A few days ago, I got a letter from the HOA telling me that I have to replace the warped, twisted, sagging 2x4s forming the roof of my pergola.

I have absolutely zero confidence that any PT 2x4 I buy from Home Depot or Lowes won't warp/twist/sag the exact same way within a matter of months, let alone years. Three weeks ago, I replaced the first one with a normal 16' PT 2x4 from Home Depot as a science experiment. I didn't have to wait long for the results.... it's not even a month old, and it's ALREADY bowed.

I did more research, and discovered LVL (and LSL and PSL) engineered studs. Except nobody seems to make a 2x4x16' (1.5"x3.5"x16') stud that's officially approved for outdoor use.

The boards won't ever have to support any weight besides their own, so life-safety and code-compliance aren't concerns. However, I've also seen what happens to OSB and particleboard when they get wet.

From what I've gathered, PSL swells when it gets wet, and never goes back to its original form... but Weyerhauser has a product (Parallam Plus PSL) that supposedly IS marketed for submerged use. Unfortunately, the smallest size you can get is 4x10 (3.5x9.25), and apparently you can't just take a 4x12 and slice it into multiple 4x2 pieces with a table saw.

I read on another site that LVL studs will swell and cup when wet, but return to their original straight form once they dry out. I got a quote on Boise-Cascade Versa-Stud LVL from Home Depot, but when I talked to Weyerhauser's support rep, he was ADAMANT that it could not be used outdoors (but wouldn't elaborate on whether it was because it would swell/warp/disintegrate, or just a matter of regulatory approval).

Has anybody seen what actually happens to a LVL (like Versa-Stud) when it gets used somewhere that's directly exposed to the elements (albeit painted)? Likewise, if I sliced a Parallam Plus into multiple faux-2x4s and used them instead, would their APPEARANCE be compromised, or would they just be useless for loadbearing applications (but perfectly capable of supporting THEIR OWN weight if they're just laid horizontally on top of a pergola)?

I actually did a second simultaneous experiment 3 weeks ago... I took two 1x2 6' strips and glued & screwed them together to make a 2"x2"x6' homemade glulam. It warped into a graceful arc that would be pretty if it were intentional, but unfortunately renders it unusable for its intended purpose (a straight 1.5" x 1.5" x 6' board, to replace the original PT 2x2s that all eventually warped and bowed at one or both ends)

I'm definitely open to other suggestions, if anybody can think of a better alternative to standard 2x4x16 PT studs. I looked at cellular PVC, but the ones I found were either a) not strong enough to support their own weight aross a 10-foot span without sagging, or b) so breathtakingly expensive, I didn't even get to the point of investigating their sag-resistance.

As far as cost goes, I can live with spending 2-3 times as much as I'd have spent on regular pressure-treated 16' 2x4s if I can feel confident that they'll look good for at least 8-10 years.

Update #2

Re the suggestion to use cedar. Home Depot can order it, and the nominal price per board isn't bad... until they hit you with a $79 surcharge for lumber special orders under $700, which completely nukes the economics of buying 10 boards, and risks total expense meltdown if I ended up having to place a second order to get a few more.

Other ideas I'm currently exploring (but not really satisfied with):

  • Attempting to use two short in-stock cedar 2x4s fastened end-to-end with a pair of mending plates. Not something I really want to do, because it wouldn't save much money and would probably look awful.

  • Using PT 2x4s, with added bracing... a 1-1/2" x 1/8" x 8-foot aluminum plate screwed into one side of the new boards that are laying flat to strengthen the 10' unsupported span.

Yeah, I'm at that dangerous, frustrated point where I'm roaming Home Depot looking for anything I can creatively repurpose as an exoskeleton to brace a 2x4... vinyl trim strips, steel stud tracks... ;-)

Update #1

Re the suggestion to reduce the span. In theory, I could possibly add a third pair of horizontal beams parallel to (and halfway between) the existing two and reduce the unsupported span from 10' down to about 4.5'. Two problems, though:

  1. The "house" side of the beams is fastened to the front wall of the house with what appears to be a custom-machined steel bracket. I know Simpsons' Strong Ties would probably be functionally equivalent, but then it wouldn't match the appearance of the original brackets on the two existing beams.

  2. The attachment point ON the front of the house is the front edge of a cast in place reinforced concrete suspended slab. Maybe I'm over-thinking the problem, but I'm pretty sure that the recessed area (where the existing ones are anchored) and the dimple along the underside isn't purely ornamental, and that whatever is behind the dark-brown recessed area is likely to be pretty hard to drill through.


Illustration of how the existing beams are anchored to the front of the house

enter image description here

The frame of the pergola after I removed the warped boards enter image description here

What it looked like before I removed them enter image description here

  • 1
    Having seen pergolas degrade and rot, that sounds like an ideal place to use white powder-coated aluminum tubing.
    – Tim B
    Commented Mar 6, 2017 at 21:01
  • RE: update #2 - home depot is not your only option for cedar boards. OR IS IT? find lumber yard that has them in stock.
    – Alaska Man
    Commented Mar 8, 2017 at 11:00

4 Answers 4


The problem isn't that there's anything wrong with PT 2x4s. The problem is that you're spanning 10 feet with, presumably, no support or bracing. A 2x6 is much more appropriate for what you're doing.

Another option would be to stick with 2x4s and run boards the other direction, maybe less often or maybe using smaller lumber, to create a grid structure. This would be stiffer and resist warpage better.

LVL lumber, like any laminated wood product, will eventually come apart if exposed to the cycles of weather. It's also not substantially stiffer than solid wood for your purposes.

(P.S. Those aren't "studs", which isn't a fancy word for all 2x4s. Studs are a vertical wall component.)

  • The problem with 2x4 vs 2x6 is that the original builder used 2x4s across a 10-foot span, and I'm not allowed to do anything that deviates significantly from what was originally there. The HOA isn't going to come by with a ruler and complain if the new boards end up being something like 1.25" x 3.75", but 2x6 would definitely look out of place compared to the neighbors & get their (negative) attention.
    – Bitbang3r
    Commented Mar 6, 2017 at 20:53
  • 2
    Any HOA that would object to (or disallow a variance for) a minor improvement to the original design that follows its style needs their bell rung. Fact, not opinion. I'd ring it softly by asking, at least.
    – isherwood
    Commented Mar 6, 2017 at 21:11
  • 4
    2x4 laid flat? The builder was a bleeping fool.
    – Ecnerwal
    Commented Mar 7, 2017 at 1:51
  • Just about any lumber will twist or bow if it is not properly braced. As Isherwood points out the unbraced span on this design is too long. Whatever lumber you choose, add more horizontal bracing to keep them straight.
    – ArchonOSX
    Commented Sep 13, 2017 at 9:13

I think one would use redwood or cedar for a pergola exposed to the weather.

  • Cypress is another naturally water resistant wood Commented Mar 7, 2017 at 0:34
  • Hmmm... I totally forgot about cedar. They don't sell 16-foot cedar 2x4s at Home Depot, but I'll go tomorrow and see whether they can special order them. IMHO, an ideal engineered product would probably be two or three 16-foot rectangular steel tubes filled with high-density styrofoam (for low-weight compressive strength), embedded in a block of low-density styrofoam that's slightly smaller than a 2x4 (for shape & bulk), then skinned with some kind of woodgrain-embossed plastic. But AFAIK, that's pure fantasy. ;-)
    – Bitbang3r
    Commented Mar 7, 2017 at 7:02
  • In Dallas cedar would be used. In California redwood. Cypress would be far more expensive, if you could get it. (BTW "cypress" in the US south means bald cypress en.wikipedia.org/wiki/Taxodium_distichum.) What is the center to center span between adjacent vertical posts? What is the total length of the pergola? In Dallas 2x6 cedar would be used for this application, laid on edge not flat. Surely yours were not laid flat? Where is this property located? Commented Mar 7, 2017 at 11:01
  • The HOA may require you to use 2x4s to keep the same appearance as the other houses; be sure to check with them before you buy any wood. Why get 16' ones? Can't you join them over the supports? Again, check with the HOA, don't assume you can make a change, even if it would be considered an upgrade. Commented Mar 7, 2017 at 13:22

I'd use vertical grain 2x4 cedar or redwood kiln dried down to 6-8% (interior standards).

The problems: 1) pressure treatment: I would NOT use pressure treated material. they are adding liquid to the member...which creates more drying out and more warping, and 2) larger boards (use 2x6's rather than 2x4's) which will twist more than smaller boards as more moisture will need to be removed, (incidentally, increasing the width might help, but increasing the depth is exactly the wrong thing to do), and 3) using lvl boards: while dimensionally stable, they are susceptible to delaminating if not "perfectly" maintained., and 4) grade: Using vertical grain material will reduce warping. (Go to lumberyard and hand-pick each piece out...don't rely on bulk delivery.) and 5) moisture content: the single biggest factor to keeping the members straight and true. Make sure the lumber is dried to 6-8% or so (unless you live in a high relative humidity area). Ideally, it would be great if you could "acclimate" the wood too. However, you run the risk of it twisting if not "held" correctly during the acclimation period. However, as the relative humidity changes, the board will gain/lose moisture and therefore grow/shrink/warp.

Staining or painting will slow the drying-out/warping process.

  • This is spot on, also the OP will likely have much better luck at a lumber yard than a big box store. Commented Mar 10, 2017 at 21:26

I personally think you are over-complicating the situation. You should think less about how to find an engineered product (Such as LVL and PVL) to solve the problem and instead look at why the wood itself bends and twists. Wood will in itself act in line with its grain and organic structural make up. The biggest factor in how a piece of wood 'settles' into its shape is if it contains the 'heart.' This is the center of the tree and any piece of wood that contains it will go nuts over time twisting like crazy. Study the grain fluctuation and order wood as 'FOHC' (Free of heart clear) with a larger waste factor and then vet the wood you use avoiding large changes in the grain and you will get a superior result in your product with a much lower price point than going to engineered wood products. I would also say to keep in mind that increasing the amount of wood used by pairing up joists doubles your weight which your structural base may not be engineered to handle and could potentially create a failure point. This means more structural work and everything that comes with it (To include the foundation/footings) which you do not want to mess with. If you want to go nerd nuts on the perfect solution calculate the weight distribution you are looking for and put wood under that load for 30 days. This will tell you how that wood will settle and use the pieces that fit your standards. Understand that the price of LvL and PvL is not the only issue. You have to get framers/electricians/plumbers that understand how those engineered products function and how to build with them. If you cut into 30% from either side of a LVL (Like any uneducated subcontractor would) you totally wreck its structural integrity and ruin any benefit the engineering of the product ever provided. Its not just the product that matters, its how its used. The price point difference to properly use these products you are talking about is 10-15 times material cost + the expertise to know how to work with them. Focus on pragmatic simple solutions and you will end up with a better end result with a lower cost.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.