Quantcast

Orion's "Composite Think" in Plywood???

HomeBuiltAirplanes.com

Help Support HomeBuiltAirplanes.com:

BBerson

Light Plane Philosopher
HBA Supporter
Joined
Dec 16, 2007
Messages
14,011
Location
Port Townsend WA
I was thinking thin plywood could be made in corrugated sheets to be used as sandwich core instead of the usual foam or honeycomb.
It would be exactly the same as any ordinary cardboard box (which is made with cheap corrugated paper core).
 

highspeed

Well-Known Member
Joined
Apr 10, 2011
Messages
275
Location
Spring, TX
As far as that goes, no one actually "knows" the properties of any material---aluminum, steel, wood, or composite!
I beg to differ when it comes to the metals. Every batch (heat) of material from the mill gets tested for mechanical properties and chemistry. The basic sample undergoes a standardized tensile test that tells you the yield strength, the ultimate tensile strength, the reduction of area, and the elongation at the fracture point. This is the fundamental stuff. Often a charpy impact test is performed to determine the toughness of the material. So you can "know" the properties of a material. Knowing the loads the material is subjected to might be another matter entirely, but the material properties can be pretty well known.
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,904
Location
Evansville, Indiana
I beg to differ when it comes to the metals. Every batch (heat) of material from the mill gets tested for mechanical properties and chemistry. The basic sample undergoes a standardized tensile test that tells you the yield strength, the ultimate tensile strength, the reduction of area, and the elongation at the fracture point. This is the fundamental stuff. Often a charpy impact test is performed to determine the toughness of the material. So you can "know" the properties of a material. Knowing the loads the material is subjected to might be another matter entirely, but the material properties can be pretty well known.
Are you willing to bet your life and those of passengers on those numbers without a safety factor? I am not and neither are trained engineers. The other issue is that the strength of a material when it comes out of the mill in sheet form and its strength after being milled, shaped, drilled, riveted and then exposed to the fatigue of use are often two disparate numbers.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,623
Location
Marion, Ohio
So you can "know" the properties of a material. Knowing the loads the material is subjected to might be another matter entirely, but the material properties can be pretty well known.
One of my previous incarnations was as a Measurment Scientist for the USAF, so I can speak with some authority on this subject. We calibrated test instruments that were used to calibrate the instruments that were used to test the aircraft.

Every instrument had a range of uncertainty. We "knew" that the value was between 5.6 and 5.8 with 95% confidence, for example, but we still did not "know" the exact value. (Note that there was still a 5% chance that we had no idea at all!)

The same thing would apply to testing a piece of aluminum. The instruments used have some uncertainty. Maybe the tech was having a bad day. Maybe it was raining outside. Same sample, different lab, different numbers. Same lab, different sample, different numbers. And exactly how many atoms of copper are in the piece of 2024? Can't say, can you?

If we were able to "know" the properties of the material exactly then we would also be able to "know" the forces on a plane, and then use that safety factor of 1.00001! But, of course we don't, so we use the factor of 2.0, most of which is from not knowing the forces I would guess.
 

highspeed

Well-Known Member
Joined
Apr 10, 2011
Messages
275
Location
Spring, TX
Are you willing to bet your life and those of passengers on those numbers without a safety factor?
Absolutely not. I have always assumed a factor of safety of at least 2 in my design exercises. And you're absolutely right about the properties of the part after milling, drilling, riveting, etc. I wasn't trying to say that you can look at a test report and design the component so the failure point is a hair past the yield point of the material. That would be foolish. I'm in Quality Assurance for a manufacturer in the oil and gas industry and most, if not all, parts are designed to require a minimum yield strength. The minimum includes the factor of safety, of course.

One of my previous incarnations was as a Measurment Scientist for the USAF, so I can speak with some authority on this subject. We calibrated test instruments that were used to calibrate the instruments that were used to test the aircraft.

Every instrument had a range of uncertainty. We "knew" that the value was between 5.6 and 5.8 with 95% confidence, for example, but we still did not "know" the exact value. (Note that there was still a 5% chance that we had no idea at all!)

The same thing would apply to testing a piece of aluminum. The instruments used have some uncertainty. Maybe the tech was having a bad day. Maybe it was raining outside. Same sample, different lab, different numbers. Same lab, different sample, different numbers. And exactly how many atoms of copper are in the piece of 2024? Can't say, can you?

If we were able to "know" the properties of the material exactly then we would also be able to "know" the forces on a plane, and then use that safety factor of 1.00001! But, of course we don't, so we use the factor of 2.0, most of which is from not knowing the forces I would guess.
Measurement uncertainty always rears its head. I agree, you can get close and know the strength of the material to within a certain range, but the sample is just that, a sample. You will always see variation in material. We'll see it from one end of a 10 foot bar of steel to the other quite readily with the appropriate test. Engineering is all about assumptions and generalizations, hopefully to the side of caution. One of the important things about all that material testing is that is shows if the mill is producing a consistent product with variation that is of an acceptable level. We often order supplemental testing of the material, and is always comes back with a slightly different number. If the difference is too great, we'll investigate further.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,623
Location
Marion, Ohio
The minimum includes the factor of safety, of course.
A table of properties for a material usually has a statement about 'minimum' or 'typical' values. Sometimes they give a range.



[Quote... always comes back with a slightly different number. If the difference is too great, we'll investigate further.
I should have added to my monologue on measurement uncertainty that the variation in wood is several orders of magnitude greater than aluminum, and I would guess that composites are somewhere inbetween.
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,904
Location
Evansville, Indiana
Absolutely not. I have always assumed a factor of safety of at least 2 in my design exercises. And you're absolutely right about the properties of the part after milling, drilling, riveting, etc. I wasn't trying to say that you can look at a test report and design the component so the failure point is a hair past the yield point of the material. That would be foolish. I'm in Quality Assurance for a manufacturer in the oil and gas industry and most, if not all, parts are designed to require a minimum yield strength. The minimum includes the factor of safety, of course.
That's what I figured you recognized but I did not want to assume and risk something bad happening.

I should have added to my monologue on measurement uncertainty that the variation in wood is several orders of magnitude greater than aluminum, and I would guess that composites are somewhere inbetween.
So what safety factor is recommended for a wood structure? I have always used 3-4 depending upon the specific structure.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,623
Location
Marion, Ohio
Ever seen two pieces of wood that even looked the same?

The variation between one piece of wood and the next is a lot more than the variation between one piece of aluminum and the next.

Whatever material is used, the structure should be designed with the minimum specification in mind, along with an appropriate safety factor.
 

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,623
Location
Marion, Ohio
So what safety factor is recommended for a wood structure? I have always used 3-4 depending upon the specific structure.
I'm sure the 'experts' on here would have an opinion on that, but my opinion is that a FOS of 2 would be OK if you base it off the minimum specification of the particular type of wood. 3-4 would be even better.
 

autoreply

Well-Known Member
Joined
Jul 7, 2009
Messages
10,755
Location
Rotterdam, Netherlands
I'm sure the 'experts' on here would have an opinion on that, but my opinion is that a FOS of 2 would be OK if you base it off the minimum specification of the particular type of wood.
Even 1.5 would do then.
3-4 would be even better.
With the possible complication that your aircraft is so overweight that it won't fly anymore. Bumping up the F of S is what you often see with "amateurs". Doesnt' work, because overall the design gets worse, not better. Design for the appliceable loads, not more. Some "extra reinforcement" is pointless, unless that part is under strength of course.
 

WonderousMountain

Well-Known Member
Joined
Apr 10, 2010
Messages
2,124
Location
Clatsop, Or
Personally I like to work from the design envelope perspective, identifying what possible maneuvers stress the airframe to it's limit, then I apply material fatigue for 1000 hours minimum, and finally top it off with the 1.5 safety factor. This sometimes comes out to the 3-4 "safety factor" and sometimes is not much bulkier than the original 1.5, depending on the component. Heavy parts are downsized, or redesigned if possible. It takes longer for initial calculations, but is much better for understanding in the end. Vne is critical component of this, so choices have to be made early on.

Blessings,

Wonderous Mountain
 

SVSUSteve

Well-Known Member
Joined
Aug 20, 2007
Messages
3,904
Location
Evansville, Indiana
With the possible complication that your aircraft is so overweight that it won't fly anymore. Bumping up the F of S is what you often see with "amateurs". Doesnt' work, because overall the design gets worse, not better. Design for the appliceable loads, not more. Some "extra reinforcement" is pointless, unless that part is under strength of course.
If done carefully and only for extremely critical components (in the case of the Vireo, this would be the spars and the roll cage which are directly connected), it should not be an issue. Everything else that is critical to safety, it is a factor of 2. All other structures were designed with a factor of 1.5 I went with that approach and still came out with an aircraft that meets the LSA standard.
 

Barracuda

Member
Joined
Jan 6, 2012
Messages
8
Location
ACT
Those tophat stringers are perfectly suited to being moulded in composites, to make them in wood would be a pain. There's no reason you can't make/buy the tophat sections and bond them to plywood, it would work well. It would be versy easy to make up a mould on a table and lay up stringers like that out of biaxial etc, then just cut to length.
Cored plywood structures have been used in boats for ages, often cores of Nomex or even honeycomb are faced with a ply to give strength and stiffness to a bulkhead with a huge reduction in weight. As long as the ply is prevented from deflecting, it can do it's job as well as a much thicker, solid ply section.
You'll find ply moulded furniture is made in large, bespoke hot presses where the adhesives can be applied uncured, the ply loaded in wet and the whole lot heated and pressed until adhesive cure. To replicate it in ply, you would be vacuum bagging and trying to ensure correct adhesion between all the various materials.
 

Dan Thomas

Well-Known Member
Joined
Sep 17, 2008
Messages
5,555
1.5 safety is what Cessna uses when they build their airplanes. For instance, the 172 is rated at 3.8 positive Gs, IIRC, so it should take at least 5.7 Gs to break one. Now, they might be looking at yield rather than ultimate, so maybe it will bend some first at that load before it breaks.

I wouldn't want to go find out.

As far as compound curvature in plywood: this can be very hard to achieve. My son and I built a small boat that had some minor compound curves and it was a bear to fight that 1/4" pine ply onto the frames. Steaming may have helped, but those fibers aren't inclined to stretch much. Bending a really thin piece along the grain will place enormous stresses on the grain and it will permit that, but adding another dimension of curvature is really difficult.

Some boatbuilders do it by laying up the hull with several layers of thin veneer strips. I think the deHavilland Mosquito was built this way. The builder is basically making his own plywood in the shape he wants.

See this: Gentry boat plans a

Dan
 

Attachments

Aerowerx

Well-Known Member
Joined
Dec 1, 2011
Messages
5,623
Location
Marion, Ohio
Seems that this thread has gone far beyond my initial simple question!

Refering back to Orion' truss-type sparless composite structure, all I wanted to know is, if it was done in plywood how it would compare with the typical built-up wood ribs and spar.

I dont' see why the plywood would have to be molded. Why not just cut some long plywood strips and place them inside the wing spanwise, arranged in the truss type configuration? Glued directly to the skin at top and bottom. Yes, some type of reinforcement might be needed at each joint.

As for the skin, maybe 1/8th or 1/16th inch ply that was steamed or soaked could be bent enough to form the surface. Although some type of form or mold would be needed to keep it in place while drying.

Or maybe a geodetic type arrangement of 1/16th inch ply strips, and then cover that with a solid layer of 1/16th inch ply?
 

wsimpso1

Super Moderator
Staff member
Log Member
Joined
Oct 18, 2003
Messages
7,291
Location
Saline Michigan
Good schemes exist for each material. The one Aerowerx is asking about is a good one for composites, but I suspect a very difficult and somewhat heavy one for wood. The ultimate way to know is to design an optimized one in both carbon and in veneer, and compare the weights. The way that some folks would propose is to compare the strength to weight ratios of the materials, but that does not take into account things like minimum gauges for each, adhesive weight, and the many regions of each that are not optimized...

I would build plywood in the conventional way, and if I wanted to build a hollow tail, I would build it via the method in the original post.

Billski
 

orion

Well-Known Member
Joined
Mar 2, 2003
Messages
5,800
Location
Western Washington
Keep in mind though that the original idea of this concept/discussion was the minimizing of part count. Doing this in wood would be very time consuming and would most likely end up with as many parts as in a conventionally built up wing. Then given the issues of wood working, the resulting structure will probably be measurably heavier and one would still need to address how to form the skin shape. Bottom line, this is unlikely to work in wood.
 

TFF

Well-Known Member
Joined
Apr 28, 2010
Messages
13,832
Location
Memphis, TN
Building truss wood ribs is really easy and for wood light. only a little slow. If the trusses get replaced with longitudinal ply that go straight to the skin, its going to need some blocking to keep in place all the way down the skin. To replace a solid wood spar requires a lot of ply which is weight. You would end up with a bunch of wasted strength in not useful directions with the ply, along with the weight. The Fokker Triplane used a box spar which is I bet where the jodels get their ideas. 2 sets of 2 spruce spars boxed with ply and then boxed together. Fokker DR-1 Photo Gallery I bet it could be simplified.
 
Top