Twenty Pump Card Signatures, Drawn and Measured
What each sucker-rod pump condition does to the pump card — and the one measurement that separates it from the conditions it resembles
How to read these
A pump card is a closed loop of load against plunger position over one stroke. Its four corners are the four events of the pump cycle: the standing valve closing and the fluid load transferring onto the travelling valve at the bottom, the plunger lifting the column, the travelling valve opening and the load transferring back at the top, and the plunger falling through the fluid to the bottom again. On every plot below, the dashed rectangle is the card's bounding box — the reference the corner clearances are measured against — and the orange marker is the bottom-right corner, where the travelling valve finishes opening. Net plunger travel is measured to that corner, and pump fillage is net plunger travel over gross plunger travel, on the plunger stroke, never on the surface stroke.
An ideal card is a rectangle and no real one is: fluid-load transfer is not instantaneous and it straddles the stroke turnaround, so a healthy card's corners are visibly cut away. Watch what happens to those corners. Nearly every signature below is a corner that has been gouged out, a corner the card now reaches that it should not, or — for a few of them — nothing to do with shape at all.
The signatures are grouped by which measurement separates them, because that is the honest grouping: modes inside a family look alike, and modes in different families are rarely confused. Full method, equations and limits are on the dynamometer card method page; to run these against a card of your own, use the signature screen.
Family 1 — fluid load shed late on the downstroke
All four gouge the card's lower-right corner: the rods keep carrying fluid load past the top of the stroke and shed it lower down. The width of that transfer window is what separates them.
Fluid pound
What the card does. Full fluid load is held into the downstroke and then collapses almost vertically. The upstroke also droops slightly as the stroke proceeds.
Mechanism. The barrel is only part filled, so the plunger falls through empty space carrying the whole fluid column until it slaps the fluid surface. A starved barrel also lets fluid slip past the plunger for the whole upstroke, which is the droop.
What separates it. The collapse is abrupt — under about a fifth of the stroke, against better than 0.4 of it for gas interference. That width is the whole difference between the two.
Gas interference
What the card does. Load bleeds off over a long slanted lower branch instead of dropping at the top of the stroke. The lower-right corner is rounded away and the lower branch slopes upward with position.
Mechanism. Free gas in the barrel has to be compressed back up to discharge pressure before the travelling valve can open, and compression is gradual.
What separates it. Transfer width, and the upward slope of the lower branch. A fluid pound collapses; gas interference bleeds.
Standing-valve leak
What the card does. Load sheds slowly down the downstroke while the top of the card stays square.
Mechanism. Fluid bypasses the standing valve, so it cannot take the fluid column back off the rods at the top of the stroke; the rods keep carrying part of it until the plunger is moving down fast enough to out-run the leak.
What separates it. The shelf carries much less load than a fluid pound's, and it is held nearly to the bottom of the stroke rather than released part-way down.
Pump tagging up
What the card does. A distinct impact spike at maximum position, standing clear of the upstroke plateau, on a card body that is otherwise recognisable.
Mechanism. The plunger strikes the top of the barrel or the pull tube. While it is jammed there the fluid load has nowhere to go, so it does not transfer at the turnaround at all — it comes off as the plunger backs away, which also opens the lower-right corner.
What separates it, and the repair. Peak-load prominence at the top of the stroke. Space the pump down.
Family 2 — load picked up late on the upstroke
Both round away the card's top-left corner while the lower-right stays square — the mirror image of family 1.
Travelling-valve leak
What the card does. Load builds slowly up the first third of the upstroke instead of arriving at the turnaround. The top-left corner is rounded away; the lower-right stays square.
Mechanism. Fluid bypasses the travelling valve at a rate set by the pressure across it, while the rate the plunger displaces fluid is set by its velocity — which is zero at the turnaround. Near the bottom of the upstroke the leak wins outright and no load develops.
What separates it. Which corner is missing. This one loses the top-left; family 1 loses the bottom-right.
Delayed travelling-valve closure
What the card does. An exponential load pickup confined to the very start of the upstroke, and then a normal card.
Mechanism. A weak or fouled valve spring closes the travelling valve late.
What separates it — and where it fails. The curvature of the pickup, over a short window. When the delay is short the affected window is a tenth of the stroke and the card is otherwise healthy, at which point this signature is genuinely lost. It is one of two modes this method will read as normal.
Family 3 — fluid load lost part-way up the upstroke
Both take load off the upstroke before it reaches the top. A cliff or a decay — that is the difference.
Plunger out of barrel
What the card does. The upstroke falls off a cliff mid-stroke and runs flat and low to the top. The top-right corner is missing entirely.
Mechanism. The pump is spaced so long that the plunger leaves the top of the barrel before the upstroke ends, dumping the fluid load the moment it clears.
What separates it, and the repair. Top-right corner clearance, which is several times a normal card's. Space the pump down.
Worn barrel
What the card does. Upstroke load decays progressively away from the plateau; card area is lost gradually.
Mechanism. A worn plunger-to-barrel fit slips more as the fluid column builds.
What separates it. Gradual decay, not a cliff. The upstroke branch slope is negative and the top-right corner is soft rather than absent.
Family 4 — mechanical contact at one end of the stroke
These two were one label until it was noticed that they are opposite mechanisms with opposite repairs. Collapsing them loses the only thing the diagnosis is for.
Pump tagging down
What the card does. Load dips sharply below the downstroke line at minimum position. The card reaches its own bottom-left corner, which a healthy card never does.
Mechanism. The plunger strikes the standing valve or the bottom of the barrel and the rods go into compression, so minimum load coincides with minimum position.
What separates it, and the repair. Bottom-left corner clearance near zero, and the stroke position of the load minimum. Space the pump up — the opposite of tagging up, which is why the two are separate classes.
Family 5 — the card's scale, not its shape
A normalised card shape cannot see these at all. They are read off absolute load and absolute travel, which is why a card must arrive in real units to be screened for them.
Rod parting
What the card does. A perfectly ordinary-looking loop — at a small fraction of the load a working string carries. Load range collapses against mean load.
Mechanism. Below the break nothing is lifted, so the surviving string carries only its own buoyant weight plus friction and inertia. That still traces a closed loop: friction opens it, and rod inertia tilts both branches upward with position.
What separates it. Absolute load, not shape. Read the axis, not the outline.
Gas lock
What the card does. A low-load loop with almost no enclosed area — the valves never open.
Mechanism. Gas trapped in the barrel is compressed and expanded without ever reaching discharge pressure. This is gas interference past the point where the card still has corners to gouge.
What separates it. Card area at low absolute load. A parted rod string draws a proper loop; a gas-locked pump draws barely any.
Stuck pump
What the card does. Near-zero position range; a tiny collapsed card at working load.
Mechanism. Debris, scale or a mechanical obstruction stops the plunger while the surface unit keeps stroking.
What separates it. Plunger travel of a few inches at loads that are otherwise normal. Gas lock has the travel and not the load; this has the load and not the travel.
Family 6 — stroke length against the expected stroke
These three differ only in absolute plunger travel. Under a normalised card shape they are literally the same measurement — which is why a screen that works on shape alone cannot separate them, and one that keeps real units can.
Normal
What the card does. Full expected plunger travel, tight corners, flat branches, high fill ratio. The corners are cut away slightly — that is the valve-transfer window, not a fault.
Mechanism. The reference case: the pump fills, lifts, transfers and falls as designed.
What separates it. Nothing distinguishes a healthy card except the absence of every other signature — which is exactly why absence of a signature is weak evidence.
Tubing movement
What the card does. An elongated card — plunger travel longer than the pump geometry accounts for — with a load slope across the loop.
Mechanism. Unanchored tubing stretches and contracts with the fluid load through the stroke.
What separates it, and the repair. Absolute stroke length against the expected plunger stroke. Consider a tubing anchor.
Plunger undertravel
What the card does. A correctly shaped card over a badly truncated position range.
Mechanism. The plunger never travels the stroke the surface unit is making — rod stretch, tubing movement, or a spacing problem.
What separates it. Absolute stroke length again, in the other direction. On a normalised plot this card and a healthy one are indistinguishable.
Family 7 — loop thickness and load bias
Neither is a valve or fillage problem. Both change what the loop encloses rather than which corner it keeps.
Excessive friction
What the card does. A thick loop: the upstroke carries extra load and the downstroke carries less, opening the card and squaring its corners.
Mechanism. Rod-on-tubing drag, a tight stuffing box, or a deviated hole.
What separates it. Fill ratio approaching one with corner clearances near zero — a card squarer than valve transfer permits.
Bent barrel
What the card does. Load biased with position, shifting the card's centroid off centre. The bias runs in either direction.
Mechanism. A bent barrel binds at one end of the stroke — which end depends on the bend, so this one fault draws two mirror-image cards. Both are carried as reference signatures for exactly that reason.
What separates it. Branch slopes of the same sign on both branches, with the loop otherwise intact.
Family 8 — the texture of the trace
Roughness alone cannot separate these: sand and vibration both spoil a clean trace. Periodicity does — vibration is a wave, sand is scatter.
Sand abrasion
What the card does. A jagged, aperiodic load trace over a slightly slanted card.
Mechanism. Grains passing through the plunger-barrel fit; the fit itself wears open, which adds the slant, because a worn fit slips at a rate that follows the differential across the plunger.
What separates it. High roughness with low periodicity. The jaggedness is grain-by-grain and therefore not a wave.
Excessive vibration
What the card does. A regular high-frequency oscillation riding on an otherwise normal card.
Mechanism. Mechanical resonance or imbalance in the surface unit — not a pump condition at all, which is worth establishing before anyone pulls a pump.
What separates it. High roughness with high periodicity: the residual autocorrelates because it is a wave.
Paraffin restriction
What the card does. Concave dents in the trace plus a raised friction signature on an otherwise normal card.
Mechanism. Paraffin narrows the tubing and grabs the rods intermittently.
What separates it — and where it fails. Dent count and depth. When the dents are shallow this card is a normal card with a little noise on it, and the signature is genuinely lost. It is the other of the two modes this method will read as normal.
What these examples are not
- They are not field cards. Every plot on this page is synthetic, drawn at the centre of a published parameter range. A real card carries measurement noise, a rod string that does not match the model, and often more than one fault at once.
- Matching a shape is not a diagnosis. Several conditions here draw nearly the same card: gas interference, fluid pound and a standing-valve leak all gut the lower-right corner, and two of them are told apart by the width of one transition that measurement noise can move.
- Two of the twenty can read as normal. A shallow paraffin restriction and a briefly delayed valve closure both sit on an otherwise healthy card body. Absence of a signature is not evidence of a healthy pump.
- No accuracy figure is published for this method, deliberately. Checking synthetic cards against their own generators measures self-consistency, not field performance. A number belongs here only when there is a benchmark against labelled real cards to put behind it.
- Nothing here predicts production. Fillage and displacement give an inferred rate for the stroke recorded; what a well would make after an intervention is not on the card.
Where these come from
The shapes are drawn from published sucker-rod pump signature descriptions and from the geometry of the pump cycle itself — the valve-transfer window that rounds a healthy card's corners, the compression of free gas that slants a lower branch, the rod inertia that tilts both branches of a parted string. Each card is generated deterministically, so the same condition always draws the same reference card and the measurements underneath it are reproducible rather than asserted. The signature screen runs the same generators and the same measurements against a card of your own, and the method page carries the wave equation, the fillage and pressure relationships, and the full limits.
Open the signature screen Read the method Run this on your own cards