Sucker Rod Back-Off: Why Rod Connections Unscrew Downhole

What is sucker rod back-off?

Sucker rod back-off is a rod connection that unscrews downhole. The pin and coupling come apart at the threads, and the string below the joint is lost, just as if a rod had broken. Most back-offs start when the connection loses the pin preload it was made up with, so the joint works loose stroke by stroke. ¹²

A sucker rod joint, a pin screwed into a rod coupling, is held together by clamping force, not by the threads alone. When the coupling is tightened past hand-tight, the pin stretches like a spring and presses the coupling face hard against the pin shoulder. ³ The friction on those two faces is what keeps the joint from turning. The threads mainly supply the squeeze; on their own they play only a minor role in keeping the coupling tight. ¹

The word comes from drilling, where "back off" means unscrewing pipe downhole on purpose, usually to recover part of a stuck string. The same glossary notes that backing off can also happen by accident. ⁹ On a rod-pumped well, it is always the accident.

A back-off is not a parted rod. A part is a fracture; a back-off is a connection that turned until it let go. The two are related: a loose joint that doesn't unscrew fully often breaks at the pin or coupling instead. ² For breaks, fatigue and reading a break face, see why sucker rods part.

Why sucker rod connections loosen

Connections loosen when the clamping force between the pin shoulder and coupling face is lost. Under-makeup, dirty or damaged shoulders, too much lubricant, coupling-to-tubing impacts, rod whip, shock loads from fluid pound and pump tagging, and flexing of the lower string from buckling all do it. Once the faces separate, the joint can work loose. ¹²³

The clamping pressure on the shoulder varies through every pumping cycle. In some joints the pumping load overcomes it, and the pin and coupling faces separate. Under those conditions, one make-up paper says, the coupling "eventually loses all prestress and backs off." ¹ A later paper puts the rule plainly: if the pin preload stress is greater than the applied load, no failure should occur. ³

Thread geometry works against you once the clamp is gone. A thread is an inclined plane wrapped around a shaft, and its slope tends to turn the coupling loose when there is little or no friction on the shoulder. ¹ The usual ways that friction is lost:

• Under-makeup. A joint made up short of the right preload starts with too little clamp. Hand wrenches and torque readings are unreliable ways to get it; see make-up below. ¹³
• Over-makeup. Too much displacement can yield the pin. A yielded pin can't hold preload, and the joint goes loose under well load. ¹
• Dirty or damaged shoulders. The joint depends on friction at the shoulder, so it must be clean, dry and undamaged. Hammering a coupling to break it out damages it. ²
• Too much lubricant. Lubricant belongs on the pin threads. On the shoulder it lowers the friction that keeps the joint from turning, much like sliding a box across an oily deck. ¹³
• Rod whip and tubing impacts. A whipping, vibrating string loosens couplings. Each coupling impact on the tubing helps the thread slope walk it loose, like vibrating a box down a ramp. ¹
• Shock loads. Partial pump fillage, stuck or sticking plungers and tagging all send shock through the string. ² Fluid pound jars the whole rod string and has been tied to coupling and pin failures. ⁵
• Unanchored tubing, deviated holes and over-pumping. All three appear on the same list of possible causes of lost displacement. ²
• Prior failures. Connections rerun without being broken out, cleaned and remade are already loose. ²

Compression and buckling flex the joint

When the bottom of the string goes into compression on the downstroke, the rods bow and flex. That bending works directly on the connections: it pries the shoulder faces apart, drives couplings into the tubing wall and adds a bending load to the pin. Rod buckling is on the published list of causes of lost pin preload. ²³

A rod connection is designed for tension. Its clamp resists loads that try to pull the joint apart along the rod axis. ³ Bending is a different load. Once the pin shoulder and coupling face open on one side, a bending moment adds to the tension in the pin, and the joint is on its way to a loss of tightness. ²

One connection-failure paper shows pin failures from lost displacement due to "excessive pumping, rod buckling or fluid pound," and its remedy for the typical case is short: eliminate shock loads. ² The revised make-up guidance lists rod buckling tendencies, fluid pound, vibration and contact wear among the operating conditions that affect pin preload. ³ A bowed string also knocks couplings against the tubing, the impacts the make-up paper says loosen couplings. ¹

For why the lower string buckles, where the neutral point sits and how to spot it on a card, see the sucker rod buckling guide. For the usual mitigation, see sinker bars and rod guides.

Rotation and torque: when the string turns the wrong way

Turning a rod string the right way tightens its connections; turning it the wrong way unscrews them. Rod rotators on beam-pumped wells turn the string to the right so couplings keep tightening. On progressing cavity pump drive strings, backspin after the drive stops can unscrew right-hand connections. ⁶⁸

API rod connections are right-hand threads. A rod rotator on the walking beam turns the string a little on each stroke, and it is set up so that the rotation "is directed to the right so as to apply a continuous tightening effect on the couplings." ⁶

Rotary-driven strings are a different case. On a progressing cavity pump, the string carries torque all the time. When the surface drive shuts off, the string backspins, and one patent notes that this can unscrew one or more right-hand connections, leaving the rods disconnected downhole. ⁸ On a beam-pumped sucker rod pump, look first at make-up and operating loads.

How a back-off shows up

From the surface, a back-off looks like a parted rod: the polished rod load drops, the dynamometer card shrinks and production stops. The difference shows when the rods come out. A back-off leaves a coupling still threaded on one pin and no fracture face, often with produced fluid squirting from loose joints above and below. ²⁵⁷

On the card, a separated string behaves like a part, because the string below the joint is no longer attached. For a part in the upper string, the card area decreases and the minimum load usually falls below normal: the fluid load and part of the rod weight are gone from the load cell. ⁵ A deep separation is harder to see, since relatively little rod weight is lost. ⁵ For reading those shapes, see the dynamometer card guide.

When the rods are pulled, look for these signs:

• A coupling left on a pin. An unscrewed coupling stays threaded on the upper end of the next rod down and has to be fished out before repairs can go on. ⁷
• No fracture. The threads are intact and there is no crack face. If there is a break instead, read it with the table in why sucker rods part.
• Fluid in the joints. A properly made-up connection holds no well fluid. Fluid squirting out of connections as they are broken out means the pin and coupling separated in service and pumped fluid in. Crews usually find it in many joints above and below the failure. ²
• Galled or torn threads. Flat, torn threads point to poor make-up, dirty threads or a bad start before the power tongs were applied. ²

Note the depth where fluid first appears in the joints. That is the start of the loose section.

Make-up practice: circumferential displacement

The accepted way to put correct preload in a rod connection is circumferential displacement: make the joint hand-tight, scribe a line across pin shoulder and coupling, then turn the coupling a set distance. The distance comes from the rod manufacturer or API RP 11BR tables, not from torque alone. ¹³

Torque is a poor guide to preload. A rough or dirty thread needs far more torque to reach the same prestress than a clean, lubricated one, because much of the torque goes into friction. ¹ Power tong pressure drifts as the hydraulic fluid heats up, so pressure alone is "not a viable means" of reaching minimum preload. ³ Displacement measured from hand-tight doesn't depend on joint friction. ¹

The revised make-up procedure, in outline: ³

1. Pick the right coupling for the pin thread (sucker rod, polished rod or sinker bar).
2. Clean both pins and both coupling ends. Check the threads and bearing faces.
3. Put a small amount of qualified thread lubricant on the pin's lead threads only. Keep the pin shoulder and coupling face clean and dry.
4. Make the coupling up hand-tight on both ends, with the shoulder faces seated all the way round (the guidance sets a maximum gap). Replace a coupling that won't seat.
5. Scribe a vertical line across both pin shoulders and the coupling.
6. Tighten to the displacement value for the rod grade, size and condition (new or rerun), from the manufacturer or API RP 11BR.
7. Check both ends against the marks.

The older make-up paper adds field points. New rods of the softer grades should be made up and broken out once before measuring, so the threads seat. Set power tongs by the displacement method on the first few joints and recheck them as the job goes on. Don't snap the tong trigger: each extra surge adds pin stress until the pin can yield. And don't exceed the recommended displacement. ¹

Connection life depends on field handling from the first make-up on, ⁴ and the lubricant matters too: use one qualified for sucker rod make-up. ³¹⁰

How to prevent sucker rod back-off

Prevention has two halves: put the right preload in every joint at the surface, and stop the downhole loads that take it out. Make up by displacement, keep shoulders clean, remake every loose joint after a failure, and remove the shock, whip and compression that flex the lower string. ¹²³

1. Make up every joint by circumferential displacement. Use the manufacturer's or API RP 11BR values for new and rerun rods. ³
2. Keep the shoulders clean and dry. Lubricate the pin threads only, and never hammer a coupling to break it out. ²³
3. Remake the whole loose section after a failure. Segregate every rod from the depth where fluid first came out of the joints down to the pump. Break, dry, re-lubricate and remake each joint, and replace corroded pins and couplings. Otherwise, one paper warns, connection failures keep spreading. ²
4. Stop fluid pound and tagging. Shock loads are the first remedy listed for lost displacement. ² Match pumping to inflow; see the fluid pound guide and pump-off controller vs. AI well control.
5. Anchor the tubing. Unanchored or poorly anchored tubing is on the list of causes. ² See the tubing anchor entry.
6. Take the compression out of the bottom of the string. Less bending at the joints means less prying at the shoulders. Sinker bars and guides manage buckling; a pump that keeps the string in tension removes it.
7. Find the root cause of the first failure. Remaking joints fixes the symptom. Review the operating practices behind it, or the next back-off follows. ²

Where SRABS fits

SRABS does not replace good make-up. It targets one of the downhole causes above: compression at the bottom of the string, which flexes the lower connections on every downstroke.

The SRABS Pump is a drop-in anti-buckling rod pump that keeps the rod string in tension where it would otherwise buckle, so the bottom of the string never goes into compression. SRABS Brain reads dynacards, sets pump speed and prevents fluid pound, the other shock load named above. Clean shoulders, correct displacement and remaking loose joints stay with your rod program; no pump fixes those. See how it works and the solutions, and to estimate what fewer rod failures would mean for your fleet, run the ROI calculator. Its default 100-well scenario is modelled, not a field result.

The bottom line

A back-off is a connection that lost its clamp. Put the right preload in at the surface with circumferential displacement, keep the shoulders clean, remake every loose joint after a failure, and take the shock and compression out of the downstroke.

If back-offs and pin failures keep showing up in the bottom of your string, test that on one well.

FAQ

Sources

1. Morris, R. A., "Sucker Rod Make-Up," Southwestern Petroleum Short Course, 1973. https://www.swpshortcourse.org/paper/sucker-rod-make
2. Hein, N. W., Jr., Malone, S., "Sucker Rod Connection Failures: Why They Occur and What to Do About Them," Southwestern Petroleum Short Course. https://www.swpshortcourse.org/paper/sucker-rod-connection-failures-why-they-occur-what-do-about-them
3. Stevens, R., Hein, N. W., Jr., Newman, F., Conquergood, S., "API 11BR Revisions on Sucker Rod Makeup," Southwestern Petroleum Short Course. https://www.swpshortcourse.org/paper/api-rp-11br-revisions-sucker-rod-makeup
4. Newman, F., Huseman, J., "Predicting the Useful Life of a Sucker Rod Connection," Southwestern Petroleum Short Course. https://www.swpshortcourse.org/paper/2014017-predicting-useful-life-sucker-rod-connection
5. US Patent 4,509,901, "Method and apparatus for detecting problems in sucker-rod well pumps" (1985). https://patents.google.com/patent/US4509901
6. US Patent 5,433,128, "Sucker rod coupling tool and method" (1995). https://patents.google.com/patent/US5433128
7. US Patent 5,265,927, "Downhole fishing tool" (1993). https://patents.google.com/patent/US5265927
8. US Patent 10,030,453, "Device for the connection of rods for rotational drive of a downhole pumping apparatus" (2018). https://patents.google.com/patent/US10030453
9. SLB Energy Glossary, "back off." https://glossary.slb.com/en/terms/b/back_off
10. Tietz, E., Sriraman, A., "Lubrication Selection Using Circumferential Displacement of Sucker Rods," Southwestern Petroleum Short Course. https://www.swpshortcourse.org/paper/lubrication-selection-using-circumferential-displacement-sucker-rods