THE SAGA OF A CARBON SPAR DESIGN BY JIM MARSKE

When I went about designing the wing spar for the Genesis 2, I tried to find design strength values for the carbon rovings to be used in the spar caps, The manufacturer's data sheet claimed 310,000 psi tensile strength but gave no compression data. A call to the manufacturer produced a claim of 100,000 psi in compression. Talking with others who have constructed carbon wing spars, I was advised to be careful as such values cannot be obtainable and to back off to 90,000 psi in tension and 60,000 in compression. To satisfy myself, I had several test strips made up of hand laid-up carbon rovings and sent them to an independent test laboratory for strength evaluation. The results of the five test samples was disturbing. In tension the values ranged from 152,000 to 190,000 psi. In compression the values ranged from 48,000 to 74,000 psi. The reason why there was so much scatter and the values lower than expected is that it is almost impossible to lay the rovings down without zigzag waves every few inches. You can only pull the tow till it's shortest filament pulls tight and the rest lay in small waves. As a result of this condition, and to acquire the necessary strength, the G1 prototype spar is fairly heavy and very stiff. To prove the spars strength, we static loaded it to +5g's and -3g's. Wing deflection, measured at the tip, was 25 inches at+5g's.

We needed carbon that was stronger and more consistent in strength values, Therefore, we looked into pulltruded carbon rods, Samples were ordered and examined, but waves were found in those filaments also, At about this same time, I found a brief article in Sport Aviation concerning a carbon rod expounding very straight filament alignment. A sample rod was obtained and after examination of the filaments, we saw that they were indeed very straight. Furthermore, the manufacturer claimed a tensile strength of 315,000 psi and a compressive value of 200,000 psiwhich was impressive. When bending the 1 /8 inch diameter rod to first sign of fracture, it was the tensile filaments that were failing one at a time. In addition to the high strength properties of these rods, the automated manufacture of the rod controlled the resin content, which is not possible in a hand laid-up situation, and dimensions of the rod assured consistent strength properties.

The next step was to prove that the rod would function as a spar cap, and that it would carry the required loads without delamination. A Genesis spar segment of the aircraft center section was made and tested in the Sportine Aviacija laboratory. No failure occurred during a load sequence to a load limit of +8.3g's. This load represents an aircraft design load of 1200 pounds, times a safety factor of 1 .5 as required by JAR-22. Impressed with this results, the load was increased past the required load limit of +8.3g's to +1 Og's without incident. Going to +10.5g's we reached the maximum output load of the test machine, and again no degradation of the rods was observed. This load was nearly twice the required spars design load of +5.55g's.

Satisfied with the static results, we did not however have a history of dynamic cyclic endurance testing for ihis parTicufar rod. Since fihe majority of the rods do not span fuii lengfih of the spar we had conc:ernas to what would happen at the end of each rod end in the mid section of the spar where a stress riser may occur. So we embarked upon a cyclic endurance test at an elevated load to force an early failure.

The first run was a 4g positive loading. We hoped for 5,000 cycles but stopped at 10,000 cycles. We then increased the load to 6g's expecting a failure in a few hundred cycles, We stopped the test at 5,500 cycles. The test spar was then inverted in the fixture to apply negative loading. The test lab director insisted that we start at -3g. We started at -4g's and ran for 5,000 cycles. No degradation was noted. To finish the test we repeated the static loading test again. One cycle to +8.33g's and two cycles to -5,33g's. Again no degradation was visible.

So we asked Klemas, Sportine Aviacija's chief engineer as to just how many flight hours all this cyclic testing is equivalent to, Klemas gave me a report on recorded accelerations made on one of their LAK-12's during 50 hours of flying, which included towing, takeoffs, landings and ground handling. The accelerations were all counted and grouped together to form a 50 hour flight period. The cycles were then multiplied by 200 to find the life of 10,000 flight hours for the LAK-12.

THE SAGA OF A CARBON SPAR DESIGN continue

This data was then transferred into chart form. I overlaid the Genesis data on the same chart to obtain a comparison. A diagram of the results appears below.

picture: Genesis spar test results compared to the LAK-12 spar test results.

After completing a quick calculation which still requires further evaluation, I feel that we have acquired an excess of 5,000 flight hours (probably 6,000 hours) in positive loading and an excess of 10,000 flight hours in negative loading. I understand that a survey of various glider clubs around the world responded to an inquiry as to the maximum flight hours that had been accumulated on any of their gliders. Only a few gliders had accumulated near 5000 flight hours. However one Australian club reported nearly 6,000 flight hours.

G2 WING SPAR DEVELOPMENT

As mentioned previously, the Sportine Aviacija facility has an extensive engineering test lab and experienced engineering staff. These capabilities in combination with our own engineering efforfs have produced some amazing results in the area of the wing spar development for the G2.

For example, the main spar on the G1 prototype was constructed in the usual manner using hand laid-up carbon fiber roving. However laboratory tests have shown that there is plenty of room for improvement in this process. So we decided to look into using prestressed carbon fiber rods as a replacement to the carbon roving used in the wing spars. These rods alone are five times stronger than conventional hand laid-up carbon roving.

Using the Sportine Aviacija test lab, we prepared a sample for cyclic fatigue testing and took it through 10,000 cycles at +4 g's, 5,000 cycles at +6 g's and 5000 cycles at -4 g's. Then as required for Jar-22 certification, we loaded this same spar sample twice for 10 seconds (once for 3 seconds is all that's required) at +8,3 and -5,3 g's and experienced absolutely no degradation or failure whatsoever, We then took it to 10 g's and also experience no degradation or failure. Certification test results like this are practically unheard of in sailplane development, but that's not to say they shouldn't be.