Garage Door Slams Shut or Closes Too Fast

The most serious version and the easiest to confirm. A broken torsion spring shows a clean gap in the coil above the opening, roughly an inch or two, as though a section had been removed.

With that gone there's no counterbalance resisting anything. The door doesn't close — it falls, and the only things slowing it are the opener's drive train and whatever friction the rollers contribute. Neither is a braking system, and the drive train in particular is a nylon gear that was chosen precisely because it's cheap to replace.

What makes this the urgent one is that the descent is the dangerous part. A door with a broken spring that's currently closed is stable and harmless. A door with a broken spring that's raised, or that gets raised, is a few hundred pounds with a clear route to the floor.

The full picture on broken springs is covered on its own page. What matters here is the instruction: don't open it again.

How you can tell: A visible gap in the coil, usually with a bang somebody in the house remembers. The door came down markedly faster than normal, or came down entirely on its own.

The gradual version, and the one that reaches this page after passing through several milder symptoms first.

Springs lose lift as they fatigue. Long before one breaks, the counterbalance is producing measurably less resistance than it used to, and on the way down that shows up directly as speed. The door still opens fine — the opener has authority to spare — so nobody connects the fast close to anything until it gets alarming.

The characteristic pattern is acceleration. The door starts down at something like its usual pace and picks up speed all the way, because the leftover weight is pulling on it for the full height of the opening. By the last few feet it's moving considerably faster than it ever did, and it lands hard.

Repeated hard landings do their own damage. The bottom section takes the impact every time, and a section that has been slammed into concrete a few hundred times develops bends and cracks that eventually become their own repair.

One variant worth separating out, because the fix is different: sometimes the springs never had the lift in the first place. Winding a torsion spring means applying a specified number of turns, and DASMA names coming up short as a real failure mode — “insufficient torque can present problems, as well, resulting in a door drifting down from its open position over time.” A door that has drifted down from fully open since the day the springs went in, and that closes fast on top of it, is describing turns rather than fatigue. The tell is history: fatigue gets worse from a known baseline, an under-wound door was like this from the start.

How you can tell: Descent has been getting faster over months. The door lands harder than it used to. It probably stopped holding position when raised by hand some time before this started. If instead it has drifted down from fully open since the springs were last replaced, suspect the winding turns rather than fatigue.

Worth ruling out early, because it can produce a startlingly fast descent on a door whose springs are only moderately tired.

When the emergency release is pulled and never re-engaged, the door is on the springs alone. The opener's drive train had been contributing a small amount of drag and a lot of positional control, and without it the door does whatever the counterbalance tells it to. On a door with healthy springs that means a controlled descent. On a door with springs that have quietly fallen behind, it means the shortfall is suddenly visible.

This is genuinely common after a storm evening, when people pull the release to work the door by hand and never reconnect it. Reconnecting is straightforward, and the method plus the safe sequence are covered on the won't-stay-up page rather than repeated here.

The important part is what happens after you reconnect. If the door then closes normally, the trolley was the whole issue. If it still comes down fast, you've just confirmed a counterbalance problem the opener had been covering for.

How you can tell: The release cord hangs slack and the carriage on the rail looks separated. The opener may run without moving the door at all.

Doors with extension springs — the long ones running back alongside the horizontal tracks — behave differently here, and worse in one specific respect.

Extension springs act independently on each side. If one parts, that side loses its counterbalance entirely while the other keeps working, so the door comes down fast and unevenly. It drops on the failed side, racks in the opening, and the loading on the rollers and track goes somewhere they weren't designed to take it.

There's a second concern specific to this setup. A stretched spring under load that parts can travel along its own length, which is what safety containment cables are for — a steel cable threaded through the middle of each spring and anchored at both ends, whose only job is to catch the pieces. Plenty of older Valley doors don't have them, either because they predate the practice or because somebody replaced springs and left them out.

How you can tell: Springs alongside the upper tracks rather than a shaft above the door. One visibly broken or slack. The door drops on that side and twists as it goes.

If the door drops with one corner running ahead of the other, this isn't a spring fault and it's arguably the most urgent thing on the page.

What that describes is a lift cable that has come off its drum, snapped, or a drum that has slipped on the shaft. That corner has lost its support, so it falls faster than the supported side, and the door twists in the opening as it goes. Every occurrence drives the rollers into the track sideways, and the door is now one or two cycles from leaving its rails entirely.

It's worth being explicit, because the mistake gets made constantly: a broken torsion spring does not cause a one-sided drop. Both springs on a two-spring door sit on the same shaft and act as one counterbalance, so a spring failure makes the door uniformly fast, not lopsided. One-sided means cables and drums.

How you can tell: One corner clearly leading. The door looks twisted mid-travel. Often a slack cable visible, or one sitting outside its drum groove.

The deceptive one, because it presents as an intermittent fault on a door that's otherwise behaving.

Something is binding — a roller with a seized bearing, a bent track section, a hinge that has worked loose, debris packed at one point in the vertical track. On the way down the door meets that resistance and hesitates. For a moment the friction is holding weight the springs no longer are. Then the door overcomes it, all at once, and drops from that point with everything it has been accumulating.

What makes this genuinely worse than a steady fast close is the surprise. A door that's consistently fast at least tells you what it's going to do. A door that hesitates and then lets go does the dangerous part without warning, and it does it at a different height each time.

It also almost always means two faults rather than one, since a healthy counterbalance would have held the door regardless of what the friction did.

How you can tell: The descent stutters, pauses, or catches at a point in the travel, then the door drops. Often a scraping or clunking noise at the moment it catches.

Worth separating out, because it brings a lot of people to this page and it's a much smaller problem.

If the door travels at a normal speed and then bangs hard at the very bottom, what you're usually hearing isn't a falling door. It's the opener's down travel limit set past where the floor actually is. The door seats on the concrete, the opener hasn't reached the position it was given, so it keeps driving — and the noise is the motor loading the door into the slab.

Two things follow from that. First, it's an opener setting rather than a counterbalance failure, and it's covered on the opener travel limits page. Second, it's still worth correcting: the bottom section takes that load every cycle, and repeated hard seating is how a section ends up bent or cracked.

The tell that separates it cleanly is the speed of everything before the bang. Normal travel plus a loud finish is this. Fast travel plus a loud finish is the rest of this page.

How you can tell: The descent looks like it always has. The noise is confined to the last inch, and often the door bounces slightly or reverses after touching down.