LVL verses plywood

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

Dana

Super Moderator
Staff member
Joined
Apr 3, 2007
Messages
9,195
Location
CT, USA
Why not just use aircraft grade plywood? It has well documented properties.

And, as mcrae0104 said, LVL will probably be heavy for the strength.
Different application. Plywood is flat sheets, LVL is dimensional lumber like a 2x4, etc. Plywood is laid up with different grain orientations from ply to ply to optimize shear strength. LVL is, I presume, all with grain in the same direction, optimizing tensile and bending strength.
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
Gluelam with the right wood and glue would be perfect....
The problem with gluelam is that they are made of thick boards, so if you rip to size there will be a lot of relatively poorly supported butt joints in a given piece.

There would be butt joints in a piece of ripped LVL as well, but because the laminations are thin there will be a lot of supporting material around them to provide strength.
 

proppastie

Well-Known Member
Log Member
Joined
Feb 19, 2012
Messages
4,501
Location
NJ
I seem to remember the cautions of non aircraft plywood such as voids, and knots.....some talk above about might be one lamination bad and not really that bad, but might there be more than one bad lamination?...Is the fact that these are very thin laminations mean that perhaps there are less large voids? How many knots or voids will you not see in that spar you might be building.
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
I seem to remember the cautions of non aircraft plywood such as voids, and knots.....some talk above about might be one lamination bad and not really that bad, but might there be more than one bad lamination?...Is the fact that these are very thin laminations mean that perhaps there are less large voids? How many knots or voids will you not see in that spar you might be building.
In the typical manufacturing process for LVL they grade the logs before cutting the veneers, and the veneers also individually graded using automated machines and defects are cut out before they are laid up into billets. The processes are subjected to very rigorous quality control. The resulting product is very uniform in quality, generally far more uniform that conventional lumber, probably more uniform than lumber that has been heavily inspected as well.

Here's a link to a video of the production process at one company.
. It's worth watching. There are videos from other companies as well.
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,260
Location
BJC
Regarding strength values, fb for coastal sitka spruce is listed as 1300 in the 2018 National Design Supplement by the American Wood Council (souce: https://awc.org/pdf/codes-standards/publications/nds/AWC-NDS2018-Supplement-ViewOnly-1805.pdf). Looking at some values for LVL, Westfraser lists fb for their lowest grade of LVL as 2750...
While I salute you for looking up the values, NDS is not the right place to look. We're building airplanes, not houses. NDS is for construction lumber of very specific sizes (not used in aircraft) and the grading is very different from aircraft wood grading. Even select structural construction lumber is not necessarily up to snuff for aircraft (although you will find the occasional stud that meets aircraft grading criteria). The NDS values are design values, not average test values, and as such have a factor of safety built in for variation in members. Buildings--especially light wood framed--have massive factors of safety built in compared to aircraft, largely due to the wide variation in the material.

The reference you want is ANC-18 Table 2.3. You will notice when you look there that it is not simply a matter of comparing \(F_{b} \) (allowable bending stress) values from NDS. ANC-18 lists \(F_{bp}\) (stress at proportional limit--akin to the yield point of metals \( F_{y} \)) and \(F_{bu}\) (modulus of rupture--akin to ultimate strength \(F_{u} \)). It's unclear which of these values the NDS \(F_{b} \) corresponds to. It's really worthwhile to read all of ANC-18 chapter 2, and probably to test some specimens of aircraft wood and LVL to ensure the values are apples-to-apples. Arthur W.J.G. Ord-Hume authored some articles on home testing methods in Sport Aviation in the late fifties or sixties that you can probably find if you search the EAA site.

BTW, ANC-18 lists \(F_{bu} \) for spruce at 9,400 psi--way higher than the 1,300 psi listed by NDS. Its strength-to-weight is almost certainly superior to LVL, but should you choose to do some testing I will be interested to read the results.
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
While I salute you for looking up the values, NDS is not the right place to look. We're building airplanes, not houses. NDS is for construction lumber of very specific sizes (not used in aircraft) and the grading is very different from aircraft wood grading. Even select structural construction lumber is not necessarily up to snuff for aircraft (although you will find the occasional stud that meets aircraft grading criteria). The NDS values are design values, not average test values, and as such have a factor of safety built in for variation in members. Buildings--especially light wood framed--have massive factors of safety built in compared to aircraft, largely due to the wide variation in the material.

The reference you want is ANC-18 Table 2.3. You will notice when you look there that it is not simply a matter of comparing \(F_{b} \) (allowable bending stress) values from NDS. ANC-18 lists \(F_{bp}\) (stress at proportional limit--akin to the yield point of metals \( F_{y} \)) and \(F_{bu}\) (modulus of rupture--akin to ultimate strength \(F_{u} \)). It's unclear which of these values the NDS \(F_{b} \) corresponds to. It's really worthwhile to read all of ANC-18 chapter 2, and probably to test some specimens of aircraft wood and LVL to ensure the values are apples-to-apples. Arthur W.J.G. Ord-Hume authored some articles on home testing methods in Sport Aviation in the late fifties or sixties that you can probably find if you search the EAA site.

BTW, ANC-18 lists \(F_{bu} \) for spruce at 9,400 psi--way higher than the 1,300 psi listed by NDS. Its strength-to-weight is almost certainly superior to LVL, but should you choose to do some testing I will be interested to read the results.
Thanks. Where can I find ANC-18? Never mind. I found a source for ANC-18. Is it the 1944 version I should look at?
 
Last edited:

Doran Jaffas

Well-Known Member
Joined
Jun 25, 2019
Messages
151
When I say "100% tested" I don't mean tested to failure. I mean that all of the parts are tested to some specified value, such as the strength values of sitka spruce. Since most grades of LVL are stronger than sitka spruce the LVL would presumable pass without breaking, and if a given piece were to break it would be discarded. It seems to me that this testing scheme might be possible.

Your comment about statistics is well-taken. Of course, the same comment about statistics applies to solid lumber as well. One only has a degree of statistical confidence that a given piece of aircraft grade sitka spruce would meet spec if it were tested to failure. There is always a chance that it is a statistical outlier and will break under load. In the spirit of statistical reasoning, LVL lumber is generally more uniform in properties than solid lumber, and this should give greater confidence that an individual piece would pass muster, even if not individually tested, as long as the test results for similar pieces were good.

Regarding strength values, fb for coastal sitka spruce is listed as 1300 in the 2018 National Design Supplement by the American Wood Council (souce: https://awc.org/pdf/codes-standards/publications/nds/AWC-NDS2018-Supplement-ViewOnly-1805.pdf). Looking at some values for LVL, Westfraser lists fb for their lowest grade of LVL as 2750, a little more than twice as strong as coastal Sitka spruce. Their highest grade is listed as 3100 for fb (source: https://www.westfraser.com/sites/default/files/products/LVL/LVL Users Guide - Canada v0415.pdf) There are of course other strength values to consider as well. Also, other vendors will have different specifications for their LVL products.

With regard to weight, a year or so ago I picked up a couple of scraps of LVL from a construction site. I measured the density of the pieces and they came out to be 35.0 and 35.2 lb/ft3, which is heavier than Sitka spruce (typically 27 lb/ft3) and close to but a little heavier than douglas fir (typically about 32 lb/ft3). It is lighter than another aircraft wood, white ash (about 42 lb/ft3). I should say that I don't know what the strength specifications are for the two random pieces of LVL are that I picked up at the construction site because they weren't big enough to read the full stamp information on the boards, but the wood in the LVL I picked up looks like douglas fir.

Also, the laminations in the pieces I picked up are about 1/8" thick, so a piece cut to, say 3/4", will have 6 laminations. If one layer were bad it might decrease the strength by about 17% worst case (naive calculation, assuming the bad layer had 0 strength). Actually, it typically wouldn't be that bad.

One more thing, this with regard to weight. I just calculated the density of the 3100 fb listed by Westfraser and it came out to be 40 lb/ft3, which is a little heavier than the two pieces of LVL that I picked up. However, in terms of strength to weight ratio (3100/40=77.5) that is still quite a bit better than sitka spruce (roughly 1300/27=48)
I may be incorrect here but...there seems to be a lot of reinventing of the wheel in some of these discussions.
Using tried and true and proven materials tested by those in the business...unless those materials are getting scarce is really the only way to do it safely. The other factors such as glue type, longevity, strength to bond ratio etc all come into play here. I for one, for the sake of saving a few $$ do not want to be wondering if that thump or veer I just experienced is turbulence or something coming undone. Also, when flying through turbulence, we all have or will, I also want complete faith that my aerial steed is not going to start shedding parts
 

Attachments

RogFlyer

Member
HBA Supporter
Joined
Jun 27, 2017
Messages
14
Location
Lewisham
It is practicable to non-destructively test every piece of timber to go into a fabrication. Timber needs to get checked in 4 directions - see following. One testing station is needed, with constant force/variable deflection being easiest - figure out the allowed max deflection for the size and test force being tested, apply that force using a spring tensioner and then a visual check or limit switch can be used as a pass/fail indicator.

So much timber these days is laminated or finger jointed, so I'm not sure if this is still the case. When I was dealing with milled timber for construction use each stick sawn from the log was dimensionally checked along its length and stress graded. Visual grading by trained (and tested) graders happened on the fly and good looking sticks were put aside for what was called low defect engineering lumber (LDEL - apologies for the americanism) and dodgy sticks were also put aside for non-stress graded sale. My experience was with non-destructive machine grading.
If you consider the x-direction to be running along the length of a piece, then the timber was tested in bending in the +/- y and +/- z directions using sets of three rollers. Each measuring station was a metre long, four stations in total over 6 metres. Depending on the grade band expected from the species being milled, we could use either fixed force/variable deflection, or fixed deflection/variable force. A stress grade F4 to F32 was allocated to each 300mm of length. The stick was then run out onto a table and passed to the cut-to-length station. The computerised saw (my bit) figured out the maximum yield grades/lengths practicable (we used to get 2 lengths of logs - cut to length 12.1 metres, or uncut up to 25 metres - and were then sawn to preferred lengths of 2.42 m 4.23m or 6.05m, with possibly short lengths of very low grade timber cut out to the reject bin (our min cut length was 600 mm) then printed with their stress grade group (S1 - S7) and sorted into bins. Knots (hard or loose), dags, resin lines, checks, splits, sapwood, borers, tearout and grain orientation were immaterial as long as the stress test was in range. Logs were typically wet when sawn, so an allowance was made for the gain in strength of dried timber, which in hardwood can be well over 100% from the green state.

At the truss plant, each piece in a job bundle was regraded, numbered and a cut list made up. To take advantage of actual strength gain from seasoning and also to avoid optimistically(!) graded sub-lengths, egregious-looking faults, twisting and bowing.
 

Mad MAC

Well-Known Member
Joined
Dec 9, 2004
Messages
607
Location
Hamilton New Zealand
What we care about here is the standard deviation as applied to youngs modulus and stress, ANC-18 and associated standards apply to visually graded timber, the resultant spread of the properties (AKA standard deviation) is higher than machine graded as I understand it.

A lower strength species that is machine graded can have a higher design allowable than a stronger species that is visually graded (the design allowible is set as x number of standard devations below the average strength). It also means if you can find machine graded spruce its possible to use higher design allowables than ANC-18.

It is possible to buy the plywood sheets that they make LVL out off. Could we treat it a bit like aluminum billet, just hog it out till we get the parts we want.

If you want to really disappear down a dark hole, note that the fatigue strength of timber is determined by how it is felled, handled & milled. Are modern processes less or more detrimental to the timber than 70 years ago.
 
  • Like
Reactions: alr

BrianW

Active Member
Joined
Jul 2, 2018
Messages
36
Location
Altus SW Oklahoma
I've been thinking about different types of woods for a project. I like plywood and have used it many times in building pretty much anything. While doing some napkin engineering, I thought of LVL. Does anyone have any experience in the use of LVL?
LVL is uniaxial or oriented which is a distinct difference from ply. The biggest criticism I have read is that in outdoor applications, where the glue line hinders chemical penetration, internal rotting has been observed (in a deck) rarely.
 

wktaylor

Well-Known Member
Joined
Sep 5, 2003
Messages
184
Location
Midwest USA
The APA Engineered Wood Association Home - APA – The Engineered Wood Association

has MANY technical/practical-oriented references/doc/engineering/use-guides/etc for engineered wood products of all types imaginable... plywood [all types], particle board, glulam beams, Oriented Strand Board OSB, etc... construction, marine, etc

HOWEVER there is strict/limited consideration for these products in aviation...
 

alr

Active Member
Joined
Nov 24, 2004
Messages
35
One source lists the modulus of elasticity for LVL as being 25% greater than spruce and the modulus of rupture as being twice that of spruce.


However, that source is short on details, such as what species and grade of spruce is used in the comparison and what type of LVL is being used in the comparison.

Also, I found a document "Non-Certified Wood Testing and Selection" by Drew Fidoe that describes the author's methods of selecting and testing wood for homebuilt aircraft. From the descriptions in the paper he tests 100% of the pieces used in the construction.
 

mcrae0104

Well-Known Member
HBA Supporter
Log Member
Joined
Oct 27, 2009
Messages
3,260
Location
BJC
One source lists the modulus of elasticity for LVL as being 25% greater than spruce...
That is interesting; perhaps that might be an advantage for a sailplane spar or some other deflection-critical member, if the weight pencils out.
 

JamesF

Member
Joined
Mar 30, 2015
Messages
11
Location
Clovis, NM
Brief description of LVL:

Laminated Veneer Lumber (LVL)
First used during World War II to make airplane propellers, laminated veneer lumber (LVL) has been available as a construction product since the mid-1970s. LVL is the most widely used structural composite lumber (SCL) product and provides attributes such as high strength, high stiffness and dimensional stability. The manufacturing process of LVL enables large members to be made from relatively small trees, providing efficient utilization of forest resources. LVL is commonly fabricated using wood species such as Douglas fir, Larch, Southern yellow pine and Poplar.

LVL is used primarily as structural framing for residential and commercial construction. Common applications of LVL in construction include headers and beams, hip and valley rafters, scaffold planking, and the flange material for prefabricated wood I-joists. LVL can also been used in roadway sign posts and as truck bed decking.

LVL is made of dried and graded wood veneer which is coated with a waterproof phenol-formaldehyde resin adhesive, assembled in an arranged pattern, and formed into billets by curing in a heated press. The LVL billet is then sawn to desired dimensions depending on the end use application.

The grain of each layer of veneer runs in the same (long) direction with the result that LVL is able to be loaded on its short edge (strong axis) as a beam or on its wide face (weak axis) as a plank. This type of lamination is called parallel-lamination and produces a material with greater uniformity and predictability than engineered wood products fabricated using cross-lamination, such as plywood.

LVL is a solid, highly predictable, uniform lumber product due to the fact that natural defects such as knots, slope of grain and splits have been dispersed throughout the material or have been removed altogether during the manufacturing process.

The most common thickness of LVL is 45 mm (1-3/4 in), from which wider beams can be easily constructed by fastening multiple LVL plies together on site. LVL can also be manufactured in thicknesses from 19 mm (3/4 in) to 178 mm (7 in). Commonly used LVL beam depths are 241 mm (9-1/2 in), 302 mm (11-7/8 in), 356 mm (14 in), 406 mm (16 in), 476 mm (18-3/4 in) and 606 mm (23-7/8 in). Other widths and depths might also be available from specific manufacturers. LVL is available in lengths up to 24.4 m (80 ft), while more common lengths are 14.6 m (48 ft), 17 m (56 ft), 18.3 m (60 ft) and 20.1 m (66 ft). LVL can easily be cut to length at the jobsite.

james fuller
 
  • Like
Reactions: alr

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
12,785
Location
Memphis, TN
The use of the plant quality control would be nice for a special run for aircraft woods. Limit the splices to aircraft requirements. Glulam style with 8-15 ft pieces. Thicker laminate of the LVL. Spars are the no compromise part. Finding wood long enough Is becoming an issue no matter the budget. Use the process but aviation grade.
 
  • Like
Reactions: alr
2
Top