On Fri, 25 Jan 2013 01:50:29 +0000, Danny D. wrote:
FINAL UPDATE ... LESSONS LEARNED:
The evidence leads us to conclude a bad left bearing caused the torsion bar to shift 1/8" to 1/4" laterally leftward when the 16-foot wide by 7
1/2 foot tall steel Wayne-Dalton Foamcore garage door opened.
That tiny lateral slip apparently caused the steel cable to slip off the right cable drum & then to tangle around the torsion bar, halting movement of the garage door on one side, causing about a foot in tilt.
In summary: a) Both bearing end plates steel bearings were removed & replaced. b) The left cable drum was damaged so it was also replaced. c) Both 1/16" x 8'5 3/8" lift cables were replaced (w 1/8" thick cables). d) The spring nylon bushing was replaced with a steel 1 3/4" bearing. e) Both bottom bracket steel rollers were straightened and replaced. f) An additional lag screw was used to stiffen each bearing end plate. g) The entire job had to be done with the door static, on the ground.
Total out-of-pocket cost was about $25.
No need to read further for a summary; however, there were many lessons learned - some of which are detailed below (with explanatory photos).
Undoing the tangles took effort because I had to unwind the spring and then lift the door by hand and support it on the ladder, as I manually untangled the cables.
Initially rewinding the cables onto the drums and rewinding the spring and manually operating the door, twice, caused BOTH cables to tangle on the drum within a minute or two of hand operation.
Paradoxically, the torsion bar could not be twisted by hand without adding leverage, which lead me to inspect the left bearing, which was (later) found to have disintegrated. These semicircular scrape marks were a clue that only later did I understand the significance of:
formatting link
The 0.250x35x1.75LH spring, which was recently replaced (May 2011), took only 27 1/4 quarter turns to unwind, but re-winding it took 29 quarter turns (go figure) to balance the 6 1/2 foot tall door with 4" cable drums.
Inspecting all parts, I noticed a deep gouge was left in the torsion bar from the tangled cables, but I had no access to a new torsion bar.
formatting link
Removing the spring revealed the torsion bar had been crimped vastly too deeply, which caused problems later on because neither the new steel bearing for the spring anchor plate nor the winding end of the spring would pass the deformed section (so everything had to be assembled from the left side).
formatting link
A few rollers were mangled & had to be removed & replaced. This turned out to be easier than I at first thought it would be. After some experimentation, I found the easiest way to R&R the bottom bracket and cable with the door closed (hence limited space) was to simply flip the bottom bracket UPSIDE down as shown in this photo below:
formatting link
Inspection of the bearing end plates showed that the left drum had scraped against the steel end plate, and, in the end, with some difficulty, I pressed out the old bearing and inserted new bearings.
In hind sight, the bearing R&R would have been tremendously easier had I the right tools, which consist of a 1.25" ID x 2"OD half-inch tall steel collar and a 2" ID inch-tall steel pipe or 2" hole cut into wood.
Or, better yet, the lesson learned is to simply buy two new bearing end plates with the bearings already inserted. Duh. The bearings were NOT replaced when the spring was put on professionally in May of 2011, so the lesson learned is always replace the bearings when you replace a spring.
Inspection of the spring anchor bracket showed a nylon bushing which was replaced with a steel bearing, but this was not a necessary step.
formatting link
One cable was found to be very slightly frayed as shown in this photo:
formatting link
both cables were replaced. The length of the new cable was about 5/8" longer and twice as thick as the original, which, much to my surprise, did not seem to cause any adverse effects when wound on the cable drums. Time will tell.
Replacing the cable was, at first, problematic, because I didn't know how to get the cable onto the bottom bracket. It took about ten minutes to figure out the easiest way was NOT to try to fit the loop over the post with the bracket in place - but to remove and flip the bracket upside down, and then loop the cable onto the post and replace the roller at the same time.
The deep dimples in the torsion bar prevented the spring and bearing from sliding from the right, and in the process of removing the torsion bar multiple times, I accidentally punched a hole in the garage wall.
formatting link
The bearing end plates each only had one lag screw on the supporting wall, so I added a second lag screw as my prior experience indicates problems can occur when bearing end plates move laterally (long story).
formatting link
A vise grip was handy for keeping the torsion bar in the spring anchor bracket while assembling parts repeatedly (preventing the 17-foot long torsion bar from falling down).
formatting link
The spring had to be bent in order to fit it onto the torsion rod with the limited side-wall space in this cramped garage:
formatting link
It's hard to see in situ, but there was originally some slop in the distance between the cable drums and the bearing in the end plate flags - but this photo of the bearing pressed against the cable drum shows how it is supposed to fit once on the torsion rod (there is not supposed to be any room between the bearing and the drum):
formatting link
Keying off a lesson learned in the past, a red line was painted on the spring AFTER it was wound 29 quarter turns, so that, over time (years?), any fatigue in the spring will show up as a slowly forming spiral (at least that's the plan):
formatting link
Having two ladders (three would be even better) made a huge difference in convenience, as was having good lighting all around (I bought and installed new fluorescent tubes just so that I could see better).
formatting link
A trick I used to ensure I only tightened the set bolts 1/2 to 3/4 turn after hand tightening was to paint one of the four flats with nail polish, so that the movement of that one flat was easily noted:
formatting link
Normally, the torsion bar is centered so that an equal amount is sticking out of each bearing end plate - but to move the spring winding cone set bolts away from the deep dimples in the torsion rod, I pushed the torsion rod to the right an extra inch or two:
formatting link
Well, that's about it for lessons learned in this, my second torsion spring style garage door repair. Hope this helps others NOT make the same mistakes I did.
In summary, since the spring was replaced less than two years ago, I fault the installers for NOT replacing the bearings at that time, which if they had, none of this would have happened.
However, I'm not without blame, as I really should have replaced the torsion rod, and, truth be told, I feel badly that I didn't convert the single spring to a two-spring setup - but that would have taken more time and money (about $80 extra or so).