Strength of Balsa Spars

telnar1236

Elite member
The assignment was submitted this morning! I am so glad to be finished. It was a very interesting project, but a lot of work. A formula was derived, but I'll post that with an explanation later. By using the formula to graph bending strength against different dimensions, the following two discoveries were made.

1. Strength does not increase proportionally with spar cap thickness. A graph of this is shown below for a spar 12mm wide and kept at a constant total height of 18mm, simulating the thickness of a wing. This is because a bending moment is a force x distance, and the added balsa wood on the inside is at a smaller distance than the rest of the balsa spar caps.
View attachment 206645

2. Bending strength increases greater than proportional to the total spar height. This means when you double the total spar height, the bending increases by a factor slightly greater than 2. This is caused by a similar reason to the spar cap thickness disproportionality.

If you haven't already, you should look into how the area moment of inertia affects bending strength. It might help you combine those two observations into a single statement.

I look forward to seeing your results!
 

quorneng

Master member
1. Strength does not increase proportionally with spar cap thickness. A graph of this is shown below for a spar 12mm wide and kept at a constant total height of 18mm, simulating the thickness of a wing. This is because a bending moment is a force x distance, and the added balsa wood on the inside is at a smaller distance than the rest of the balsa spar caps.

Well done. There is nothing like actual practical experiments to find out what works and what doesn't.
What you have discovered is it the distance between the tension and compression elements that is significant. This suggests the best strength to weight is likely to be achieved with thin but wide spar flanges positioned right at the top and bottom surface of the wing. However a thin spar flange needs support to hold it shape both when in tension and compression. This support also has to resist the shear forces that are created between the spar flanges when under load. This further suggests the material used in the supporting shear web will need different properties to the spar flanges. Things then get rather more complicated.

I look forward to seeing your results in due course.
 

Jackson T

Elite member
If you haven't already, you should look into how the area moment of inertia affects bending strength. It might help you combine those two observations into a single statement.

I look forward to seeing your results!
Yep, done that. The problem is balsa wood is bi-modulus, so the neutral axis is not the geometric centre of the spar. The neutral axis depends on the total height, the compressive and tensile moduli and the spar cap thickness. It is not a constant fraction of the spar height unless you consider the spar cap thickness as having an insignificant effect. For some combinations their effect would be insignificant, but their significance increases as they increase relative to the total height of the spar. I discuss and derive the location of the neutral axis, moment of inertia, and overall bending strength in the assignment, which I'm trying to figure out how to upload here. I did the best I could be researching online, but I'm not sure if it's 100% correct. When I figure out how to upload it I'd love some feedback on any potential errors/weaknesses.
Well done. There is nothing like actual practical experiments to find out what works and what doesn't.
What you have discovered is it the distance between the tension and compression elements that is significant. This suggests the best strength to weight is likely to be achieved with thin but wide spar flanges positioned right at the top and bottom surface of the wing. However a thin spar flange needs support to hold it shape both when in tension and compression. This support also has to resist the shear forces that are created between the spar flanges when under load. This further suggests the material used in the supporting shear web will need different properties to the spar flanges. Things then get rather more complicated.

I look forward to seeing your results in due course.
Thanks @quorneng. It was good to actually build and break some spars, I only wish there was time for more spars for more valid results. I will get the results published as soon as I figure out how to.
 

Jackson T

Elite member
This should work: Strength of Balsa Model Aeroplane Spars. This is the full research report. Some of the formatting and the formulas went funny when transferring from Word to Docs. The abstract is an overview of the whole project, and the method and results sections explain how I tested the spars and how strong they were. The discussion analyses the results, derives the formula and discusses other relevant stuff. The conclusion summarises the discussion in one paragraph, and the appendix includes some more derivation stuff. Please ask questions, tell me what I did wrong/should have done better, and any other constructive feedback. Enjoy!
 

Piotrsko

Master member
Nice work, gave me the proper headache reading the math squiggles, appropriately researched.

Question: how do you assess density? I presume it is by weight for a specific size unless you did the unqualified fingernail smash assessment. And: most aircraft use cantilevered free end unsupported beam structure which your test methods didn't address (which also includes random torsion stresses related to flutter) finally: do your analysis include the latest Prandtyl (?) Loading data where the loads assume a more bell mouth increase of load towards the ends of the wing?
 
Last edited:

quorneng

Master member
Jackson T
A very thorough and well presented report.
If I have a criticism is that your method was likely to demonstrate that the most important factor in the strength of a beam was the distance between its top and bottom surface. Almost any strength of materials text book would indicate that. You did also usefully identify that increasing the thickness of the beam flange does not proportionally increase the strength of a beam.
In the report you identified several different types of spar construction but gave no explanation as to their strengths or weaknesses or the reasoning behind your choice of a particular spar design.

I appreciate you obviously had time and resource limitations and the primary object was to produce a well documented report but I do think you rather skipped over the fact that the detail design of any beam has a very significant effect on both its strength, stiffness and weight, all rather important factors in a cantilever wing application which was the 'target' of the report.

I am not in any way wishing to belittle your work but you did ask for comments.
 

Jackson T

Elite member
Nice work, gave me the proper headache reading the math squiggles, appropriately researched.

Question: how do you assess density? I presume it is by weight for a specific size unless you did the unqualified fingernail smash assessment. And: most aircraft use cantilevered free end unsupported beam structure which your test methods didn't address (which also includes random torsion stresses related to flutter) finally: do your analysis include the latest Prandtyl (?) Loading data where the loads assume a more bell mouth increase of load towards the ends of the wing?
Density was measured by weighing the sheet of balsa and dividing by the calculated volume. While I didn't test any cantilevers directly a bending moment is a bending moment, although other factors like torsion would have an impact. I probably should have acknowledged torsion stresses, even just from the pitching moment. With flutter I guess by the time the wing starts fluttering you're in pretty bad shape anyway. Thanks for your feedback, every little bit helps in developing a better understanding of flight. I haven't been able to fully wrap my head around Prandtl yet, but I look forward to when I do! I haven't heard of any recent developments, got a link?
 

Jackson T

Elite member
Jackson T
A very thorough and well presented report.
If I have a criticism is that your method was likely to demonstrate that the most important factor in the strength of a beam was the distance between its top and bottom surface. Almost any strength of materials text book would indicate that. You did also usefully identify that increasing the thickness of the beam flange does not proportionally increase the strength of a beam.
In the report you identified several different types of spar construction but gave no explanation as to their strengths or weaknesses or the reasoning behind your choice of a particular spar design.

I appreciate you obviously had time and resource limitations and the primary object was to produce a well documented report but I do think you rather skipped over the fact that the detail design of any beam has a very significant effect on both its strength, stiffness and weight, all rather important factors in a cantilever wing application which was the 'target' of the report.

I am not in any way wishing to belittle your work but you did ask for comments.
Thanks.
The different spar dimensions weren't selected to prove any one thing specifically, just a few differences to provide a small range of sections. While it is obvious that spar height is an important factor, it's exact effect (greater than proportional) was not previously known, at least by me. Good point on the lack of discussion on different spar designs, that really should have been included. Thanks for pointing it out ;).
 

Jackson T

Elite member
New question: will this lead somewhere, or is it a one shot school paper? One hopes for the first answer.
It was a one-off school paper, but there's no reason why myself or anyone else couldn't further it. Where in particular should this lead? I'll certainly use it when designing my planes, but as a rough guide only. Applying the formula to my 4m glider, it should have been good for 2.5G's without the carbon tow or sheeting. 2.5G's is pretty small, but I probably would have called it good enough with the sheeting included considering it's mission.
 

Piotrsko

Master member
Well considering that a 747 was only good for + 2.5 G & -1.5G....... However you're talking a lot of weight loaded.

Keep the technical stuff up, you now know more than 99.5% of everyone else about spars
 

quorneng

Master member
You have to be careful considering the possible G loadings in models. Remember there is no 'feel' through the controls or by the seat of the pilots pants.
I consider 4G is a safe figure for normal RC flying but throw in energetic fast aerobatics and you might need 2 or 3 times that.
 

telnar1236

Elite member
You have to be careful considering the possible G loadings in models. Remember there is no 'feel' through the controls or by the seat of the pilots pants.
I consider 4G is a safe figure for normal RC flying but throw in energetic fast aerobatics and you might need 2 or 3 times that.
I definitely agree. I tend to design for +10 -5 on anything that will be performing aerobatics.

The other thing to keep in mind is damage on the ground. Frequently the highest loads are seen due to incautious handling and transport.
 

Piotrsko

Master member
I built some aerobatic gliders once that had engineered spindly looking spars supposedly good to 20 g. Static tested one and got to 25 G. Broke both of them flying. Went back to SWAG engineering, haven't had a failure since, but I do tend to build too heavy.
 

telnar1236

Elite member
I built some aerobatic gliders once that had engineered spindly looking spars supposedly good to 20 g. Static tested one and got to 25 G. Broke both of them flying. Went back to SWAG engineering, haven't had a failure since, but I do tend to build too heavy.

Wow! What did you do that poor plane? Obviously I joke, but pulling 20g is impressive!
 

Jackson T

Elite member
I built some aerobatic gliders once that had engineered spindly looking spars supposedly good to 20 g. Static tested one and got to 25 G. Broke both of them flying. Went back to SWAG engineering, haven't had a failure since, but I do tend to build too heavy.
What was the design and materials of those spars? 25G's is massive! How did you manage to break them, square loops or something :p.
 

Piotrsko

Master member
Attempting squareish loops after diving from height. Never got past the first leg of the loop, either fluttered during the dive, or the pull up broke and folded the wings always at the fuselage. Not hard to get super high loading going really fast and adding abrupt direction changes. And while control line combat does do square loops, adding 1970's vintage RC to that wing is too much mass. Remember we didn't have carbon fiber back then, glass was too heavy and 2" thick wings were too slow.
 

Piotrsko

Master member
Oh and back then you could still buy 3/16 certified aircraft plywood in 4' x4' sheets for $10 USD. so 36 inch spar was 1/2h x 3/8 thick 2 strip lamination with top and bottom 1/2 w x 3/16 caps all plywood aliphatic resin glue. So about a 1" thick spar. If I wanted it really strong I added front and back balsa box webbing between ribs but that was generally way too much.
 
Last edited:

Jackson T

Elite member
Oh and back then you could still buy 3/16 certified aircraft plywood in 4' x4' sheets for $10 USD. so 36 inch spar was 1/2h x 3/8 thick 2 strip lamination with top and bottom 1/2 w x 3/16 caps all plywood aliphatic resin glue. So about a 1" thick spar. If I wanted it really strong I added front and back balsa box webbing between ribs but that was generally way too much.
No wonder your spars were so strong! If only we had access to that now. Even balsa wood is getting expensive here. I can remember when it was half its current price, and I've only been building for 5 or 6 years!