laminated veneer lumber (lvl)

[massmanute]

I am wondering about laminated veneer lumber (LVL) for structural members.

Depending on the grade of LVL chosen, the modulus of elasticity for LVL products can be as high as 2E6, which exceeds the modulus of elasticity for Sitka spruce or Douglas fir.

However, density of LVL grades with modulus of elasticity of 2E6 is typically higher (e.g. about 40 lb per cubic foot vs. about 27 for Sitka spruce or 32 for Douglas fir) which is a negative factor. It's slightly lighter than white ash on average, with a higher modulus of elasticity than white ash, and I understand that white ash is considered a decent aircraft wood, though perhaps a bit on the heavy side. A grade of LVL with modulus of elasticity of 1.7E6 is lighter (about 36 lb per cubic feet).

I don't know how LVL compares in modulus of rupture, shock resistance, and other structural parameters.

Vendors of LVL claim less variability in structural parameters than natural wood, which is a good thing. Vendors claim that the veneers are graded (by various automated means) before the laminated products are built up, which is probably also a good thing, and because a slight weakness or defect in any single lamination has relatively little effect on the overall structure, that is also a good thing.

LVL tends to be expensive compared to regular lumber, but I will bet it is not expensive compared to aircraft grade lumber.

According to some historical notes, LVL was actually used in World War II to build aircraft, especially the Mosquito bomber, although it is likely that the LVL used then is not the same as LVL used in the building trades today.

There would be a number of things to consider of course, (more than can be discussed in an introductory post on the topic), but I thought it might be a good topic to discuss, particularly since I can't find anything on the topic in searches of this forum.

[Dana]

I believe Steen Skybolts use laminated spars nowadays.

[FlyingRon]

Yep, the first thing I think (based on construction material) is LVL is heavier. Not as big a concern if you're building a house, but problematic on airplanes. You can get all the LVL parameters from their manufacters. ModE only one parameter (and even ModE has different modes of measurement). There's also flex pressures (the actually load carrying ability) and compression factors. I've had to go through all this when trying to size a beam over a 16 sliding door with limited room for a header above. I ended up having to add a steel filch to the thing.

Of course, the other building idea, trusses, have been used in airplanes for far longer.

The Skybolt does use a laminated spar, but it's a far cry from LVL. It's just a small number of plies (2?) of spruce glued together. It just primarily gets around the issues of finding a hunk of spruce big enough that meets the quality requirements.

[massmanute]

Yep, the first thing I think (based on construction material) ...You can get all the LVL parameters from their manufacters...

Thanks for the comments. I have not been successful in finding all of the LVL parameters from manufacturers online sources. For example, I have not yet found an online manufacturers information source that lists the modulus of rupture (MOR) of LVL their products. This surprises me a little because it is often claimed that LVL is stronger than natural wood. That's supposedly one reason why it is often used in headers in place of natural wood.

Would it be generally true that, other than density, the two most important structural parameters for aircraft wood are modulus of elasticity (MOE) and modulus of rupture (MOR)? What other structural parameters are of particular importance?

[massmanute]

Adding one more thought to my last post, what is the relationship between allowable bending stress and modulus of rupture? I assume that allowable bending stress is probably the modulus of rupture multiplied by some kind of safety factor. Is that right?

I can sometimes find allowable bending stress for LVL, but not modulus of rupture. I can find modulus of rupture for Sitka Spruce, but not allowable bending stress. It would be good to have parameters in a form that are directly comparable.

[FlyingRon]

Not entirely related at all. Usually LVLs are rated in elasticity (deflection) and overall point strength (in psi). Those are stamped right on the beam. I've never had a problem once you know whose product you're using. Here's the one form Weyerhaesur's microllam lvl: https://www.weyerhaeuser.com/application/files/4415/1128/9357/TJ-9000.pdf

[massmanute]

Not entirely related at all. Usually LVLs are rated in elasticity (deflection) and overall point strength (in psi). Those are stamped right on the beam. I've never had a problem once you know whose product you're using. Here's the one form Weyerhaesur's microllam lvl: https://www.weyerhaeuser.com/application/files/4415/1128/9357/TJ-9000.pdf
Thanks for the comment and data sheet. The data sheet is pretty typical of the ones I have found on the web.

The problem with using these data sheets to compare LVL specifications to (for example) Sitka spruce, is that the specifications are not necessarily reported in comparable terms. For example, Wood Database (https://www.wood-database.com/sitka-spruce/) reports the modulus of rupture for Sitka spruce as 70 MegaPascals or 10,150 psi. The webpage http://www.matweb.com/search/DataSheet.aspx?MatGUID=1e56abdf98904f2ca53bff4bd12 50cab&ckck=1 gives the same values. The weyerhaeuser data sheet does not specify the modulus of rupture, so the data in that sheet can't be directly compared to the bending strength for Sitka spruce.

I am pretty sure that the strength values from manufacturers data sheets are given in terms of allowable design values rather than actual average values for the material, but I believe that various wood databases usually give average strength values. Generally speaking, an engineer will take into consideration various factors (including statistical variability of material specifications between different specimens among other factors), so they do not design to the average strength of the material but to a derated strength. That way their designs are not flirting with getting too close to structural failure. The easy way to do this is to use allowable design values rather than average material values

The modulus of elasticity is one thing that can be compared between the two sources, but of course it is less interesting than strength values. This is not to say that it is of no interest because it relates to the flexibility or "whippiness" of the design, which I suspect is of at least some interest to aircraft builders. If one wants a less whippy design then, all else being equal, LVL would be a better choice than spruce because its modulus of elasticity is greater. I am pretty sure its strength is also higher, but as I mentioned before, I can't find strength figures quoted in a way that allows direct comparison between LVL and spruce.

[Bill Berson]

How thick is each veneer?
Something like 1/8" would have less glue weight.

[massmanute]

How thick is each veneer?
Something like 1/8" would have less glue weight.
I have a couple of scraps of LVL that were left over from a construction site. I just did some measurements on it, and the veneers averaged 0.125" thick and the density was 32.4 lb/ft^3. I don't know the grade or the material type, but visually it looks like Douglas fir, and the density would be consistent with it being Doug fir.

There are knots in the veneers, but because the veneers are only 1/8" thick and are randomly distributed, they have relatively little effect on the strength of the full-size material. Of course, if the LVL boards were cut down to smaller dimensions (e.g. to make stringers) the statistics of the knot distribution would require that the material be de-rated somewhat below the full-size specified values.

[FlyingRon]

Cutting down LVL is going to be problematic if you're talking about perpendicular to the plies. You're not going to be able to make any strength guarantees.

Yep, comparing LVL numbers to spruce specs is going to be problematic given the spruce specs are generic and LVLs are pretty specific to using them for beams. In construction we're only much concened about the point strength and mode and we only measure that downward on the top of the beam.

[massmanute]

The document "PROPERTIES OF LUMBER PRODUCTS" by Hernandez and Green, found at https://pdfs.semanticscholar.org/e901/3beccfea6aa2a6804b6d38eedacf48d7750a.pdf?_ga=2.136 770553.875849538.1561754811-168971952.1559122033 gives some engineering data that seems relevant. They list design values for various wood species of various grades. On page 6-122, for the highest grade of Douglas-fir/larch (select structural grade) they list a bending strength (fb) of 1500 psi. Note, this is basically an allowable design value which is quite a bit less than the values for average bending strength that you might see listed in some places. For the highest grade of spruce/pine/fir they list 1250 psi.

On p. 6-127 also give an example of bending strength for an LVL product of 2800 psi. This is fairly typical of values you will find in manufactures literature, but it is by no means the highest value one can find. For example, the Murphy LVL technical design guide lists a bending strength (fb) of 3100 psi for their LVL products, which is more than twice as high as the best grade of doug fir/larch. On page 6-125 they also make the comment "Generally the engineering design properties compare favorably with or exceed those of high-quality solid dimension lumber."

In terms of what would happen to the structural parameters if LVL is cut down, it is good to keep in mind how LVL is manufacture. It is manufactured in wide billets, which are then cut to less-wide sizes to give the standard widths sold in commerce. Manufacturers provide formulas for converting their standard values for bending strength, which are generally based on a 12" wide format, to other widths. For example, Murphy uses a multiplier based on the expression (12/d)^0.18. Thus, a their 5.5" wide board would be rated at 3567 psi whereas a 12" wide board would be rated at 3100 psi.

Manufacturers don't tell you how to rate an LVL board that you cut yourself. Some in fact tell you not to cut. That's probably lawyer talk to limit their liability. However, keeping in mind how LVL is manufactured, in which less-wide boards are cut from wider billets, which is similar to what one would do at home to make a less-wide board from a wider one, it seems reasonable to apply the same formulas they use to calculate fb for their standard sizes, e.g. (12/d)^0.18. Even if this is a not quite perfect predictor, it seems to me that the allowable bending strength of LVL is so far above that of natural spruce or Doug fir that it would provide an inherently large effective safety factor that would easily take up any slack in the analysis. And of course, one could also do one's own testing for structural members that are not too thick, such as material for stringers, just to be sure.

[Bill Berson]

32 pounds per cubic foot is same as Doug fir (from memory). So the glue weight is minimal. Sounds good to me.
Could boil a sample a few hours to test.