The repeated openings most likely come from the rupture disc’s temperature basis, an unmeasured pressure transient, reverse differential pressure, or cyclic fatigue—not the steady 4–5 kg/cm2 steam-header reading. The disc was specified at 100 °C but operates at 140–150 °C, and the shell itself has no pressure transmitter. Measure pressure on both sides of the disc during startup, trap cycles, and shutdown before changing the temperature loop or relief device.
What do the symptoms rule in and rule out?
The failures began after the preheater and ethylene de-oxo bed were returned to service. That timing directs the investigation toward operating transients and installation conditions introduced by active steam and condensate flow.
A stable LP steam header at 4–5 kg/cm2 does not prove that shell pressure remains at that value. The header measurement is upstream of the temperature control valve, while the rupture disc responds to pressure at its own process connection. Valve movement, trapped condensate, steam collapse, trap discharge cycles, and local hydraulic shocks can produce short events that a header trend never records.
The tube side operates at 40 bar, with ethylene heated from 30 °C to 100 °C. A tube leak remains a credible overpressure case even though no leak was found during the initial check. It is not, however, the only mechanism capable of damaging the disc.
| Signal | Source or measurement point | Wrong-value symptom |
|---|---|---|
| Steam pressure | Shell nozzle near the rupture disc | A short positive spike identifies local overpressure that the 4–5 kg/cm2 header indication misses. |
| Flare pressure | Disc discharge connection or correlated flare-header measurement | Changing backpressure alters differential pressure; a reversal can flex an unsupported disc. |
| Steam temperature | Shell operating region | The measured 140–150 °C exceeds the 100 °C burst-temperature specification and invalidates the selected calibration basis. |
| Condensate behavior | Shell outlet and pressure-pump steam trap | Flooding, intermittent discharge, or hammer points to condensate accumulation and transient loading. |
| Control-valve output and position | Temperature controller and steam valve | Repeated large movements can correlate steam admission with pressure and thermal cycles. |
| Ethylene outlet temperature | Tube-side outlet | Oscillation or delayed recovery can indicate shell flooding, trap cycling, or an oversized/poorly behaving steam valve. |
How does differential pressure determine disc loading?
A rupture disc responds to differential pressure, not to shell pressure alone. For the normal flow direction, the relevant relationship is:
Disc differential pressure = shell-side pressure − flare-side pressure
The records list 2 kg/cm2 flare backpressure, a 13 kg/cm2 data-sheet burst pressure, and an 11 kg/cm2 stamped bursting pressure. Before using those values in a calculation, resolve three items from the certified disc documentation: whether each pressure is gauge or absolute, whether 13 kg/cm2 is the rated differential pressure at 100 °C, and whether the 11 kg/cm2 stamp represents a backpressure correction or a different specified condition. Adding or subtracting the 2 kg/cm2 without those definitions can produce an unsafe conclusion.
At the reported steady conditions, a 4–5 kg/cm2 shell pressure and 2 kg/cm2 flare pressure would give only 2–3 kg/cm2 forward differential pressure if both readings use the same pressure reference. That calculation explains why the steady readings do not account for rupture, but it does not exclude a brief local spike. It also says nothing about reverse loading when flare pressure exceeds shell pressure.
Flare pressure changes can load the disc in both directions. Falling flare pressure increases forward differential pressure for a given shell pressure. Rising flare pressure during low shell pressure, steam isolation, or condensation can create reverse differential pressure. A disc requiring vacuum or backpressure support may deform, buckle, or accumulate fatigue damage under that reversal before it opens in the forward direction.
Why does the burst-temperature mismatch matter?
The disc’s documented burst condition is 100 °C, while its operating environment is 140–150 °C. Rupture-disc strength and burst tolerance depend on temperature and material. The certified burst pressure therefore applies at its stated temperature, not automatically at the higher service temperature.
Do not calculate a corrected burst pressure from those two temperatures without the manufacturer’s certified temperature-correction data for the exact disc design and metallurgy. The direction and magnitude of the correction are design- and material-dependent. Ask the manufacturer to confirm the rated burst pressure and tolerance at 140–150 °C, then compare the resulting minimum burst value with the maximum allowable pressure and the required relief scenario.
Also verify the actual temperature at the disc connection. A branch or dead leg can run cooler during steady operation yet see rapid heating during steam admission. Those thermal cycles can combine with pressure cycles and accelerate fatigue. The remedy may require a disc selected for the real operating temperature and cycling duty, but material selection must come from compatibility, temperature, corrosion, and certified performance data rather than a generic alloy preference.
Can condensate flooding create the failure transient?
Yes. If the pressure-pump steam trap does not remove condensate at the rate it forms, the shell can partially or completely flood. Heat transfer then falls, ethylene outlet temperature drops, and the controller drives the steam valve farther open. When the trap cycles or a condensate path clears, steam can enter a cold, liquid-filled region rapidly.
That sequence can create water hammer or another short pressure impulse. The impulse may be local to the exchanger and relief branch, so neither the upstream steam-header trend nor a slow pressure recorder will show its peak. Repeated impacts also impose mechanical loads on the disc holder and piping.
Look at the temperature, valve-position, shell-pressure, and trap-cycle trends together. A repeating pattern of falling outlet temperature, increasing valve demand, sudden condensate discharge, pressure disturbance, and rapid temperature recovery points toward flooding and clearing. Tuning does not fix a restricted condensate line, failed trap cycle, poor drainage geometry, or a valve that admits steam too aggressively.
Condensing steam can also create sub-atmospheric shell pressure during isolation or cooldown. If the flare remains pressurized, the resulting reverse differential can load the disc from the discharge side. Confirm whether the installed design tolerates full expected reverse pressure or requires a support device.
What diagnostic procedure separates the causes?
- Preserve the failed parts. Record the disc orientation, tag data, holder condition, gasket condition, bolt state, and the appearance of the opening. Do not flatten or clean the disc before manufacturer examination.
- Reconcile the documentation. Compare the purchase specification, certified burst record, stamp, holder identification, material, 100 °C specified temperature, 140–150 °C service temperature, 13 kg/cm2 data-sheet value, and 11 kg/cm2 stamped value. Resolve pressure reference and backpressure treatment in writing.
- Measure the actual differential. Add suitable pressure measurement at the shell near the disc and correlate it with flare pressure. Use acquisition fast enough to capture control-valve movement, pressure-pump trap cycles, startup, shutdown, and condensate-clearing events.
- Trend the complete signal chain. Record ethylene inlet and outlet temperatures, controller output, actual steam-valve position, shell pressure, flare pressure, and trap activity on one time base. Start with the trend; adjusting the controller first can hide the initiating event.
- Check condensate removal. Confirm the pressure-pump steam trap cycles correctly, discharge is unobstructed, isolation valves are fully positioned, check valves behave correctly, and the piping drains without retaining condensate. Examine supports and low points for evidence of hammer.
- Check the control valve. Compare commanded position with actual travel. Investigate stiction, abrupt opening, leakage when closed, and excessive capacity relative to the heating duty.
- Evaluate reverse pressure. Determine the maximum flare pressure during normal operation, upset, startup, and shell depressurization. Compare it with the installed disc’s certified reverse-pressure capability and support requirements.
- Recheck the tube-leak case. Use an approved leak-test or process-composition method suited to the exchanger. A previous negative check does not replace evaluation of the 40 bar tube-side overpressure scenario.
- Review the protective-device choice. Have the process relief engineer compare a rupture disc, a pressure safety valve, or an approved combination for capacity, backpressure, cycling, leakage, fouling, discharge containment, and allowable pressure. Changing device type requires a relief calculation and formal change control.
How should the corrective action be verified?
Verify the correction across the operating states that preceded the failures: initial steam admission, warm-up, steady heating, trap cycles, temperature-demand changes, steam isolation, and cooldown. The shell and flare trends must show differential pressure remaining within the certified operating envelope in both directions. The trap must discharge without shell flooding, hammer, or correlated pressure impulses.
For a replacement disc, match the certified burst temperature to the measured service condition and document the applicable burst tolerance, pressure reference, backpressure treatment, reverse-pressure rating, holder, orientation, and installation torque requirements. Confirm that the protected equipment’s allowable pressure and the required relief capacity remain satisfied. A month without failure is useful operating evidence, but the pressure and temperature trends provide the engineering verification.
Inspect the next removed disc even if it has not opened. Permanent deformation, fretting, corrosion, holder marks, or fatigue features can expose continued reverse loading or installation stress. Maintain a record linking each disc serial identification and certificate to operating trends and inspection findings.
Which recurring pitfalls lead to repeat failures?
- Using header pressure as shell pressure: the control valve and condensate system separate those measurement points dynamically.
- Treating backpressure as a fixed subtraction: the flare acts on the opposite side of the disc and may vary or reverse the load.
- Ignoring calibration temperature: a pressure marked for 100 °C cannot be assigned unchanged to 140–150 °C service without certified correction data.
- Calling every opening a true overpressure: thermal cycling, reverse pressure, mechanical damage, incorrect installation, and fatigue can produce premature failure.
- Retuning before checking the trap: controller changes cannot remove pooled condensate or correct hydraulic shock.
- Replacing the disc with a valve by substitution: a pressure safety valve has different capacity, backpressure, leakage, stability, and installation requirements. Recalculate and approve the protective system before changing technology.
- Discarding the failed disc: fracture pattern and deformation are valuable evidence for distinguishing pressure burst, reverse buckling, corrosion, and installation damage.
Frequently Asked Questions
What happens if a rupture disc rated at 100 °C runs at 150 °C?
Its certified burst pressure cannot be assumed to remain at the stamped value. Obtain the exact design’s correction data and certified burst tolerance at the measured 140–150 °C service temperature.
What happens if the exchanger shell fills with condensate?
Heat transfer falls, the temperature controller can open the steam valve farther, and subsequent trap discharge can admit steam rapidly into a liquid-filled region. The result may be water hammer and a short local pressure spike.
What happens if flare pressure rises above shell pressure?
The disc sees reverse differential pressure. If its design lacks the required reverse-pressure capability or support, it can deform and accumulate fatigue damage.
What happens if the steam header stays at 4–5 kg/cm2?
That reading still does not rule out a shell-side transient because the header is upstream of the control valve. Measure shell pressure near the disc and correlate it with the 2 kg/cm2 flare reading and trap cycles.
When should the unit be stopped and official support contacted?
Stop under the approved operating procedure if unexplained pressure impulses continue, shell-to-flare differential approaches the certified disc limit, hammer occurs, or the disc shows deformation or installation damage. Escalate to the rupture-disc manufacturer and the responsible process relief engineer when the 13 kg/cm2 data-sheet value, 11 kg/cm2 stamp, 100 °C rating, backpressure treatment, or reverse-pressure capability cannot be reconciled; do not return the protective system to service until its basis is documented.