work with -- in terms of energy/pound, we are there now. Some lithium cells have about the same
energy density as high explosives.

I'll have to fire up the miniature torch I built that I mostly use here and get a picture -- I used a MIG tip for it, and miniature needle valves. I use MAPP gas in normal
propane torches for bulk preheating, and that to work the details. I rarely reach for the oxy-acetylene unless it's quartz. The MIG tip torch is just right for things up
to about an inch, and doesn't eat so much gas or oxygen. Very tiny point flame.

Homebrew torch for glass work

Found an old pic. If I were making another one, I'd put the valves back on the handle instead of where they are -- they get hot there. I was a little concerned about having any amount of mixed gas volume...the thing on the right of the pic is a set of crossfires I made. Believe me, at that diameter a flashback is all too exciting, that one's air-gas only, can't do oxygen. In this picture, the torch is at full snot -- I usually don't do that for real work. The pic also demos why you want the fancy glasses that kill just the yellow sodium lines (which is what I was doing when I took it, I was showing a visitor how cool they are).

[Jerry]

When I mean oxidizing flame I dont mean using oxygen, just a flame where the air is turned up from a normal neutral flame.

Yeah, if it is getting dark on the inside you have gotten too hot.

[Doug Coulter]

Right. Well, it's not much of a problem for the intended use, as in making neon signs, where you bend it around pegs in a jig anyway, a little at a time, and it can be pretty stiff and still do that fine. Heck, it might be a "feature" where you do those sharp bends between letters you'd have to put black paint over anyway.

The one real use I've had for it here is making discharge tubes (duh) and the fact that it matches the tempco of Ti wire (pure) well enough to make an airtight seal. In other words, real high tempco as glass goes. I'd guess the neon sign guys like it because the high index of refraction makes them look a little more "sparkly". At any rate, the tempco match with Ti gives another new cheap way to get into and out of a vacuum with some power. I use pyrex and tungsten a lot for that, it's easier. And best of all if you can tolerate an O ring is the thing I made that just uses a piece of cut pyrex with an o-ringed conductor shoved in. That way, when you wreck it with hot ions, it's simple to replace, and very low cost.

But boy, if you want to melt an end shut and blow a bubble in that lead stuff, it's more or less impossible to do without turning it black (and electrically conductive in the bargain). Ditto making it soft enough to squish a seal with unheated long nose pliers.

I had heard that in the early days, this was one of the ways they made channeltrons....I'd guess they now use something fancier than lead to produce secondary electrons and conductivity down the channel. The ones I bought don't look like this, quite. Haven't needed them yet, so they are still vacuum packed.