Direct answer: A coupling leaks when the helical gap between pin and box threads is not fully sealed by interference and thread compound, so pressurised fluid finds a path along the thread. The usual causes are insufficient or excessive make-up torque, a dry or wrong thread compound, galled or damaged threads, cross-threading, a coupling that does not match the pipe, and thermal or pressure cycling that relaxes the connection over time. Prevention is procedural as much as metallurgical: correct compound, torque to specification, careful make-up, and inspection before running. When those are controlled, leaks through an otherwise sound oilfield coupling are rare and usually traceable to a specific handling error.
On a standard API round-thread or buttress connection there is no separate elastomer seal. The seal is formed by the interference of the tapered thread flanks plus the thread compound, which fills the spiral gap and resists the passage of fluid. This is an interference-and-compound seal, so its integrity depends heavily on how the joint is made up and what fills the thread. A premium connection adds a metal-to-metal seal and sometimes a torque shoulder, but even there make-up quality governs whether the seal engages. Understanding this explains why so many “leaking couplings” are really leaking connections — the geometry was fine, but the make-up or the compound was not.
| Root cause | Effect at the connection | Prevention |
|---|---|---|
| Under-torque | Incomplete interference, weeping or gas leak | Torque to the specified value |
| Over-torque | Flank yielding, damaged seal surface | Respect torque ceiling; calibrate tongs |
| Poor compound | Voids in the thread path | Correct compound, full-length application |
| Galling/damage | Non-uniform flank contact | Inspect and dress before make-up |
| Cross-threading | Deformed helix, jump-out risk | Slow, aligned stab and first turns |
| Thread/grade mismatch | Improper seating and seal | Match coupling to pipe exactly |
| Cyclic loading | Progressive relaxation | Design margin and periodic re-check |
Gas has far lower viscosity than liquid, so it migrates through a thread path that a liquid would not pass. A connection that holds liquid pressure perfectly can still weep gas, because the molecule is small enough to travel the fine spiral gap between flanks. This is why high-pressure gas wells place greater demands on oilfield couplings: the same interference and compound that seals a liquid joint may be marginal for gas, and a connection with a dedicated seal is often specified. It also explains why a gas leak found during commissioning must be treated seriously rather than as a minor weep — the path will not close on its own.
Leak-tightness is verified, not assumed. Common methods include hydrostatic pressure testing of the joint, gas testing where gas-tightness is critical, and factory leak-detection techniques such as bubble emission under water or pressure-decay measurement. For field assurance, the connection should be pressure-tested before the string is fully committed, and any weeping or pressure drop investigated immediately rather than monitored. Where a specific test is required, agree the method, the pressure and the acceptance criterion in writing before the order is placed, so the result can be checked against a stated standard.
Where connections are critical, a torque-turn plot gives a positive record of how the joint behaved: the shoulder should be reached, and the torque should then rise in the expected manner. A joint that never shoulders, or whose torque rises abnormally fast, is a warning sign that the threads are not engaging correctly — even though it may have passed a final torque reading. Keeping torque-turn records for critical strings turns an invisible make-up process into data you can review, and it is one of the fastest ways to find a repeat leak cause.
Minor weeping sometimes stabilises as the compound redistributes, but a genuine leak path through damaged or under-made threads usually requires the joint to be broken out and re-dressed or the connection replaced. Do not rely on a leak “sealing itself” in a pressurised well.
Round thread can be acceptable on many wells, but for high-pressure gas a connection with a dedicated seal is usually specified, because gas migrates more readily through a thread path than liquid does.
Not beyond the specified value. Over-torque damages the flanks and makes leaks more likely. The fix is correct torque with correct compound.
The make-up torque, the compound used, the joint serial, the operator and the test result. Traceable records are the fastest way to find the common factor when a connection leaks.
Set the specification, then control the make-up. For oilfield couplings, the specification and the field procedure are two halves of the same reliability guarantee. A leak-prevention routine is short: match the coupling to the pipe, inspect and dress the threads, apply the correct compound, torque to the table, pressure-test, and record the result. Catalogue examples for reference include the API 5CT L80 STC short coupling (4 1/2 in) for gathering-pipeline duty and the API 5CT L80 STC short coupling (5 1/2 in); both are published listings, and both rely on correct make-up for their seal. See the full range of oilfield couplings on the Oilfield Couplings page, and specify your test requirement when you enquire so the evidence you need is built into the order.
Direct answer: A coupling leaks when the helical gap between pin and box threads is not fully sealed by interference and thread compound, so pressurised fluid finds a path along the thread. The usual causes are insufficient or excessive make-up torque, a dry or wrong thread compound, galled or damaged threads, cross-threading, a coupling that does not match the pipe, and thermal or pressure cycling that relaxes the connection over time. Prevention is procedural as much as metallurgical: correct compound, torque to specification, careful make-up, and inspection before running. When those are controlled, leaks through an otherwise sound oilfield coupling are rare and usually traceable to a specific handling error.
On a standard API round-thread or buttress connection there is no separate elastomer seal. The seal is formed by the interference of the tapered thread flanks plus the thread compound, which fills the spiral gap and resists the passage of fluid. This is an interference-and-compound seal, so its integrity depends heavily on how the joint is made up and what fills the thread. A premium connection adds a metal-to-metal seal and sometimes a torque shoulder, but even there make-up quality governs whether the seal engages. Understanding this explains why so many “leaking couplings” are really leaking connections — the geometry was fine, but the make-up or the compound was not.
| Root cause | Effect at the connection | Prevention |
|---|---|---|
| Under-torque | Incomplete interference, weeping or gas leak | Torque to the specified value |
| Over-torque | Flank yielding, damaged seal surface | Respect torque ceiling; calibrate tongs |
| Poor compound | Voids in the thread path | Correct compound, full-length application |
| Galling/damage | Non-uniform flank contact | Inspect and dress before make-up |
| Cross-threading | Deformed helix, jump-out risk | Slow, aligned stab and first turns |
| Thread/grade mismatch | Improper seating and seal | Match coupling to pipe exactly |
| Cyclic loading | Progressive relaxation | Design margin and periodic re-check |
Gas has far lower viscosity than liquid, so it migrates through a thread path that a liquid would not pass. A connection that holds liquid pressure perfectly can still weep gas, because the molecule is small enough to travel the fine spiral gap between flanks. This is why high-pressure gas wells place greater demands on oilfield couplings: the same interference and compound that seals a liquid joint may be marginal for gas, and a connection with a dedicated seal is often specified. It also explains why a gas leak found during commissioning must be treated seriously rather than as a minor weep — the path will not close on its own.
Leak-tightness is verified, not assumed. Common methods include hydrostatic pressure testing of the joint, gas testing where gas-tightness is critical, and factory leak-detection techniques such as bubble emission under water or pressure-decay measurement. For field assurance, the connection should be pressure-tested before the string is fully committed, and any weeping or pressure drop investigated immediately rather than monitored. Where a specific test is required, agree the method, the pressure and the acceptance criterion in writing before the order is placed, so the result can be checked against a stated standard.
Where connections are critical, a torque-turn plot gives a positive record of how the joint behaved: the shoulder should be reached, and the torque should then rise in the expected manner. A joint that never shoulders, or whose torque rises abnormally fast, is a warning sign that the threads are not engaging correctly — even though it may have passed a final torque reading. Keeping torque-turn records for critical strings turns an invisible make-up process into data you can review, and it is one of the fastest ways to find a repeat leak cause.
Minor weeping sometimes stabilises as the compound redistributes, but a genuine leak path through damaged or under-made threads usually requires the joint to be broken out and re-dressed or the connection replaced. Do not rely on a leak “sealing itself” in a pressurised well.
Round thread can be acceptable on many wells, but for high-pressure gas a connection with a dedicated seal is usually specified, because gas migrates more readily through a thread path than liquid does.
Not beyond the specified value. Over-torque damages the flanks and makes leaks more likely. The fix is correct torque with correct compound.
The make-up torque, the compound used, the joint serial, the operator and the test result. Traceable records are the fastest way to find the common factor when a connection leaks.
Set the specification, then control the make-up. For oilfield couplings, the specification and the field procedure are two halves of the same reliability guarantee. A leak-prevention routine is short: match the coupling to the pipe, inspect and dress the threads, apply the correct compound, torque to the table, pressure-test, and record the result. Catalogue examples for reference include the API 5CT L80 STC short coupling (4 1/2 in) for gathering-pipeline duty and the API 5CT L80 STC short coupling (5 1/2 in); both are published listings, and both rely on correct make-up for their seal. See the full range of oilfield couplings on the Oilfield Couplings page, and specify your test requirement when you enquire so the evidence you need is built into the order.
english
français
Deutsch
Italiano
Русский
Español
português
Nederlandse
ελληνικά
日本語
한국
العربية
हिन्दी
Türkçe
indonesia
tiếng Việt
ไทย
বাংলা
فارسی
polski