Search the forums

Loading

Garage Door beam sagging

Castanea's picture

After 33 years and a new tile roof 6 years ago, the beam across my garage door is sagging about 1 1/2 inches in the middle. The door sets behind the beam and is not binding.


Can it be repaired by jacking it up to level or slightly above, and laminating two thicknesses of 1/2 inch hardwood plywood on both sides with construction adhesive and wood screws?


If that is feasable, where would the seams fall between the 8 foot lengths of plywood and what type of screws and screw pattern would be best. 


I appreciate your input and any other ideas 

(post #99045, reply #1 of 58)

To fix this on our garage I removed the double 2x14 header and installed double microlams. Shimmed up in the middle a hair so that it's dead level after the load is on it.

(post #99045, reply #7 of 58)

Dan, how long is the span on the garage opening?


I am in a similar situation with a single door two car garage and the sag is about an inch, hasn't changed over the years. Thinking of just shimming the head jamb to make it look presentable. Think it would work?

(post #99045, reply #8 of 58)

Guys, you're killing me.  It takes a heck of a lot more information than just the span to figure out if you're looking at similar situations.  Equally as important as the span, is the load above the opening.  One of you may have a garage door in a gable end carrying minimal weight, while the other may have a door in a basement garage that is carry two floors of living space as well as a roof load.


Both of you should make notes of as much information as you can muster describing the situation, along with a few pictures of your house and the opening in question and take them down to your local lumber yard.  They ought to be able to size an appropriate replacement header/beam for your openings or steer you in the direction of someone who can.


Either that or check it once a year like that clown replied earlier.  You gotta be kidding me.

(post #99045, reply #10 of 58)

DP, keep in mind that we're discussing what to do about an EXISTING header here. In general, the existing header has demonstrated "adequate" (but clearly marginal) ability to handle the load, and presumably the header was "legit" for the load when installed (if this is in a locale with real inspections).

So the question is what can be done to increase the load-bearing capacity of the beam to correct an essentially cosmetic defect. And, more to the point, what options are available/reasonable without a total reconstruction project, major engineering costs, etc.

Maybe the "engineer" at a lumber yard can size something for you, but what he comes up with may not be practical for the existing situation. If what he comes up with is "practical", it's about 99% certain to involve replacing the dimension lumber with the largest microlams that will fit the opening. Duh!!

Granted, the tile roof does add another "interesting" dimension, but mostly this is only if the original roof was not tile (and this point has not been established, one way or the other). If the original roof was not tile, I'd recommend that the OP get an engineer to look at the overall setup and make sure that the rest of the framing is sufficient before investing a lot in the door opening.

I agree that taking annual measurements is probably a waste of time. In my experience there will be no appreciable increase in sag after the first 3-4 years, and, besides, what good does it do you to know that the sag is getting worse -- it's bad enough already.

(post #99045, reply #11 of 58)

It appears to have been OK for 27 years, then sagged when the tile went on 6 years ago.  That sounds like it wasn't designed for tile.  Junkhound's Flitch conversion will probably work, but I'd run the actual numbers before trying it.  Another question -- Is this in earthquake country?  Anything marginal is the first to go when it gets shook.


 


 


-- J.S.


 

 

 

-- J.S.

 

(post #99045, reply #12 of 58)

Structural members in failure are now cosmetic issues?   I missed alot this weekend.


What ever you say.  It's your house.  What do I know?

(post #99045, reply #13 of 58)

Ha ha,  Don't worry about that beam it's only sagging 1 or 2 inches it's mostly cosmetic anyway.


What happened to "Castanea"?


If that house didn't have a tile roof prior to the time it was replaced I would be concerned about the other parts of the house too.  Maybe that's why Castanea hasn't posted anymore he's busy shoring up the place.


 


 

(post #99045, reply #21 of 58)

The thinking seems to be, "since it hasn't fallen down yet, it must not be in a state of failure" so this one here is structurally sound I guess.

 

 


Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!

 

 

Oh Well,

We did the best we could...

PreviewAttachmentSize
tumbledown.jpg
tumbledown.jpg74.67 KB

(post #99045, reply #26 of 58)

Help the ones that want to be helped, I guess.


Thank goodness for thermo-formed Azek trim boards.... don't know how else you case out a garage door opening with a 2" sag in it.  Maybe you shim that too?


I'll know when I'm in Dan's neighborhood..... when I'm driving around and start to wonder why I get the feeling that all of the garages are "smiling" at me.


Whatever.... glad I ain't parking my rig in any of 'em.


You were missed this weekend.  Hope you and yours are well.

(post #99045, reply #14 of 58)

1 1/2" over 16', the longest span the poster could be talking about, would be equivalent to 1/128 of the span.  Most designs are done to 1/360 maximum deflection.  Roofs can be done to 1/240 but that is bouncy.  Interiors with long spans are often designed for 1/480.  Those are all maximum live load plus dead load numbers.  The original poster is talking about JUST a dead load deflection of L/128 or more.  Definately more than cosmetic. 

 

 

(post #99045, reply #15 of 58)

Yet, I see this much deflection in a large number of the garages around here. Have yet to see one fail.

(post #99045, reply #29 of 58)

"Most designs are done to 1/360 maximum deflection."

That's a common misunderstanding about roofs and deflection ratios.

Remember that there's LIVE load deflection and TOTAL (live + dead) load deflection.

Floors are typically done at L/360 live, and L/240 total. (Although a lot more people are using L/480 for live lately)

Roofs are trypically done for L/240 live, and L/180 total.

These are the minimums called for under BOCA. Other state/local codes may require different things. But those numbers are what's commonly used in the midwest.

Every father should remember that one day his son will follow his example instead of his advice

(post #99045, reply #43 of 58)

I stand corrected, thanks.

 

 

(post #99045, reply #30 of 58)

"1 1/2" over 16', the longest span the poster could be talking about"

Why do you say that?

18ft is a common size door and you can get larger.

. William the Geezer, the sequel to Billy the Kid - Shoe

(post #99045, reply #44 of 58)

"1 1/2" over 16', the longest span the poster could be talking about"


Why do you say that?


18ft is a common size door and you can get larger.


 


I should have said 16' is the longest span I've seen or built.  Most doors around here are 8', 9', or 10'.  Never installed an 18 footer. 


 


Mike

 

 

(post #99045, reply #46 of 58)

Agreed, adding the plate on the bottom of the beam would prevent stretching - but - to be realistic. doing the math, the difference between the flat 16 foot span and the 16 foot span with a 1 1/2 inch deflection is .012 inch, less than 1/32, assuming an 8 foot radius. That would require some pretty solid and precisely fitted lag bolts.


Yes, I have a 1958 Porsche Convertible D in that sagging garage which I would like to sell and give me room for serious woodworking.

(post #99045, reply #48 of 58)

You are right, according to Junkhound; those lag bolts must be precisely placed and they must be in holes with no slop. He mentioned that in his initial post.


Art (Junkhound) is an engineer, something I are not. I understand the reasons for structural behaviour; he can calculate it.



Dinosaur


A day may come when the courage of men fails,when we forsake our friends and break all bonds of fellowship...



But it is not this day.

Dinosaur

How now, Mighty Sauron, that thou art not brought
low by this? For thine evil pales before that which
foolish men call Justice....

(post #99045, reply #20 of 58)

Any structural member in a residence that has an inch and a half deflection is definitely NOT "adequate" as you say. It is in a state of progressive failure and needs to be remedied. This is far more thana cosmetic issue.

One of the good qualities of wood is that it does not fail catastrophicly like steel, which bears the load until one day when the provervial straw is added to the load, it collapses suddenly.
Wood, on the other hand will slowly deflect beyond the design parameters, and show other signs of distress long before it finally collapses.

This header is showing those signs, one of which is that it has exceeded acceptable deflection, a demonstration that it is incapable of resisting the loads being placed on it. Ignoiring the condition and calling it cosmetic is like driving on a nearly flat tire all month with the assumption that the scruchy sidewalls are only a cosmetic issue. It is like going aaround all month with a bloody noise that won't quit and thinking that it is only a cosmetic concern.

It is a visible sign that something is wrong.

 

 


Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!

 

 

Oh Well,

We did the best we could...

(post #99045, reply #23 of 58)

Like I said, I could point to a dozen homes around here (all over 20 years old) that have about that much deflection above the garage door.

(post #99045, reply #25 of 58)

Putting shims in the middle between the header and the top plate will shim the plate up or the header down. It will not shim the middle of the header up

so since every garage on the neighborhood was poorly built, and has headers in failure because some developer/builder lined his pockets by doing shoddy work, you think it is OK to do?

Yes? No?

I recommend doing it right instead of repeating a problem. Think about the logic of what you are saying

 

 


Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!

 

 

Oh Well,

We did the best we could...

(post #99045, reply #19 of 58)

sanity has arrived

 

 


Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!

 

 

Oh Well,

We did the best we could...

(post #99045, reply #9 of 58)

Mine was a 16-foot opening with the truss ends of a standard asphalt roof resting on the beam. Original beam was doubled 2x14 (13 1/4" actual, IIRC), and was sagging about 1 1/2". I replaced with double microlam that was 1 3/4 x something like 12 5/8 -- the closest I could get. When tentatively let down it sagged about 3/8", so I shimmed that much in the middle to get it perfectly straight.

I was redoing the siding at the time, so it was a lot simpler to do the work from the outside. Doing from the inside would require removing the door in most cases.

Whether you can shim or not depends on the structure. In my case the bottom of the beam was pretty much touching the door frame, so there was no way to simply shim and still have the frame straight.

(post #99045, reply #17 of 58)

I'm trying to figure out how you can shim up the middle of a clear spanning beam...???

 

 


Welcome to the
Taunton University of
Knowledge FHB Campus at Breaktime.
 where ...
Excellence is its own reward!

 

 

Oh Well,

We did the best we could...

(post #99045, reply #22 of 58)

> I'm trying to figure out how you can shim up the middle of a clear
> spanning beam...???

You put the shims between the beam and the top plate.

(post #99045, reply #2 of 58)

THe plywood as mentioned is next to worthless.


Assume the span is now a 4x12 or up to a 6 x14??  If so get a 15 ft long piece of 1/8 thick steel, 5 or 6" wide, lag bolt it to the bottom face* of the existing beam while the cneter is jacke up to about 1/4" above level (you may need to leave the jack in place for a few days to get the sag out without cracking any drywall)


*  You need a lot of 3/8 inch 3 inch long lags, double row at the end at about 1" spacing, single row at about  12 ' spacing near center span, shear loads get higher toward the ends, so you can graduate the bolt density up as you get near the ends. TIGHT bolt holes _NO slop allowed. Takes about a half hour to do the shear calculations by hand to get the optimum number and spacing of bolts, a linear progression of the spacings above is a good rule of thumb for a 16 ft beam. 


Edited 8/14/2005 11:22 pm ET by junkhound

(post #99045, reply #28 of 58)

My apologies for taking so long to get back, went to Laguna Seca for a historical car race but thats way off the subject.


Your idea of the plate on the bottom sort of suprised me. Not being a builder I would think the steel plate would be lag bolted to the vertical surface.


Accepting your idea, where can I get info on the lag bolt spacing?


Answering some of the questions earlier.  The tile roof (heavy) replaced a wood shake (light) roof after an engineering report and a tough permit process.  Yes, I am in Southern California which is EQ country. The beam is 17 feet overall. There is just the roof above the opening.


I would rather not get into the beam replacement process which would require shoring up the roof and removing the door.


Further comments?


Thanks

(post #99045, reply #31 of 58)

[Junkhound's] idea of the plate on the bottom sort of suprised me. Not being a builder I would think the steel plate would be lag bolted to the vertical surface.


The function of the steel plate in this configuration is to prevent the bottom of the beam from stretching. It's a combination of fibre stretch on the outside of the arc and compression on the inside that allows a piece of wood to bend.

Based on what I've read in this thread, you may or may not be able to solve your problem with a Flitch plate. You're not going to get a definitive answer here, unfortunately, because we can't see and measure the garage, and not all of us are engineers. While I would venture to say there are numerous posters here who could fix that problem on site without breaking a sweat, the reality of your situation is that eventually, you are going to need to have an engineer (not an architect) examine the situation and design a remedy. This is especially true since you are in earthquake country. You may very well wind up having to replace the entire front wall of the garage with a unitized shear panel. But I can't tell from here....


Your interest in historic cars indicates to me you are probably parking some fairly valuable hardware in that garage. So you really oughta bite the bullet and get on with it, seems to me....



Dinosaur


A day may come when the courage of men fails,when we forsake our friends and break all bonds of fellowship...



But it is not this day.

Dinosaur

How now, Mighty Sauron, that thou art not brought
low by this? For thine evil pales before that which
foolish men call Justice....

(post #99045, reply #33 of 58)

It would seem to me that there is a simpler non-engineering approach:


Can't he just flip the beam upsidedown, so that the sag becomes the crown?  This would not change the structures architecture, nor it's engineering (other than give a sudden 1.5" boost to the front roof height).  A Flitch Plate could be installed on the underside after the beam has settled back into a more desirable shape.


The big beam (4x14 in my garage) sagging seemes to be caused by load and time... but doesn't wood deforme over time when exposed to heat, humidity, and load?  I thought that is how you shape wood into curves on purpose!  In this case, it happenes slowly over time.


Both of the big beams in my garage (6x16 ridge and 4x14 garage door header) had some sag - the garage door had about an inch.  My roof coverings (since 1960) were asphalt gravel (original), asphalt shingle (80's), and asphalt shingle again (90's).


BTW, I'll bet your garage door will come off more easily than you think.


Rebuilding my home in Cypress, CA


Also a CRX fanatic!

YAY!  I love WYSISYG editing!  And Spellcheck!

____________________________________________________

(post #99045, reply #35 of 58)

How do you hold the roof up during the flipping process?  What about the exterior stucco or whatever is there over the beam?  This sounds to me like a lot more work than Art's in-situ Flitchification idea. 

 


 


-- J.S.


 

 

 

-- J.S.

 

(post #99045, reply #36 of 58)

"flitchification"   That's awesome.