Login Form

Q-Talk 166 - Composite Build Competition at FOD?

Guys,
I would like to propose a composites building competition at FOD this year.

I'm thinking give everyone 30 days or so to build a Rutan confidence layup (bottom of page) using whatever epoxy or process they like, with the requirement that they document it fully. You bring it with you to FOD for testing with the group.

At the field, we have a gauge to verify it does not exceed the cross-sectional area limitation, and a scale to weigh the part. A couple of bricks and some weights (with a larger scale) would suffice to test the parts. You can enter several parts, but only your top result is used for scoring. Score is based on *specific* strength, ie failure load normalized by part weight.

I can probably put together a lot of the competition except some of the parts: the heavy duty scale (300-350lbs max), bricks, and a bunch of weights that I'm not going to fly there with (barbell weights would suffice). It'd be nice if someone with access to a laser cutter could cut out the gauge for accuracy.

In any case, this seems like a good way to both get a sense for the build quality that people are turning out, and to encourage good practices (eg, VARTM process generates stronger parts, etc). I've specifically set up the requirements so that it can be tested safely and at home so that people can go practice themselves.

If all this sounds a bit wordy--I took a stab at this process a while back myself (Shown Below.) Let me know if you think this would be a popular event and we can start formalizing the rules and maybe some kind of prize. I know I'd have a blast testing parts with you guys but we need about 10 entries to really start learning from it.

Best,
Imraan


LEGAL DISCLAIMER: This document is provided as an incomplete record of events to inform those conducting similar tests and is not intended to serve as design advice. Author does not warrant that it is complete, comprehensive or accurate, or commit to its being updated. You agree that making this information available shall not be seen as the provision of design advice, and therefore I, my heirs, employer, and/or any other agent acting on my behalf are not liable in any way for its use or for the consequences of any actions taken on the basis of the information provided.

Load Testing of a Composite Fiber-Reinforced Laminate Beam

Imraan Faruque
This email address is being protected from spambots. You need JavaScript enabled to view it.
3182 Glenn L Martin Hall
College Park, MD, 20742, USA

I. Introduction

Due to my inexperience at composites work and the need to perform a structural repair on my aircraft, I elected to evaluate both the original load characteristics of a composite fiber reinforced laminate structure and the characteristics after subsequent failure and repair.a

II. Theoretical Stress Estimate

The Rutan confidence layup described in Moldless Composite Construction was selected as a standard test specimen. The beam is composed of a 1/4in urethane foam core with 2 layers each of bidirectional and unidirectional woven glass on the top and bottom as seen in Fig 1(a). The relative simplicity of this structure allows for analytic results and straightforward reproduction by other researchers of the results described herein, and the sample lends itself well to simply supported moment testing.

The structural member was modeled as a simply supported beam loaded in the geometric center of the beam length as seen in Fig 1(b). This load condition imposes both sheer stress and a bending moment stress, however, the bending moment stress will be shown to be higher and is expected to cause failure, as glass-fiber and resin structures have a lower compressive yield stress.

 
Figure 1. Rutan confidence layup cross sectional dimensions and load test.

Figure 1. Rutan confidence layup cross sectional dimensions and load test.

 

A. Shear Stress Calculations

The beam has an estimated cross sectional area of Aw = 0.18818in2. Shear force is maximal at the supports with Vx = F/2, and sheer stress Shear Stress Formula.

B. Bending Moment Stress Calculations

Equivalent peak bending stress at break force Fb is

Break Force Formula

If modeled as a solid structure, an equivalent section modulus for a solid structure is Sx, s = 0.06913in3 .

However, the structure is not solid. Approximating as a hollow rectangle, Sx = 0.03387in3 .

III. Experimental Methodology

HexCell's EZ Poxy was selected, due to its favorable structural properties. Epoxy was mixed by weight with a ratio of xxxx. Initial layup was conducted at a temperature of 77°F and 40% relative humidity on a glass preparation surface cleaned with acetone. The sample was allowed to cure for 18 hrs before being removed from the surface and an initial trim conducted using aviation shears.b The sample was then postcured in an unventilated glass enclosure subjected to solar heating for 42 hrs.

 
Figure 2. Layup construction.

Figure 2. Layup construction.

 

Load testing was conducted as a simply supported structure, using a pair of standard bricks and a load applied at the center. The load testing fixture is shown in Fig 3.

 
Figure 3. Load testing fixture under load, and failed component.

Figure 3. Load testing fixture under load, and failed component.

 



IV. Results and Discussion

A. Original load test

Load testing began at 145lbs and increased to 210 lbs. Due to lack of calibrated weights greater than 210 lbs, dynamic loading of the 210lb weight was conducted, and failure occured at an estimated 225 ± 5lbs. Failure occured at the point of load application, showing compressive failure and delamination in the upper surface. No damage was apparent visibly in the lower surface.

While supporting 210 lbs, the sample experienced peak shear and bending stresses of 0.56 ksi and 25 ksi, respectively. At 225 lbs, the sample experienced peak stresses of 0.60 ksi and 27 ksi. Room temperature cured E-Z Poxy alone has a strength of approximately 8200 psic , indicating that the glass fibers have significantly strengthened the structure.

B. Repair

The sections that were visibly damaged were removed as seen in Fig 4 (approximately 1.5 in long), and a urethane foam plug shaped to replace the removed foam. The foam was painted with a 1:1 micro slurry, and remaining voids filled with 2:1 flox mix. Actual epoxy mix was xxxx. 2 layers of BID and 2 layers of UNI measuring 3 in long were then applied, and the part allowed to cure for approximately 68 hrs in the solar enclosure.

 
Figure 4. Damaged material removed for repair.

Figure 4. Damaged material removed for repair.

 

C. Repaired load test

Load testing again began at 145 lbs, and proceeded to 210 lbs without only elastic deformation. After a 230 lb load was added,d plastic yield failure was observed. Again, failure was in the compressive loading of the upper surface as seen in Fig 5, but the failure occured adjacent to the repaired area, suggesting that the repaired area had been restored to full strength and the lack of staggered plies now created a stress concentration. Minor damage was noted to the lower plies as well, as seen in Fig 6.

At a 230lb load, peak sheer and bending stresses were 0.61 ksi and 27 ksi.

 
Figure 5. Repaired beam after testing.

Figure 5. Repaired beam after testing.

 
Figure 6. Failure after repair, including two damaged points on the lower surface.

Figure 6. Failure after repair, including two damaged points on the lower surface.

 

V. Conclusions

A composite beam structure was fabricated and loaded in bending. After failure, the sample was repaired and re-tested. The original and repaired samples exhibited failure at an overall load of 225 lbs and 230 lbs, respectively. Both held a proof load of 210 lbs, corresponding to a solid-beam-equivalent peak bending stress of 12 ksi. Modeled as a hollow rectangular beam, in both cases, the structure failed betweend 25 and 27ksi of bending stress. While he original sample failed at the point of loading, the repaired sample failed at the terminus of the plies added during repair, suggesting that the additional plies had concentrated stresses at the edge of the repair. Additional stagger of the plies could reduce this concentration. Additionally, unavailability of a loading structure with adequate capacity complicated the original test, and it is suggested that the original test should be repeated with a calibrated load to failure.

Footnotes

a This document is provided as an incomplete record of events to inform those conducting similar tests and is not intended to serve as design advice. Author does not warrant that it is complete, comprehensive or accurate, or commit to its being updated. You agree that making this information available shall not be seen as the provision of design advice, and therefore I, my employer, and/or any other agent acting on my behalf are not liable in any way for its use or for the consequences of any actions taken on the basis of the information provided.

b Minor delamination was sustained near the terminus of the lofted structure during removal. The sample was still usable for load testing, as this area is subjected to low bending moment stresses.

c Manufacturer data. Increased to 10 ksi if postcured 150◦ F for 2 hrs

d 210 lb human + 8x8x16 cinderblock


From Q2 Plans - Page 3-20


CONFIDENCE LAYUP

.....The second practice layup is one intended to give you confidence in the strength of your work. This layup is a sample of composite sandwich structure and is typical of the load carrying structures in your Q2. When this layup is finished, and completely cured, you will subject it to a simple load test, and thus demonstrate the strength of your workmanship.

.....First, tape a piece of waxed paper about 30 inches long to the top of your work table. Shape a piece of green foam as shown.

 

.....Go to your glass cutting area and cut the glass plies shown.

 

 

.....Lay up two plies of UNI, two plies of BID, paint the foam with micro slurry, and press it in the center. Then lay up the other BID and UNI plies. Be careful to work a11.air bubbles out of the corners. The best way is to stipple with the brush. The glass is oversized so that it can be trimmed to exact dimensions later. Trim to the dimensions shown after curing 24 hours, using a coping saw or band saw. Allow the piece to cure for four days at room temperature before the load test.

 

 

.....Now for the test: lay a broom handle or piece of tubing on the work bench and try to break the sample by putting all of your weight on the ends. A 200 pounder will stress the sample more than any part of your airplane is stressed at 10 g's.

 

 

 
PAGE 3-20