Guide
Acetylene generator overheating: how polymerisation and incandescent carbide start and how to stop them
An acetylene generator overheats when the roughly 400 kcal released per kg of carbide is not carried away. Above about 100–200 °C acetylene polymerises, releasing more heat, and at the extreme carbide turns incandescent at 1,000–1,100 °C. The warning signs are a tar smell and brownish-yellow gas and residues.
- Heat released
- About 400 kcal per kg of carbide yielding 300 litres; 429 kcal/kg on the more detailed calculation
- Adiabatic temperature rise
- About 730 °C if the reaction runs with no cooling
- Polymerisation onset
- From about 100 °C to 200 °C and above
- Incandescent carbide
- About 1
- Water ratio, falling-carbide generators
- 6 to 10 times the weight of carbide decomposed
Calcium carbide releases about 400 kilocalories of heat for every kilogram that reacts, and an acetylene generator exists as much to carry that heat away as to make gas. When it does not, a predictable chain follows: the reaction water warms, acetylene begins to polymerise somewhere between 100 °C and 200 °C, polymerisation releases more heat and accelerates, and at the extreme the carbide itself turns incandescent at 1,000–1,100 °C. The C.P.I. safety manual of 1968 set out this chain, its causes and its warning signs in more detail than any current standard, and its reasoning is the basis of this page.
How much heat, and where it comes from
The reaction CaC₂ + 2 H₂O → Ca(OH)₂ + C₂H₂ releases 31.0 kcal per mole, which S. A. Miller (1965) works out as 484 kcal per kg of pure carbide and 429 kcal per kg of commercial carbide yielding 300 l/kg, after adding the heat of slaking its free lime. The C.P.I. uses a round figure of about 400 kcal per kg, built up from 27,000 cal per mole for the carbide reaction plus 15,500 cal per mole for slaking the 10–15 % free quicklime. EIGA Doc 237 gives 1,793 kJ/kg for commercial carbide of about 80 % purity. All three agree within about 7 %.
Two consequences follow. First, if nothing removed the heat, the stoichiometric reaction would raise the temperature of its products by about 730 °C. Second, and less obviously, the C.P.I. pointed out that heat per cubic metre of acetylene rises as carbide quality falls, because a leaner carbide carries more quicklime, which slakes and heats without making gas. Poor carbide is hotter per unit of output and needs more water, not less.
| Quantity | Value | Source |
|---|---|---|
| Heat of reaction | 31.0 kcal/mol | Miller (1965), from heats of formation |
| Pure CaC₂ | 484 kcal/kg | Miller (1965) |
| Commercial, 300 l/kg | 429 kcal/kg; about 400 kcal/kg | Miller (1965); C.P.I. 1968 |
| Adiabatic rise, no cooling | About 730 °C | Miller (1965) |
| Worked example | 1 kg in 10 litres of water at 20 °C reaches about 60 °C | C.P.I. 1968 |
| Polymerisation onset | About 100 °C to 200 °C and above | C.P.I. 1968 |
| Incandescent carbide | 1,000–1,100 °C | C.P.I. 1968 |
The C.P.I. worked example is the basis of every water ratio: one kilogram of carbide decomposed in 10 litres of water at 20 °C can in theory heat it to 60 °C. Halve the water and the temperature rise doubles.
The runaway
The C.P.I. described polymerisation as reactions occurring at variable temperatures from about a hundred degrees to 200 °C and more, depending on temperature, pressure, rate of passage and the diluents and catalytic surfaces present, and noted that because these reactions release heat they can continue and even accelerate if the heat is not removed. That is the runaway. EIGA Doc 123 gives a much higher onset, about 400 °C, for acetylene at atmospheric pressure; the C.P.I.'s lower range describes generator conditions with lime and carbide surfaces present, and is the conservative figure to work to in a generator. If heating continues the carbide decomposes to lime, hydrogen and carbon, and the mass may become incandescent, at temperatures of about 1,000 °C, elsewhere stated as 1,000–1,100 °C, close to those found by calculation. The limiting reactions are CaC₂ + H₂O → 2C + CaO + H₂, releasing 63,000 cal, and CaC₂ + Ca(OH)₂ → 2C + 2CaO + H₂, releasing 47,000 cal.
Laboratory data confirm how easily it starts. US Bureau of Mines Report of Investigations 3794 (1945) and RI 4067 (Jones, Scott, Kennedy and Spolan, 1947), as compiled by Miller, record moist air from water at 70 °C heating 14–28 mesh carbide until it glowed at 1,100 °C, with no liquid water present. Miller also records that explosions in medium-pressure generators have been traced to wet acetylene and carbide dust: moist gas wets dust lying in a hopper or chute, the dust reacts with no liquid heat sink, and the local temperature heads for the adiabatic figure.
The warning signs are specific. The C.P.I. described the phenomenon as characterised by the stench of tar and the brownish-yellow colour taken on by the gas and the residues. In dry-lime generators, the residue should be white or pale grey with 2–14 % moisture, and discolouration indicates polymerisation. Gas yield per kilogram falls at the same time, because acetylene is being converted to heavier products instead of reaching the gasholder.
What incandescence actually endangers
It is natural to fear that glowing carbide will set off a decomposition of the acetylene around it. The C.P.I. considered this risk relatively slight, because the acetylene in a generator is always accompanied by a considerable quantity of steam, and water vapour strongly moderates decomposition, particularly at high temperature. The C.P.I. tables quantify it: at 80 °C, water-saturated acetylene needs 1.4 kg/cm² to propagate decomposition against 0.40 kg/cm² dry (Rimarski's dry-gas series).
The real danger is different. The hot points can ignite mixtures of air and acetylene, and such a mixture forms when the generator, and especially the reaction chamber, is opened. The hazard of an overheated generator is realised at the moment someone opens it. Hence the C.P.I.'s rule for drawer generators: never remove the drawer within 10 minutes of the end of gasification, recognised by the pressure no longer rising, because hot residues and premature contact with air may give rise to ignition, and the person opening it should not stand in the zone where residual acetylene is given off.
Rules that matter
The C.P.I. listed the conditions that fail to carry away the heat. Every one is an operating condition, not a design accident:
The causes
- Lime coating the carbide pieces while there is still enough water for reaction to continue; lime is a poor conductor.
- Too high a lime concentration in the reaction water.
- Insufficient movement or agitation of the reacting products.
- Clogging of the carbide mass in dry-lime generators when water delivery is excessive.
- A choked basket in contact generators.
- Re-use of the same reaction water over several charges, which the C.P.I. called a method of working to be shunned, because it raises the lime concentration and leads to an abnormal rise in temperature, a fall in performance, a risk of overproduction and even polymerisation.
- Very fine carbide of 0–2 mm attacked without vigorous agitation; in the motionless state, and even floating on water, it can give rise to incandescent residues.
Miller adds the floating dust cake in carbide-to-water machines, where reaction proceeds above the liquid with no heat removal until the cake breaks up and gasifies suddenly, and surface-slaked carbide meeting hot carbide in bins, which at 200–350 °C polymerises acetylene over lime. EIGA Doc 237 adds carbide islands from agitation failure and low-water hot spots from stone build-up on a turnplate.
The engineering answer
Each generator type removes the heat differently, and the C.P.I. gave water ratios by type:
| Generator type | Water | How heat is removed |
|---|---|---|
| Falling carbide, carbide into water | 6–10 × the carbide weight | Large water excess, regular discharge of milk of lime, replacement with water at ambient temperature, mechanical stirring |
| Contact (basket) | 6–10 × the carbide weight | Large excess of water, container designed against clogging |
| Drawer, water onto carbide | 2–4 × the carbide weight | Chamber shape with a large surface relative to volume |
| Dry-lime | About 1 litre per kg | Evaporation; runs at 90–95 °C with indispensable stirring |
KIGL generators are carbide-to-water, continuous-feed machines with a design band of 6–8 litres of water per kg and an agitator running 5 minutes in every 15, discussed under water-to-carbide ratio. Miller derives the 6:1 floor for low-pressure generators from two limits that coincide: a 75 °C temperature limit and an 18 % solids fluidity limit. EIGA Doc 237 recommends about 8:1 within 7:1 to 10:1 and a reaction-water temperature of 70–80 °C for maximum yield. IS 8471:2003 cl 8.2.1 requires at least 8 litres per kg unless water and carbide are introduced continuously in a predetermined ratio. All of these numbers assume cool feedwater; warm Indian feedwater raises every equilibrium temperature, so the ratio should be checked against measured temperatures, not assumed.
The protective systems follow: a high-temperature trip that stops the carbide feed, a low-water-level trip, agitator rotation monitoring, and air-operated fail-safe controls, as listed under interlocks and alarms.
Rules that matter
Rules to apply
- Keep the water-to-carbide ratio inside the design band and raise it when carbide is poor or the feedwater is warm.
- Discharge milk of lime regularly and replace it with fresh water; never re-use reaction water over several charges.
- Keep the agitator running its cycle and confirm the shaft turns at the far end.
- Never let fines sit on the surface; they must be wetted and stirred.
- Treat a tar smell or brownish-yellow residue as a stop signal: stop the carbide feed, follow the plant's written emergency procedure, and find out why the temperature trip did not act.
- Never open a generator, hopper or drawer while it is hot; wait, cool, purge, then open, standing clear.
For fault-finding by symptom see acetylene plant troubleshooting; for the decomposition physics see acetylene decomposition; and for the carbide hazard in storage see the carbide safety data summary.
Acetylene generator overheating — frequently asked questions
What causes an acetylene generator to overheat?
Too little water per kg of carbide, lime building up in the reaction water, too little agitation, re-using the same water over several charges, fine carbide gasified without stirring, and lime coatings on the lumps. The C.P.I. listed all of these in 1968. In a carbide-to-water machine the commonest cause is a water-to-carbide ratio that has drifted below the design band.
How hot does carbide get if a generator runs dry?
Very hot. The stoichiometric reaction with no cooling would heat its products by about 730 °C, and incandescent carbide has been measured at 1,000–1,100 °C. In laboratory tests moist air at 70 °C made fine carbide glow at 1,100 °C. At those temperatures the carbide itself decomposes to lime, carbon and hydrogen.
What is the sign of polymerisation in a generator?
A smell of tar and a brownish-yellow colour in the gas and the residues. Dry-lime generator residue that should be white or pale grey becomes discoloured. Gas yield falls at the same time, because acetylene is being converted to heavier compounds instead of reaching the gasholder.
Is incandescent carbide an explosion risk?
Not mainly as a decomposition risk, because the steam in the generator strongly suppresses acetylene decomposition. The danger is that the hot points ignite a mixture of air and acetylene, and the moment that mixture forms is when the generator is opened. Never open a generator, drawer or hopper until it has cooled and been purged.
Why is leaner carbide hotter?
Because the free quicklime in commercial carbide slakes exothermically without making any gas. The C.P.I. noted that heat per cubic metre of acetylene rises as carbide richness falls, since a leaner carbide has more quicklime. Poor carbide therefore needs more water per kg, not less.
How long should I wait before opening a drawer generator?
At least 10 minutes after gasification has ended, recognised by the pressure no longer rising, and the person opening it should not stand where residual acetylene is given off. That was the C.P.I. rule for drawer generators, and the same principle of cooling, waiting and purging applies to opening any generator.
References
- Bureau of Indian Standards. (2003). Acetylene Generators — Requirements (IS 8471:2003, First Revision).
- European Industrial Gases Association. (n.d.). Safe Operation of Acetylene Generator Systems (EIGA Doc 237).
- US Bureau of Mines reports (NIOSH mining archive)
Enquiries
Request a quotation
State the gas or equipment, the quantity or capacity, and your location. KIGL will confirm the grade, the cylinder or plant size, and whether the location can be served from Noonmati, or through the agent network.