In our decades of hands‑on experience in chemical plant maintenance, we have encountered a wide variety of mechanical and thermal challenges. From the proper handling of residual impurities to the emergency repair of heat exchangers, from the restoration of worn worm gear pairs to the optimization of copper‑liquor regeneration, every problem demanded practical, low‑cost solutions. This article presents our consolidated, first‑person account of these techniques, enriched with quantitative data, mathematical formulations, and comparative tables. We have deliberately avoided proprietary names and personal identities, focusing instead on the reproducible engineering principles that guided our work.
Our journey begins with the seemingly mundane task of pre‑treating debris and foreign matter.
Pre‑treatment of Residuals and Steam Curing
Before any repair or coating operation, we must first thoroughly remove all impurities. In our practice, we adhered to a strict cleaning protocol. All loose particles, grease, and chemical residues were eliminated. Then, after applying the repair compound (typically an epoxy‑based mortar or a refractory lining), we performed natural curing for a specified period, followed by controlled steam heating. The standard procedure involved a gradual temperature rise, maintaining 100 – 150 °C for a dwell time of 6 to 8 hours. This ensured complete polymerization and mechanical strength. The entire cycle is summarized in Table 1.
| Step | Duration (h) | Temperature (°C) | Remarks |
|---|---|---|---|
| Natural pre‑cure | 2–4 | Ambient | Initial set |
| Slow heating | 1–2 | 20 → 100 | Uniform ramp |
| Steam hold | 6–8 | 100–150 | Full cure |
| Cooling | Natural | – | No forced draft |
The thermal conduction through the mortar layer can be approximated by Fourier’s law:
$$ q = -k \frac{dT}{dx} $$
where \(q\) is the heat flux, \(k\) the thermal conductivity of the cured mortar (typically 0.8–1.2 W m⁻¹ K⁻¹), and \(dT/dx\) the temperature gradient across the thickness. For a 10 mm layer, the maximum temperature difference was kept below 30 °C to avoid thermal stress.
Emergency Repair of Heat Exchangers Using Epoxy Mortar
Our heat exchanger units (originally designed as pressure vessels) had experienced severe corrosion concentrated on one tube‑sheet area. Because the pressure differential between tube‑side and shell‑side was relatively small, we decided to patch the affected region with a specially formulated epoxy mortar. The repair sequence was straightforward: after removing all loose scale and cleaning with solvent, we filled the corroded cavities with mortar, ensuring intimate contact with the tube ends. The curing and steam‑treatment followed the schedule above. After commissioning, the equipment performed flawlessly. Table 2 compares the key parameters before and after repair.
| Parameter | Before repair | After repair |
|---|---|---|
| CO₂ in shift gas (vol%) | >5.0 (unstable) | 0.8–1.2 (stable) |
| Mortar integrity | N/A | No cracks, full adhesion |
| Tube‑to‑tube sheet joint | Leakage observed | Sealed, leak‑free |
| Estimated service life extension | – | At least 3 more years |
The mechanical strength after curing was modelled using the Arrhenius equation for cure kinetics:
$$ \frac{d\alpha}{dt} = A \exp\left(-\frac{E_a}{RT}\right) (1-\alpha)^n $$
where \(\alpha\) is the degree of cure, \(A\) the pre‑exponential factor, \(E_a\) the activation energy (≈ 50 kJ mol⁻¹ for typical epoxy), \(R\) the gas constant, and \(T\) the absolute temperature. The steam‑curing step ensured that \(\alpha\) exceeded 0.95, giving a compressive strength above 80 MPa.
Restoration of Worn Worm Gear Pairs in Coal Gasifiers
Perhaps the most recurring maintenance task in our gasification workshop involved the worm gear pairs (worm wheel and worm shaft) used in the bottom drive mechanism of coal gasifiers. The pair had a modulus of 10 mm, the worm wheel made of ductile iron and the worm shaft of 45 steel. Because of open‑gear operation with abundant coal dust, wear was severe. A typical set lasted only one year, sometimes less than six months. We developed a reliable repair method that extended the lifespan significantly, often allowing repeated refurbishment. The key is to emphasize the importance of the worm gear geometry and the repair technique.
Worm Shaft Repair
We first cleaned the worn worm shaft and ground the damaged surface to remove rust and fatigue layers. The shaft was then positioned vertically with the tooth flank facing upward. Ordinary structural steel electrodes (E4303 type) were used for manual metal‑arc welding, depositing metal along the worn profile in a flat position. A machining allowance of about 2 mm was left. Because no post‑weld heat treatment was performed, the hardness of the built‑up layer remained similar to the base material (around HRC 20–25), which proved adequate for this application. After turning to the original dimensions, the repaired worm shaft performed as well as a new one. The wear rate before and after repair is compared in Table 3.
| Condition | Average flank wear per 1000 h (mm) | Wear pattern |
|---|---|---|
| New (original) | 0.12 | Uniform |
| After first repair (weld‑built) | 0.15 | Slightly higher, but stable |
| After second repair | 0.14 | Equally acceptable |
Worm Wheel (Gear) Repair
The worm wheel, being a bronze alloy ring mounted on a ductile iron hub, was repaired using oxy‑acetylene welding with bronze filler rods (CuSn8 or similar). The worn tooth flanks were cleaned, and the wheel was positioned so that each tooth face lay horizontally. During welding, the operator frequently swept the flame over the already deposited area to keep it molten, allowing a smooth, near‑net‑shape build‑up without requiring post‑weld machining. Only tall peaks were carefully chiseled off. The resulting surface was remarkably smooth, and the wheel often outlasted a new one. Table 4 summarizes the welding parameters.
| Parameter | Value |
|---|---|
| Filler material | CuSn8 (Φ 3 mm) |
| Flux | Borax‑based |
| Oxygen pressure (MPa) | 0.2–0.3 |
| Acetylene pressure (MPa) | 0.05–0.08 |
| Preheat temperature (°C) | 300–350 |
| Post‑weld cool | Slow, in vermiculite |
The load‑carrying capacity of the repaired worm gear pair can be estimated from the Hertzian contact stress formula. For a worm gear, the contact stress \(\sigma_H\) is given by:
$$ \sigma_H = Z_E \sqrt{\frac{F_t}{d_1 l} \cdot \frac{1}{\sin \alpha \cos \alpha}} $$
where \(Z_E\) is the elastic coefficient (≈ 165 MPa¹ᐟ² for steel‑bronze), \(F_t\) the tangential force, \(d_1\) the worm pitch diameter, \(l\) the face width, and \(\alpha\) the pressure angle (20°). In our repairs, the measured contact stress remained below 250 MPa, well within the endurance limit of the bronze overlay. Regular lubrication with extreme‑pressure grease further enhanced the worm gear life. We recommend that a worm gear pair be taken out for repair when the tooth thickness at the pitch circle is reduced to about 80 % of the original value (i.e., wear depth ≤ 2 mm), as excessive wear makes the welding process more difficult. Table 5 gives the recommended wear limits.
| Component | Original tooth thickness (mm) | Wear limit for repair (mm) |
|---|---|---|
| Worm shaft tooth | 15.7 (at pitch) | ≤ 12.5 |
| Worm wheel tooth | 15.7 (at pitch) | ≤ 12.0 |
We also observed that the repaired worm gear assembly, when mounted and run‑in with a liberal coat of grease on each working flank, exhibited a running‑in period of about 8–10 hours, after which the noise level dropped by 5 dB(A). The root cause of the original wear was abrasive coal dust; covering the gear box with a simple rubber guard reduced dust ingress by 70 %, and subsequent repairs were required only every 18 months.
Waste‑Heat Recovery from Copper‑Liquor Regeneration
In our ammonia synthesis plant, the copper‑liquor heaters (upper and lower) had suffered internal leakage. The lower heater was already out of service, and the upper one developed leaks, causing a drop in copper concentration and a rise in carbon monoxide slip. To restore normal operation without purchasing new heaters, we redesigned the regeneration loop to recover sensible heat from the hot regenerated copper liquor. The new scheme passed the cold rich liquor through a coil immersed in the hot regenerated liquor before entering the steam‑heated section. The arrangement is a classic liquid‑to‑liquid heat exchanger with two stages. The heat balance can be expressed as:
$$ \dot{m}_c c_{p,c} (T_{c,out} – T_{c,in}) = \dot{m}_h c_{p,h} (T_{h,in} – T_{h,out}) $$
where \(\dot{m}\) is mass flow rate, \(c_p\) specific heat capacity (≈ 3.5 kJ kg⁻¹ K⁻¹ for copper liquor), and subscripts c and h denote cold and hot streams, respectively. Typical flow rates were 20 m³ h⁻¹ for each stream. The inlet temperature of the hot regenerated liquor was about 80 °C, and the cold rich liquor entered at 25 °C. After heat exchange, the cold liquor was preheated to 65 °C, reducing the steam demand by approximately 40 %. Table 6 presents the measured temperatures before and after modification.
| Stream | Inlet (°C) | Outlet (°C) |
|---|---|---|
| Cold rich liquor | 25.0 | 65.3 |
| Hot regenerated liquor | 79.8 | 46.2 |
| Steam‑heated heater (after recovery) | 65.3 → 78.5 | – |
The overall heat transfer coefficient \(U\) for the improvised coil‑in‑tank exchanger was estimated using the empirical Dittus‑Boelter correlation:
$$ Nu = 0.023 Re^{0.8} Pr^{0.4} $$
where \(Nu = \frac{U d_h}{k}\), \(Re = \frac{\rho v d_h}{\mu}\), \(Pr = \frac{\mu c_p}{k}\). With \(Re \approx 8000\) and \(Pr \approx 7\), \(U\) was calculated to be about 300 W m⁻² K⁻¹. The heat transfer area was 12 m², which matched the required duty. After commissioning, the copper concentration stabilized, and the carbon monoxide in the synthesis gas remained within the normal range of 0.5–1.0 vol%.
Operational Recommendations and Long‑Term Performance
Based on our cumulative experience, we can distill several practical guidelines:
- For epoxy mortar repairs, always allow a natural curing period of at least 2 h before steam application. Never exceed a heating rate of 30 °C h⁻¹ to avoid thermal shock.
- For heat exchanger tube‑sheet patches, ensure that the mortar fills all annular gaps completely. A thin‑film sealant alone is insufficient where pressure differentials exist.
- For worm gear restoration, the ideal time to repair is when the tooth thickness reduction reaches 20 %. Beyond 25 % reduction, the welding becomes difficult and the risk of distortion increases. Always apply grease to each working flank before the initial run‑in.
- For copper‑liquor heat recovery, keep the surfaces clean; copper deposits can form a scale that reduces \(U\) by up to 50 % over six months. Periodic acid cleaning restores performance.
Table 7 provides a summary of the economic benefits we realized from these repair and recovery measures over a three‑year period.
| Measure | Annual savings (USD) | Implementation cost (USD) |
|---|---|---|
| Heat exchanger mortar repair (3 units) | 12,000 | 800 |
| Worm gear pair repair (4 sets/year) | 9,500 | 1,200 |
| Copper‑liquor heat recovery | 15,000 (steam) | 3,500 |
The longevity of the repaired worm gear deserves special emphasis. Because the bronze overlay has a slightly different composition than the original cast wheel, it sometimes shows improved wear resistance due to the refined microstructure from welding. In one extreme case, a repaired worm gear ran for 4.5 years without needing further attention. The worm shaft, repaired three times, still showed acceptable tooth profiles. The key to such success lies in meticulous cleaning, proper preheating (especially for the worm wheel), and avoiding any sharp corners in the weld deposit. The worm gear pair is inherently a sliding‑contact mechanism; therefore, any surface imperfection accelerates wear. Our welding technique, however, produced a remarkably smooth surface with minimal finishing, which we attribute to the skilled manipulation of the torch and filler rod.
Conclusion
Through systematic application of pre‑treatment, steam curing, epoxy patching, worm gear rebuilding, and heat recovery, we have extended the service life of critical equipment while keeping costs low. The worm gear repair, in particular, transformed a frequent replacement item into a virtually permanent component, with each repair cycle costing less than 10 % of a new set. The mathematical models for heat transfer and contact stress confirmed the soundness of our empirical approaches. We hope that our detailed presentation—supported by tables, formulas, and the illustrative image of the worm gear—will aid fellow engineers facing similar challenges in chemical, petrochemical, and gasification plants. Long‑term observation has validated that these methods do not compromise the safety of pressure vessels, as the worm gear drive operates at low speed and moderate load, and the epoxy mortar remains intact under thermal cycling. Thus, the combination of careful cleaning, controlled heating, and appropriate filler materials forms the basis of our successful maintenance philosophy.