Lessons learned installing a torsion spring in a typical residential garage

[Danny D.]

For most of you on a.h.r, these lessons will be old hat; but for those of you (like I was) who have never relocated a torsion spring, these lessons may come in handy in the future.

WHAT DO MOST GARAGE DOOR COMPANIES DO:

- Most charge around $150 to $200 and will do a great job.

- Most promise same day and next-day service.

- I could do the entire job in an hour - so I'm sure they can too.

- Most will simply replace the old spring with the same size new spring.

- Some will 'upgrade' the spring to a longer-life spring (others won't).

- Some may charge for that longer-life upgrade; others won't.

- Most will adjust and lubricate the door and GDO as a bonus to you.

- Some may try to sell you useless extras, such as galvanized springs or warranties (the 800 numbers I called were the worst offenders).

MAINTENANCE:

- Snap a picture today of your torsion spring setup (I wish I had this).

- Lubricate the garage door rollers, hinges, GDO mechanism, and torsion bar bearings (two to three of those bearings may be installed).

- Check that the door is parallel when raised and vertical when lowered.

- Operate the door by hand to check spring balance at multiple positions.

- Close the door and check for tilt by looking for light at the bottom.

RESEARCH:

- I've watched EVERY torsion spring DIY on YouTube and none beat DDM Garage Doors - so all you need is the ddmgaragedoors.com web site.

- The Richard Kinch truetex web site is the second site you'll need.

- No other web sites are needed although I've read EVERY alt.home.repair thread that mentions garage doors that I can find in the groups.google.com archive and while there is 'some' really good information on a.h.r - most of the threads also contain contradictory garbage, and therefore you must take every thread with a grain of salt.

REPLACEMENT:

- Replacing a torsion spring is easy and requires basic tools.

- The only special tool are two 18" long 1/2" diameter winding bars.

- A few open end wrenches and a large vise grip is all else you'll need.

- Basically, to remove a single broken torsion spring you climb on your stepladder, unbolt the two set bolts on the winding cone and remove the two nuts on the spring end plate side. After marking the location of the cable drums on the torsion rod, you loosen the set bolts on both cable drums, and then you simply slide the broken torsion spring off the torsion rod, leaving the torsion rod at the top of the door and only removing the one cable drum on the side away from the spring anchor plate.

- Basically, to replace the torsion spring, you side the spring onto the torsion rod, add a bearing if desired, line up the cable drum prior marks and tighten the cable drums snug against the bearing end plates and insert the cables holding them in place with a vice grip tensioning the tension bar and then proceed to wind the spring. When wound the prescribed 30 quarter turns (7 1/2 full turns for a 7 foot tall door), you push out the spring about a quarter inch, and then tighten the winding cone set bolts. Then you check and adjust and lubricate the hinges, rollers, bearings, and GDO mechanism.

- If it's a two-spring system, the only additional initial step is to unwind the unbroken spring before touching anything. Unwinding the old spring is even easier than winding the new spring and is simply the reverse operation of 30 quarter turns (7 1/2 full turns) for a 7 foot tall door.

DANGER:

- I searched the news.google.com archives for gory stories of residential homeowners being hurt or killed by winding garage door springs at home, and, I just couldn't find much. This doesn't indicate much other than it's not a big newsworthy topic I guess - but it's a datapoint.

- EVERY (and I mean every) site says it's dangerous - just as chain saws and table saws and 220 volt motors and swimming pools are dangerous when accidents happen - so I'll repeat what the sites say. It is dangerous.

- Winding torsion springs is dangerous because "something can go wrong", and because "something can break".

- However, having said that, if you take normal precautions against both of those possibilities, you too can (easily) wind a torsion spring.

- There are plenty of things NOT to do, by the way, when winding torsion springs ... but the list of things to do are well spelled out at the DDM Garage Door web site.

- Personally, at no point did I "feel" dire danger, especially after having removed and reinstalled my torsion spring a half-dozen times. It became 'almost' routine (therein lies the biggest danger, I suspect, to garage door repairmen).

- The amount of force needed to wind a 0.250" 36" long torsion spring with 18" steel bars is well within the strength of a normal man.

THEORY:

- The torsion spring acts like a counterweight to balance the (appreciable) weight of the door.

- The GDO merely pushes the door open or closed - and in and of itself does NOT open the door.

- The only lateral movement, assuming the cable drums are tightened against the bearing end plates, as the door goes up and down is merely the distance between the coils of the springs.

- Everything else should be locked down tightly (which was my problem).

- In general, the garage door repair company skimps on the springs, by default, by giving you a 10,000 cycle spring.

- You can ask for longer cycle springs, which, if you keep to the same inside diameter, are merely thicker gauge wire.

- The only thing that matters is the IPPT (inch pounds per turn) that you need. Your garage door has a weight and a track & drum geometry that determines the IPPT you need. Period.

- So, whatever new springs you put on must exert the same IPPT as the old springs. Period.

- Most people want longer-lasting springs, so, all you need is either a wider inside diameter spring with the same IPPT or a thicker gauge spring with the same IPPT.

- In general, you likely won't opt for a wider ID so your choice is merely a thicker gauge spring.

- Adding a second spring does NOT in any way improve your number of cycles (only the spring geometry makes a difference in lifetime).

- Adding a second spring has some benefits - but they're relatively minor.

- Your limiting factor in improving lifetime will be how much room you have as thicker-gauge springs with the same ID and IPPT will be longer than the original.

[DD_BobK]

DD-

Just because you've done one garage door (or seen one done) & read a number of website sites does not make you expert enough to be writing up descriptions for other novices to follow.

Your use of the subject terminology gives you away as someone merely regurgitating / "parroting" (apologies to every parrot worldwide) poorly understood information.

Your fix (metal "plate") was an amateurish hack.

Stop fixating on "ugly", learn to do appropriate investigative demolition, drywall repair and how to take expert advice offered by the a.h.r regulars.

How's your slope doin' ? Goin' read some websites & post on slope stabilization next?

[Danny D.]

I never said (nor implied) I was an expert so I'm sadly confused about your response.

In fact, I clearly & overtly said in the very first line that I had never done it before; and in the text I said these were lessons learned from removing and reinstalling the torsion spring a half dozen times.

So, I'm sorry if you were confused (apparently) because I also said I read every DIY and howto and watched every video that I could find on the net.

I meant you no harm - and I sincerely hope nobody else (erroneously) thought I was an expert as I never represented myself as such.

[I should note I confirmed all my work with Dan Musick whom I've spoken with only a half dozen times, yet whom I 'would' consider an expert.]

[Danny D.]

Agreed!

That reduced-severity action actually works both ways.

1. When the door is going down (as you noted). And,
2. When the door is coming up: When a single broken spring, especially when it breaks near the stationary cone, can cause the torsion bar to spin violently such that there can be damage the top part of the door and to the cables.

In addition, two-spring systems balance the forces on the two cable drums, which is important for systems (like mine) with weak cable drum support to start with.

Two springs also enable the advantage that MORE SPRING SIZES (i.e., thicknesses) are instantly available to the homeowner, who then has more options to choose springs of the desired life cycles.

A minor advantage of two-spring systems is that winding each one is half the work of winding a single-spring system; but that's a one-time bonus.

Of course, two spring systems typically being longer (combined) than single-spring systems have disadvantages too (e.g., they take up more room so that you may have to move the spring anchor plate, they can cost more, there are more things to break and replace, they can be heavier, etc.).

All in all, very few professional installers told me by phone that they would change out my single-spring system for a two-spring system, and, Dan Musick advised me against it (even though his web site explains all that I've said above).

Nonetheless, if you wish to convert from one spring to two, this handy calculator tells you all that you need to know!

For example, here is my current spring: $51.56 36,000 cycles 0.250"x36" 14.14 pounds Lift=123.3# SPB-250-36-00R

If I converted that to two springs, Dan's site recommends: $34.56 77,000 cycles 0.207"x28.5" 9.09 pounds Lift=61.9# SPB-207-28-50R $34.56 77,000 cycles 0.207"x28.5" 9.09 pounds Lift=61.9# SPB-207-28-50L

So, the two-spring system would only cost about $15 more and would weigh only about 4 pounds more, yet it would lift the same 123 pound door and each spring would last more than twice as long as my single spring system.

[jloomis]

I never said (nor implied) I was an expert so I'm sadly confused about your response.

In fact, I clearly & overtly said in the very first line that I had never done it before; and in the text I said these were lessons learned from removing and reinstalling the torsion spring a half dozen times.

So, I'm sorry if you were confused (apparently) because I also said I read every DIY and howto and watched every video that I could find on the net.

I meant you no harm - and I sincerely hope nobody else (erroneously) thought I was an expert as I never represented myself as such.

[I should note I confirmed all my work with Dan Musick whom I've spoken with only a half dozen times, yet whom I 'would' consider an expert.]

[Danny D.]

Thanks for understanding.

I need help when I ask, and then, when I'm done, I've been on the net long enough to know to give back to the group (hopefully in spades), by paying it forward.

It was interesting though to read constantly about how dangerous it was, yet to not find in the literature very many documented examples of how exactly these people get hurt.

To be sure, I'm positive people get hurt every day (but most are likely commercial accidents due to common mistakes the professionals make when they're not careful doing something they do 10 times a day, every day).

For example, Dan Musick himself told me that he hurt his leg simply by stepping off the ladder onto the old spring on the floor. I'm sure he put old springs on the floor hundreds of times - but if you do that day in and day out, one of those days you're gonna trip on that spring and break your leg.

In addition, I'm sure that homeowners do some really really really dumb things, e.g., Dan Musick says on his web site that one of his customers unbolted the spring anchor plate without first untensioning the torsion spring! That customer was lucky to get out of that one alive!

And not everyone survives their dumb mistakes.

For example, I read this 2004 OSHA Fatality Assessment of a NY maintenance man who got killed in maintaining a commercial torsion spring.

However, if you read that report closely, you'll see MANY compound mistakes piled up one upon another - with the result being his eventual death.

I even searched the bestgore web site expecting to find garage door accidents galore - but alas - it was to no avail.

If you look at garage door accident statistics, there are 20K injuries in American garages every year - but most of them appear to be to the consumer and not to the repairman working on the garage door.

the statistics are a PR stunt for a garage-door company.)

This garage door company mimics those dire statistics:

Another door company publishes vastly different "statistics": they say 10K people are hurt every year by having their fingers pinched off in the door panels or having the whole door fall on them.

Perhaps more reliably, this scientific study of shows 85 children killed or seriously injured since 1974 when garage door openers didn't reverse on time (notice the numerical difference with the manufacturers' statistics):

And, perhaps most apropos, this study shows the percentage of DIY accidents: DIY accidents were 1610 out of 13,325.

Personally, I suspect all (or almost all) the DIY accidents were from people doing dumb things like using screwdrivers to wind the springs, or using flimsy ladders, or unbolting the anchor bracket without detorsioning the spring, or failing to disconnect the GDO before working on the door and then someone pressed the button - or - even this - forgetting to gently move the black widow spider from the web-strewn upper door area!

YES! That's 'my' black widow spider. I found it while I was setting up a safe and clean environment to work safely in on 'my' garage door repair!

[DD_BobK]

DADD-

I'm sorry but you're too inexperienced to even identify & recognize an expert.

Watching videos, listening to "experts" & reading websites .... none of this makes you an expert or able to identify an expert

The fact that you butchered an oak tree to satisfy your desire for "a view" speaks volumes about your personality. Your BS behavior since then confirms it. The fact that you're "confused" is only more evidence.

Repeating information w/o understanding it makes you no better than a parrot.

[Danny D.]

I corrected that assessment in a subsequent detailed response to rightfully recyclebinned who also disagreed as you did.

The advantages of two torsion springs are (summarized):

1. When one spring breaks, the door goes down softer.
2. When one spring breaks, the door goes up softer too!
3. Balanced spring forces are gentler on the cable drum flags.
4. With each spring doing half the work, you have greater spring choices.
5. Winding two springs is only half the work per spring.

The disadvantages weren't listed, but some of them are:

1. Two springs generally costs (slightly) more than one spring.
2. Two springs are generally (slightly) longer overall than one spring.
3. Two springs are generally (slightly) heavier than one spring.
4. Two springs are (slightly) more effort to install than one spring.
5. Converting from one spring to two spring has to be done correctly.

Let me know if I missed any pros and cons.

[Vic Smith]

Probably better to say when a one spring door spring breaks, the door comes down with full weight of the door, or if already down, it takes major muscle power to get the door open. The danger and effort is cut in half with a 2 spring.

Almost every piece of fastening hardware undergoes more strain with a single spring - because of torsion shaft lateral movement. Remember your spring anchor bracket flex. I think the only force that would be equal with either setup would be the rotational force pushing against the top of the spring anchor bracket and attempting to pull the lag screws in the bottom of the bracket from the header. I have to look at mine and see if they used 2 spring anchor brackets or 1. And whether the bolts connect the stationary cones together. Walked the dogs. It's one bracket, and the bolts go through both cones. Two brackets back to back would be stronger, but maybe overkill. And you'd need left and right bracket designs since there's 2 holes for 2 lag screws on the bottom and one slot for a screw on top. BTW, I think I said - and know I was thinking - that with a 2-spring setup the springs counteract each other. That's wrong, except for in the case of shaft lateral movement. They apply or release torsion to the torsion shaft in whatever direction the shaft rotates. Has to be that way (-: Still don't understand how the springs can lengthen/shorten on a

2-spring when all 4 ends are locked down and the shaft can't move laterally. I can live with that.

I don't see 2 and 3 as "disadvantages." Springs and hardware are contained within the length of the torsion shaft and nothing else should be around that anyway. Weight doesn't matter unless you drive your garage.

[Danny D.]

I totally agree that your two points above are 100% correct.

When 1-of-2 torsion springs break:

1. The door going down is (much) safer!
2. Lifting the door up is (much) easier!

But, there's another not-so-obvious advantage of having two springs on a double-car door when spring breaks while the door is moving up.

1. When one spring breaks, the door goes up softer too!

It's difficult to explain, so allow me simply quote Dan Musick himself: How to Convert from One Garage Door Spring to Two

[ verbatim ] Other problems frequently ensue when a single spring is used on a double-car garage door. Many manufacturers have cut costs by using a single spring on a double-wide 16' steel garage door. If the spring breaks near the stationary cone, a large portion of the spring spins loose with the winding cone secured to the shaft. This causes a strong force to pull on the cables, often leading to one or two broken cables. If the opener up force is set too strong, it is more likely to wreck the top section if not the whole door. A second spring helps to keep the tension when the first spring breaks, resulting in fewer broken cables and less damage to the garage door itself. [ / verbatim ]

[Danny D.]

I forgot to mention that I also learned that the torsion bar is NOT supposed to move sideways when the door opens and closes (allowing for the tiny bit of unavoidable slop at each side's cable-drum-to-end-bearing interface).

What Dan Musick told me by phone was the only sideways 'movement' is in the spacing between the coils. This wasn't intuitive to me because I personally twisted that spring 7 times and watched it grow almost two inches in the process (i.e., 7 quarter-inch coils).

I kept wondering where that two inches went when the spring untwisted on the way down! It turns out that two inches is hidden between the coils!

It's intuitive once you know it - but not until you do it!

[Danny D.]

Since this is a lessons-learned thread, another not-so-obvious result is to understand the differences when a torsion spring breaks nearer to one end than to the other in a single torsion spring system.

Until I read Dan Musick's web site, I hadn't realized that the part that spins violently is the winding cone end because that's the end bolted to the torsion rod itself.

The other end just spins inside its bearing.

So, the implication is different if the spring breaks nearer to one end than to the other. If the single torsion spring breaks nearer to the winding cone, the violent spin on the torsion rod is apparently much LESS than if that same spring breaks nearer to the stationary cone.

This, of course, has implications if the door was moving upward at the same time. This particular detail, to my knowledge, has never been discussed on a.h.r, and therefore I add it as an additional lesson learned.

[Danny D.]

Thanks. This is a good 'lesson learned' from having done it yourself!

It's common to have little space at the sides so many DIYs show how to bend the spring in an "L" shape to get it back on the torsion rod.

Also, some spring anchor plates (like mine) have a "U" shaped opening, which allows more freedom of movement of the torsion rod outward.

In addition, once both cable drums are loosened, you can slide the torsion bar laterally as long as you need to get the spring off (if you have room on the 'other' side of the garage).

In my case, I simply removed on cable drum and took down the entire torsion rod.

But, there's just no way you're gonna get the torsion spring off that torsion bar, in any case, without removing at least one cable drum. :)

[Danny D.]

I agree. That's the advice from every single web site & installer, although almost all will also say that if the second spring is less than a few years old, then they'd leave it be.

The main reason to replace both if the first spring broke at the end of it's useful life is that the second one is gonna break soon so you may as well save the additional service call.

Of course, if you do it yourself, you 'might' save some money just replacing springs as they break - but - personally - I'd use a broken spring as a golden opportunity to upgrade both springs anyway.

Hmmmmmm.... I hope you first unwind both old springs BEFORE removing the assembly from the door; and I hope you replace the assembly before winding the new springs.

If you don't ... very very very very bad things are going to happen.

Once you unwind the springs, removing the assembly doesn't require the neighbors' kids. Since I was relocating my anchor plate, I had to remove mine and it was trivial to take down from the doorway. Mine was only one torsion spring, but two springs would only have been four pounds heavier, as we've already ascertained separately in this thread).

Springs are only roughly 10 pounds each. The torsion bar is simply a 9' hollow steel 1" tube. The cable drums are cast aluminum.

The whole thing doesn't weigh much at all - so I'm not sure why the need for the extra kids. Nothing wrong with having help (I wish my kids would help me sometimes), but an average guy can easily handle the removal himself.

[Danny D.]

Thanks for noticing. You were the first to respond to my query, and you were the one who noticed that I had placed the rollers in the wrong slots for my other door. You also advised me when that second door was sticking in the tracks (ironically, that was resolved by moving the rollers!).

I took (almost) all your advice, e.g., to make sure the winding bars were seated well. And you explained why nobody works on the garage door when it's open, and your humorous color-code mnemonic was a good one, and you helped me understand those fancy spring-winding tools weren't needed.

I tried your relag suggestion; but it failed due to the lack of substantial wood below the anchor plate (which you and I were both unaware of).

You even were the first one (IIRC) that pointed me to the DDM Doors web site - which - after having read every single page - I would say is clearly the best on the net!

About the only suggestion I didn't take of yours was when you explained how you manage to unwind your torsion springs all at once! (Whew!)

I try to be responsive and to post pictures, and to pay it forward. While not everyone appreciated the effort, many have helped me on this endeavor and for that help I am indeed very grateful.

You and they make the USENET a fantastic resource for all of us.

BTW, even though Dan Musick's videos are all anyone would need, this 2-part real-time series from a young homeowner is pretty good:

1. Another decent video that hasn't yet been mentioned is this one which shows in a humorous way how to lubricate your garage door: One of the best benefits of homeowner DIY is that you know the right spring combination was put on the door. This short video from the same company above shows how to do a balance test, and it even shows what happens to the opener gear when a repairman puts the wrong spring on (because that's what he had on the truck). BTW, about the only GD company that says it's safe to do is this one: (They equate the task with changing a tire & they sell you the guides.) But there is always more to learn.

For example, an enigma I haven't found the answer to on the net yet is why the OLD broken spring still shows latent winding spirals ... as if it's still wound up?

That paint on the old broken spring used to be a straight line; so why does the old spring still show 7 spirals as if it's still wound?

Also notice the winding spirals are in the opposite direction from the original. Why would that be?

[G. Morgan]

Lol.. Yeah, those are only for springs that will never be used again.

[G. Morgan]

Because when the spring is compressed, it's not exposed to the elements. When you wind the spring it shortens in length by a few inches, then you tighten it. When unwound, that extra length is visible.

[Danny D.]

I don't understand.

I guess you're saying that the old spring was exposed to the elements while it was in the wound state - and yet - the backside of the spring isn't exposed to the elements so the backside is noticeably less rusted.

So, I guess, the backside gets a straight line of less rust while the spring is still wound seven times - and when that spring breaks, that straight line turns into an opposite spiral of less-rusted spring steel as the now-broken spring instantly spins back the 7 times to unwind?

[Danny D.]

I wrote to Richard Kinch and to Dan Musick at DDM Garage doors, to ask them about this weathered line in my broken torsion spring:

Both engineers kindly responded with great information that explains this anomaly and, interestingly, which provides a useful homeowner 'trick' to know when your springs might be slipping over time!

First, Dan at DDM Garage Doors explained that many springs have this reverse wound spiral because dust collects at the TOP of the spring.

As G. Morgan astutely surmised, this linear collection of dust on a wound spring, over time, causes differential weathering, which shows up as a straight line on a wound spring.

When that wound spring breaks, the weathering shows up as a reverse spiral on the unwound spring.

Interestingly, Richard Kinch provided ways to make use of this feature:

1. We can PAINT a line on a wound spring to judge slippage over time
2. We can LOOK at an unwound spring and count the number of turns
3. We can LOOK at an unwound spring and determine the chirality

Of course, there are OTHER ways to note slippage, turns, and chirality, but this feature can be used as a doublecheck!

Indeed, there are still lessons to be learned even AFTER installing a typical garage door torsion spring!

[jloomis]

I wrote to Richard Kinch and to Dan Musick at DDM Garage doors, to ask them about this weathered line in my broken torsion spring:

Both engineers kindly responded with great information that explains this anomaly and, interestingly, which provides a useful homeowner 'trick' to know when your springs might be slipping over time!

First, Dan at DDM Garage Doors explained that many springs have this reverse wound spiral because dust collects at the TOP of the spring.

As G. Morgan astutely surmised, this linear collection of dust on a wound spring, over time, causes differential weathering, which shows up as a straight line on a wound spring.

When that wound spring breaks, the weathering shows up as a reverse spiral on the unwound spring.

Interestingly, Richard Kinch provided ways to make use of this feature:

1. We can PAINT a line on a wound spring to judge slippage over time
2. We can LOOK at an unwound spring and count the number of turns
3. We can LOOK at an unwound spring and determine the chirality

Of course, there are OTHER ways to note slippage, turns, and chirality, but this feature can be used as a doublecheck!

Indeed, there are still lessons to be learned even AFTER installing a typical garage door torsion spring!