Expert Garage Door Repair,
Service, & Installation in
Coronado, CA

Everything most people picture about an opener is common to all of them: a head unit at the back of the garage, a rail running forward towards the opening, a trolley riding along that rail, and an arm from the trolley down to the top of the door. What differs is inside the rail. A chain drive turns a sprocket at the motor and runs a loop of chain forward to an idler at the other end, with the trolley fixed to the loop. A belt drive does the same work with a reinforced belt over a pulley. A screw drive has no loop at all — a threaded steel shaft turns inside the rail and the trolley walks along the thread. Same door, same arm, three unrelated mechanisms in between.

Which is why a sound on its own is not a diagnosis. Slack is a chain's own adjustment, so a chain knocking against its rail is pointing at an adjustment rather than a part. A belt is tensioned at the trolley end and cannot go slack in that way, so an identical noise sends you somewhere else on the machine entirely. A screw drive has nothing to tension and grease along its thread instead, which is an item of upkeep the other two do not have. The drive has to be settled before any part is named, too: a sprocket, a belt and the carriage that rides a thread are not alternatives to each other, they belong to different machines. And nothing about the building decides which of the three is up there. Somebody bought a machine on a day nobody wrote down, and it went up on whatever was already overhead.

The building record here is unusually firm, so it is worth saying what it does and does not reach. The census counts 9,521 housing units in Coronado, and 2,852 of them date from 1970 to 1979 against 1,095 for the 1980s and 641 for the 1960s — differences of 1,757 and 2,211, on a sampling error of 324 and 312 for those two comparisons. So the largest single decade here is one the census will stand behind rather than merely list. The median build year is 1974, and the margin printed on that median is a single year, about as tightly as this survey ever fixes the middle of a housing stock. 67.6% of everything counted was standing before 1980.

The same table stops short in the middle of itself. Four of its ten periods come back at 901 units for the 1950s, 837 for the 1990s, 679 for the 2000s and 641 for the 1960s, and no pair among those four separates — even the widest gap in the group, the 260 between the 901 and the 641, sits inside the 269 of sampling error on that difference. Four rows the census can count and will not rank.

None of that reaches the ceiling. That 1970s row is the closest thing to a common vintage this table has to offer, and it still does not hand over a common machine, because the year dates the building while the machine over the car was a purchase in its own right, made whenever it was made. So the reading is done overhead and in this order: what the label on the head unit says, what the rail is carried by at each end, and what is inside it driving the trolley. A chain gets checked for slack and its sprocket for wear, a belt for tension and its pulley, a screw for the state of its thread. Then the door being pulled gets looked at on its own terms, which is a separate question with a separate answer.