Forges 101

[Mikey98118]

High quality rough coat diamond (#60 to #40 grit) disks are the future of small cutoff wheels; presently they are overpriced in discs smaller than 3” diameter, but competition will bring those prices down in the next couple of years.

    At present a Lenox 3” diamond coated disc costs $25. A Dremel 1-1/2” Max-Life diamond coated disc costs $30; why? The Lenox disc is aimed at the industrial market; the Dremel disc is aimed at the hobby market. The Lenox disc is rated for one thousand metal cuts. The Dremel disc is rated for a lot less.

 

1-1/4″ HDII is a heavily diamond encrusted (#46 grit) steel disc, for rotary tools (1/8” arbor hole) from Bad Dog Tools; it carries a lifetime guaranty for $32.47.

 

EZ Lock Max-Life High Performance Rotary Diamond Wheels are mounted on EZ-lock mandrels, which greatly reduce the likelihood of kickbacks; this makes them much safer than any other diamond coated steel cutoff discs. The discs are available from Home Depot and online from Dremel, for $30.

 

Lenox METALMAX diamond coated circular blades (AKA cutoff discs) in 1-1/2” size have 3/8” arbor holes, with brass spacer rings included; this allows them to be used on mandrels for 1/4” arbor holes, which is important, because rotary mandrels (with 1/8” shanks) only go up to 1/4” arbor thread versions. Metalmax blades are designed for steel cutting; they are available through Amazon.com, and at Lowes hardware stores. Do not attempt to use the 2” diameter blades with a 1/8” shank mandrel. 1-1/2” blades are already pushing the limit that such a weak shank can successfully spin. Even then, use these blades carefully; the shank probably won’t withstand kickback. So, why even mention the larger discs? Did you forget about 2” angle grinders? Steel discs can be run in 1/4” pneumatic die grinders, too. You have more choices than just between rotary tools electric die grinders.

High quality rough coat diamond (#60 to #40 grit) disks are the future of small cutoff wheels; presently they are overpriced in discs smaller than 3” diameter, but competition will bring those prices down in the next couple of years.

At present a Lenox 3” diamond coated disc costs $25. A Dremel 1-1/2” Max-Life diamond coated disc costs $30; why? The Lenox disc is aimed at the industrial market; the Dremel disc is aimed at the hobby market. The Lenox disc is rated for one thousand metal cuts. The Dremel disc is rated for a lot less.

 

1-1/4″ HDII is a heavily diamond encrusted (#46 grit) steel disc, for rotary tools (1/8” arbor hole) from Bad Dog Tools; it carries a lifetime guaranty for $32.47.

 

EZ Lock Max-Life High Performance Rotary Diamond Wheels are mounted on EZ-lock mandrels, which greatly reduce the likelihood of kickbacks; this makes them much safer than any other diamond coated steel cutoff discs. The discs are available from Home Depot and online from Dremel, for $30.

 

Lenox METALMAX diamond coated circular blades (AKA cutoff discs) in 1-1/2” size have 3/8” arbor holes, with brass spacer rings included; this allows them to be used on mandrels for 1/4” arbor holes, which is important, because rotary mandrels (with 1/8” shanks) only go up to 1/4” arbor thread versions. Metalmax blades are designed for steel cutting; they are available through Amazon.com, and at Lowes hardware stores. Do not attempt to use the 2” diameter blades with a 1/8” shank mandrel. 1-1/2” blades are already pushing the limit that such a weak shank can successfully spin. Even then, use these blades carefully; the shank probably won’t withstand kickback. So, why even mention the larger discs? Did you forget about 2” angle grinders? Steel discs can be run in 1/4” pneumatic die grinders, too. You have more choices than just between rotary tools electric die grinders.

[Mikey98118]

More on fuel efficiency

The easiest way to create a hot gas flame is to speed it up. But heat management is about more than how well fuel burns.

    The reason flames, whenever practicable, are aimed on a tangent, is to cause their combustion gasses to swirl around equipment interiors; creating a longer distance from flame tip to exhaust opening. Obviously, a lengthened exhaust path increases the amount of hang time to deposit combustion energy on internal surfaces. What isn't so clear is that the heat gain isn't added by hot gases blowing an extra foot or two, at high velocity; it’s due to a continuing drop in velocity over that added distance.

    Combustion gases begin to slow as soon as they leave the flame envelope, but small flames decelerate much faster than large flames. The smaller flames of two 1/2" burners will use the same amount of fuel as a single 3/4” burner, but will drop velocity much faster in a five-gallon propane cylinder forge, greatly increasing efficiency; because they can burn faster/hotter without creating a wasteful tongue of fire out the exhaust port.

    What about people who want to build a two-gallon forge from a non-refillable helium or Freon cylinder? They will need two 3/8" burners to do the same trick. Someone who wants to forge hand tools or cast jewelry in a one-gallon paint-can or three Lb. coffee-can will want two 1/4" burners to run their equipment with maximum efficiency; a 1/4" burner is also ideal for the so called “one-brick” forge, or  tomato can jewelry forges, which so many people attempt to run from a store-bought propane torch; without making the needed modifications.

    The law of diminishing returns reduces fuel savings in miniature equipment, but faster heating time, and increased portability remain major advantages. Portability? Yes; miniature equipment isn’t all that portable, if it must be fed from a five-gallon fuel cylinder.

    Do multiple flame burners (Giberson ceramic burner heads or homemade ribbon burners) take deceleration even further? Yes, they certainly do; unfortunately, the burners themselves tend to be large. Over time, compact multi-flame burners will be perfected, but first there must be a lot more interest in doing so; there isn’t much, because they provide the most advantage when heating large equipment. Also, the burner heads are ceramic; a lot more expensive material to play with, than

#316 stainless steel used in flame retention nozzles.

    Fortunately, Frosty has already used wooden forms for burner experiments; not very long lasting, but a whole lot cheaper than building every new design from ceramic

 

[Frosty]

Watch close they don't last long, 10-15 seconds and you can't read the flame.

Frosty The Lucky.

[Mikey98118]

Yes, I expect that the wood burning away complicates the flames, but it should still produce some valuable hints about whether or not you're on the right track, yes?

[Frosty]

Exactly, when the wood starts burning all you get are long yellow feathered flames and smoke. But for the first few seconds you can read the flames well enough to get the inducer in tuning range. 

If I had to cast a new burner block every time I tested one I'd still be thinking about it.

Frosty The Lucky.

[Mikey98118]

I have a lot on my plate, but am hoping to do several experiments with ceramic flame retention nozzles next summer. Of course, the chance to do some miniature ribbon burners at the same time will be to good to ignore

[Mikey98118]

I wonder if those few seconds can't be doubled with a little kiln wash?

[Frosty]

I don't know, I soaked one in water over night and it became unreadable in maybe 15 seconds. 10-15 was average but it was only clear for maybe 5. 

If I do it again, I'll make the plenum easier to screw to the board. 

Frosty The Lucky.

[Mikey98118]

Possibly, the plenum will take less experimentation than various flame hole plates. I expect that coming up with just the right combination of holes in just the right orifice sizes, will be what takes the most experimentation?

[Frosty]

This'll just be a tweak, no real experiment besides I did something I've been telling people NOT TO DO with the two NARBs, I welded them to their current forge. I'm blaming the tree! That's my story and I'm sticking to it.

The only real experimentation with the plenum I did was internally with the diffuser baffles. The different configurations didn't seem to make much difference, the flames in the center of the block were always longer and at the ends sometimes hardly enough flow to burn at all.

Eliminating the excessively restrictive diffusion baffle as shown in the plans you see everywhere was my aim to start with so it didn't need such high static pressure to force the mix through the tiny gaps. 

The real experiment was moving the plenum inlet port 90* to the alignment of the outlet nozzles. The flow from the inducer hits the far side of the plenum and self distributes nicely. There's still a little different in flame size between center and ends but it's not significant.

If I do this again, I'll use some of the 2" x 3" rectangular tubing for the plenum. The burner block will still be 2" x whatever. The extra height of the plenum should allow more room  for pressure to equalize and better still be a little easier to fit on a forge.

The first plenums slipped over a 2" slice of 2x4 and screwed through the sides of the plenum into the test block. The next one if I do it again will have an angle iron tab at each end I can screw through into a test block. About 1/16" of plenum past the flanges will embed in the wood for a tight seal. It only has to last a few seconds but I didn't care for the entertainment of the fireball the leaky connection made before I tweaked it. 

The tabs will b easy enough to remove if they won't work in a forge.

If the working blocks are different enough I might need to come up with a different plenum but I don't see why I couldn't weld a ledge inside to cement ceramic or cast refractory blocks if that's a better method. 

So, yeah all the real experimentation was getting the blocks right.

Frosty The Lucky.

[Mikey98118]

Good info, Frosty.

[Buzzkill]

On 10/30/2021 at 1:23 PM, Frosty said:

Exactly, when the wood starts burning all you get are long yellow feathered flames and smoke.

I can't help but wonder if those SS sleeves I used in my latest NARB could be used in wood test blocks to give you more time.

[Frosty]

Good thought, I expect they would. I was experimenting with nozzlette size as well as number and I only needed a few seconds to tell if it was within tuning range of the inducer. 

I don't know but doubt if a wood block would work for nozzlettes 3/16" - 1/8" in dia. in the first place, so the SS sleeves are probably more than an option.

Wood provides a lot more friction than cast refractory. The refractory NARBS perform quite differently than the wood test block NARBS but they were within tuning range.

Frosty The Lucky.

[Mikey98118]

Fuels (from my notes)

Safety starts at the fuel cylinder, and what choices you make there. The only fuels you should be concerned with are propane and propylene; both are LPG fuels (liquid petroleum gas); both are heavier than air. Adiabatic temperatures are the mathematically derived greatest possible combustion temperature of a given fuel; that of propane and propylene burning in air are about equal. The actual flame temperatures of air/propylene flames are a lot hotter than that of propane. Why? Because in the real world, how hot a flame burns depends on how well the burner’s design can combust the fuel; most hydrocarbon fuels will burn at about the same heat in a jet engine; the richer hydrocarbons will simply require less fuel. how hot they burn in a torch or burner varies widely.

    Propylene runs at higher cylinder pressures than propane at any given ambient temperature; therefore, propylene cylinders have thicker walls; outside of these differences, you’ll find that safety regulations are similar for both fuels. Propylene’s much higher flame temperature will call for refractory flame nozzles when burners are placed in equipment interiors; kilns, furnace, and forge interiors need use-ratings to be upgraded over what you’d normally choose for a refractory hot-face that’s only heated with propane; or you can reduce burners to the next smaller size, to what is recommend with propane fuel.

    Never store a fuel gas cylinder out of vertical-up position. The pressure relief valve on your fuel cylinder, is set to discharge gas into the air, if heat levels cause internal pressure to rise too high on hot days. The valve is normally located in the gas space above the liquid fuel level; Moving the cylinder out of vertical orientation can allow the valve’s location to end up below the liquid fuel level; it can then spew a large amount liquid into the air (instead of a small amount of compressed gas); that liquid immediately expands about 270 times in volume, and then mixes with an even greater volume of air, creating an explosion hazard.

    Propane cylinders are required to have an internal OPD valve (Over-fill Protection Device), to prevent the cylinder from being filled to more than 80% capacity.  Furthermore, cylinders are required to be safety tested every ten years; this rule is enforced in most areas, and tanks that are dinged, or rusted will not pass inspection. Do not store fuel gas cylinders in direct sunlight, or expose them to other heat sources.

    Autocrats being what they are, some states took the addition of federally mandated over-fill valves to quietly add an internal flow limiting device to the OPD valve, on five-gallon (and possibly on other small) propane cylinders; when used on a barbecue grill, flow-limiters shouldn’t create problems; when used on this kind of heating equipment (which all demand much higher flow rates), they cause major problems. Check with your local propane dealer to see if you live in one of these states, and what cylinder sizes to avoid.

    Compressed gas fuel cylinders must be physically separated from ignition sources;they must be stored and used in a shaded area outside of any building; or you must provide code approved barriers walling it away from ignition sources, and from oxygen cylinders. Even when you are working outside a building, both fuel and oxygen cylinders are required to be placed at a minimum distance of twenty feet away from your work, and any other ignition source (ex. electrical outlets), and twenty feet away from each other, unless they are housed in a torch cart. Oxy-fuel torch carts in transit, or on field sites are an exception to the separation of oxygen and fuel cylinders rule. Using such a torch set-up in your shop probably violates safety codes in most places, despite its being a common practice. The torch set should have a twenty-five-foot-long hose, allowing the fuel and oxygen to be placed at a twenty-foot distance from hot work, and other ignition sources, while leaving enough hose to do the work.

    Most local codes also require propane cylinders to be set on a concrete pad above ground level (to protect their thin shells from rusting from exposure to ground water). I recommend that the cylinder also be placed in an aerated enclosure (ex, heavy gauge wire cage). All compressed gas cylinders are required to be physically constrained from falling over (ex. chains, or metal straps).

    When not in use, the fuel cylinder valve should be capped, to keep it clear of debris. If the cylinder is attached to a piping system, a flexible hose section protected by braided stainless steel “armor” should be employed between the outdoor section of pipe and the fuel cylinder, to prevent gnawing by rodents; they can be attracted by the smell from minor leaks, and from cylinder changes. 

    Most LPG fuels are heavier than air and slow to disperse; which means these fuel gases can collect in low lying areas, gradually making their way to an ignition source. Methane (AKA natural gas) is the only member of the LPG family that is lighter than air.

Ignition Sources includes flame, sparks, or heated surfaces capable of igniting

flammable vapors or fumes; including appliance burners, burner ignitors, and electrical switching devices (light switches, electrical outlets, power switches on equipment, etc.).

    If you smell flammable gas, the proper procedure is to clear the area, shut off the valve on the outdoor fuel cylinder, and call your local fire department.

    In this chapter’s opening remarks, I mentioned what it feels like to deal with legal problems, after burning down a garage. How much worse do you think that would be if that garage was part of your house? If you don’t have a separate shed or garage to do your hot work in, put wheels on your equipment, and work out in the yard; that includes welding, and brazing jobs. You may expect it will be inconvenient, but you will instead find it is surprisingly fun—and a whole lot safer! These days, all kinds of work tables and benches are designed to be set up on job sites (and easily transported when you change addresses), so assuming that doing “shop work” out of doors will be inconvenient is outmoded thinking. A few years back, I replaced fifty-eight feet of wooden fence with ornamental iron, which was built in the back yard, and welded in place, one section at a time. The neighbors were so fascinated that people stopped their cars to look, for a week. I never heard a single complaint.

[Mikey98118]

One last thing about propane versus propylene is their costs. Propylene in 16 oz. canisters cost at least twice the price of propane in canisters. However, compressed propylene in refillable cylinders down at your local welding store only costs about a third more than an equal amount of propane. So, burning propylene that way doesn't really cost a dime more than propane, because it has about one-third more heat, too

[Mikey98118]

Of course, nothing is for nothing so...you need to use refractory flame retention nozzles on burners used to heat forges with propylene, or no nozzle at all. You also need to use refractory capable of handling more heat for the flame face, or turn the burner way down, until you can upgrade the refractory.

[Mikey98118]

Mounting burner ports

Typically, a burner port (entrance) consists of a short tube or pipe about 1/4” to 3/8” larger than a burner’s mixing tube all the way around; about 1/2” to 3/4” larger inside diameter than the outside diameter of the burner’s mixing tube. This allows enough space to aim the burner somewhat within the portal. You also want the port large enough for the burner’s flame retention nozzle to slide through it.

    The burner is held in position and aimed with two rows of thumbscrews; each row has three equidistant screws. One of the advantages of these screws is that they can hold a pipe or tube in place within the portal, and resting exactly where the flame is intended to impinge, while the portal opening is being ground into an oblong shape (to allow the tube to be aimed at a desired angle). This method ends up with a very close fit in the shell opening, to promote easy silver brazing of the port’s tube to the equipment shell. You’re building the burner first, so why not employ it to help you construct its forge?

    Alternatively, you can drill and mount a burner port in the shell with two bent flat bars and some pop rivets, or self-drilling screws. Bracketing parts together can end up looking tacky if you don’t manage to keep the shell opening tolerances close. Employing screwed brackets can be a be a minor pain, if the burner’s port tube is positioned at an angle.

    Welding equipment parts, such as burner ports unto a steel shell, takes a wire feed machine and a learning curve. Some people are reworded with distortion in the shell, because of welding contraction; it just takes a little time to learn to run a wire feed welder, and somewhat more to bridge gaps with one; but it takes a lot more time to learn where and how much to weld without creating distortion. Silver brazing, or even hard brazing, doesn’t create distortion problems, but does require close fit-ups.

    Hard brazing requires an oxy/fuel torch, or an air/fuel torch, propylene fuel, and a lot of skill, Silver brazing can be done with an air/fuel torch, propane, and close attention to fit; most silver brazing alloys won’t bridge gaps easily.

    Silver brazing by hand torch benefits from a low temperature filler with broad melting  range such as Ufhauser silver braze filler A-54N (54% silver/ ) that has a broad elastic range (250 °F), and bridges irregular gaps; it can be considered a capping alloy, but if heated too slowly it can suffer from liquation (where the alloy separates into solid and liquid zones); it will melt between 1325 °F (dark red) and 1575 °F (bright red). The high temperature portion will melt only above the normal brazing temperature afterward. For this reason, alloy A-54N should be heated rapidly through its melting range. If you are joining a thin shell from a tin can to a thicker tube, keep the flame mostly on the tube.

    This filler alloy has a good color match to steel. Reasonable care with a sanding drum or grinding stone in a die grinder or electric rotary tool, will easily produce a sufficiently close-fit in the joint between a burner portal tube and the forge shell opening. Burn polypropylene fuel gas if you employ an air/fuel torch, and employ a high temperature black flux meant for brazing on stainless steel.

     Car mufflers are zinc coated, and silver brazing parts to this kind of forge shell will ensure lots of damage to the plating. Zinc-based solder sticks may be employed afterward. Most zinc-based soldering alloys are zinc-tin-lead (avoid these), zinc-tin-copper (excellent), or zinc-cadmium (use fume rated respirator with these and follow all safety guidelines to the letter).

Note: The main ingredient in zinc flux is zinc chloride (follow safety guidelines on container); it is the only ingredient in many of them; it tends to “tin” the surface of steel, rather than just cleaning it. If steel is freshly cleaned and power buffed with stainless steel wire wheels, it can be zinc soldered without flux, but why do things the hard way? Zinc’s melting point is 787 °F; comfortably below its boiling point (1665 °F). Zinc fumes are easily seen and smelled; avoid them. Unlike lead fumes, it takes a heavier dose of zinc vapors to cause fume fever. Unlike lead, the body can tolerate a little zinc, but keep your dose tiny; none is best. No metal fumes are good for your lungs.

Caution: Metals give off toxic fumes upon reaching their boiling points. Using zinc coated sheet metal or parts (such as car mufflers) is okay if you're careful about doing it. The boiling temperature of zinc (the point at which it makes fumes) is

1665 °F (bright red heat). Your forge shell should not get higher than one-fourth that temperature. But you do need to be careful to keep the shell well away from the edge of the exhaust openings, by not making the openings in ceramic fiber, kiln shelf, or cast refractory even with, or even near the shell. But zinc coated flame retention nozzles or mixing tubes are out. There is no reason at all to avoid zinc coated reducer fittings on a burner’s air entrances. In other words, keep zinc away from part surfaces that may become incandescent (above 1200 °F or 649 °C).

Note: Preheat temperatures should be kept down to 600 °F (315 °C on zinc coated surfaces, such as car mufflers, to avoid damage to the existing coating on their surfaces, and to keep scale formation down on the steel; “tinning” the bare steel with a zinc chloride-based flux will help with this. Remove all residual flux with hot water and a clean rag after soldering.

    Larry Zoeller (of Larry Zoeller Forge) is credited for first mounting schedule #40 pipe to a forge shell with conduit locking rings; he calls it a “burner holder assembly.” If you’re looking for fast and easy, he sells them for $25 and shipping from his website. Their main limitation is that they can only be positioned at right angles.

    A zinc coated burner port should be completely external to the forge shell; it should not extend inside the forge. The internal part of conduit locking rings should be soaked in vinegar overnight to remove the zinc coating, before being screwed onto the forge shell.

[Mikey98118]

Exhaust size and shape: One thing backyard casters and blacksmiths both worry over is how large to make the exhaust openings on their equipment. Too small and you have high back pressure killing burner performance; too large and you can't get enough heat to stay in the equipment to do your work. Of course, the closer to the "right" opening size your equipment is the stronger the forge or furnace can be built. Just don't get suckered into confusing the right size for a “perfect” size. As long as burner output can by varied (turn-down range), there can't be any such thing as a perfect opening size. The right size is what is needed to accommodate the burner's highest output (the highest you are willing to take it to).

    Variable is the optimal opening size; all other dimensions can be outright wrong, but seldom just right, with a burner flame that can be varied. This is one of the many reasons for controlling exhaust with an external baffle wall beyond a larger than needed ringed opening; thus, allowing the least heat loss through radiation, while maintaining optimal back pressure in the forge. Why include a ring around the exhaust opening? To divert hot exhaust gasses away from the metal shell, where it could otherwise overheat the metal, and cause distortion.

    Moving the baffle wall closer or farther from the exhaust opening allows you to tune backpressure in the forge to the burner flame. Sound complicated? Your baffle can be as simple as bricks laid near the opening, or as complex as kiln shelves (with small drilled and cut openings) trapped in a hinged lached locked door.

Doors: Maximum clearance can be provided with a hinged and latched forge door that contains built-in changeable baffle plates. A door makes building the refractory structures of equipment much easier, and permits larger pieces to be heated then would pass through a standard exhaust opening. Best of all it allows movable internal baffles to be used, which would not pass a narrowed exhaust opening; this promotes the use of a single burner to be used for short pieces, saving money in a five-gallon propane cylinder furnace run by two ½” burners.

The door is a big step up from an exterior brick baffle wall; it should include a parts entrance that can be varied in size; for instance, with several round (or hexagonal) kiln shelves with different openings for passing stock through, which can be exchanged, and trapped in a pocket on the door. All of these improvements don’t need to be seen to at once, so long as a hinged and latched door is included in the forge shell.

    If you choose a simple brick baffle wall in front of the forge, keep the bricks at a small distance from the exhaust opening, to allow hot gases to move up and out, between forge shell and brick wall, while bouncing radiation off of a re-emissive (heat reflective) coating, back into your forge. Keep the stock entrance only as large as is needed to move parts through.

    This arrangement helps to slow the flow of expended gas in the forge interior, as it heads toward the exhaust opening; and then speed the gas up through the opening; another highly desirable trade off. So, you are gaining hang time for the heated gas in the forge, and recuperative savings from bounce back of radiant energy; another win-win situation. A baffle wall also minimizes infrared and visible light from impacting your eyes and skin, improving your health and comfort.

    While hinged and latched doors can do just as much on box shaped forges, all the examples I have seen slide up and down.

High alumina kiln shelves are seven times more insulating than hard fire brick; they are also tougher at forge temperatures, which is an important consideration for something you will end up shoving parts back and forth through. Using alternate kiln shelves, with different part openings is fine, but building an elaborate system of moving kiln shelf parts to ape the ability of bricks to change their openings comes under the heading of "gilding the Lilly." The additional energy savings it provides probably isn't worth the effort. Make up new openings in shelve baffle walls sparingly. Diamond coated and carbide coated rotary burrs (and coated hole saws) are the preferred way to drill holes in kiln shelves. Friction cutoff blades or diamond coated blades are the best way to cut lines between those holes.

    You want to coat the hot-face of the door with one of the re-emissive coatings, use a formula of 95% zirconia silicate (zircon) and 5% Veegum (or bentonite as an alternate); this mixture makes a tough re-emissive coating for wear surfaces. Zirconium silicate can also be mixed with fumed silica to make a tuff and effective coating on refractories (but not on ceramic fiber products). There are other choices, Like Plistix 900F, but none of them are as economical or easily purchased in other countries.

[Mikey98118]

Propylene costs about twice the price of propane in 16 oz. canisters, but only about one-third more than propane in refillable cylinders down at your local welding supplies store. Since it provides about one-third more heat, this fuel might seem to produce no major advantage in heating equipment; but, no matter how cleverly you design a forge, to reduce exhaust speed, the gases lower limit depends on how much fuel must be combusted to attain desired internal temperatures. So, flame temperature, sets an unexpected limit on efficiency.

[Mikey98118]

Wow, Frosty! I expected to at least get a comment out of you on this.

[Frosty]

I've never burned propylene except maybe in a soldering torch years ago. I don't know it's: flammable ratio range, flame front velocity, mixing characteristics and requirements. Zippo.

Everything you said is pretty much dead on as far as I can see. I don't know enough to discuss flame velocity in NARBs even.

Frosty The Lucky.

[JHCC]

Don't Zippos use butane?

[Irondragon Forge ClayWorks]

Some Zippos use butane but the majority (especially old ones with wicks) use lighter fluid like naphtha/benzine.

[pnut]

I've never seen a butane fueled Zippo. They've all been naptha/lighter fluid fueled lighters. 

Pnut

[Leather Bill]

I've seen people put gasoline in a Zippo,it worked so they dropped it in their pocket and went about their business.   Before long the fumes blistered their leg.