HISTORY OF FLIGHT On August 6, 1993, at approximately 1140 central daylight time, a Piper PA-32-300, N479KC, was destroyed when it broke up in flight and impacted terrain about seven miles southeast of Delight, Arkansas. The pilot and five passengers were fatally injured. Visual meteorological conditions prevailed at the site: however, the airplane was in instrument meteorological conditions when contact was lost. An IFR flight plan had been filed. The pilot contacted the McAlester (OK) Flight Service Station via radio at 0924 while in flight and, after receiving a briefing on weather conditions in the Texarkana, Arkansas area, filed an IFR flight plan. At 1035, he requested clearance from the Fort Worth Air Route Traffic Control Center (ARTCC) to descend from 5,000 feet to 4,000 feet "to see if we can make a visual approach into Nashville, Arkansas." The controller said this was not possible due to the minimum vectoring altitudes (MVA) in the area, but cleared him to "cruise" to the Nashville Airport at 5,000 feet. At 1047, after reporting he was still in the clouds and could not see the airport, the pilot was given radar vectors to intercept the localizer course at Texarkana Airport, and was cleared to descend to 2,100 feet. At 1059 the pilot asked, "Can we go ahead and go on our own and get back to Nashville now and go through the clouds. I see a shot where we can make it." He then cancelled his IFR flight plan and radar services were terminated. At 1110, the pilot reported back on the Fort Worth ARTCC frequency and said, "We've lost our vacuum. We're in the clouds. We're in a helluva bind...Nothing's working...My directional gyro I don't believe is working either. It went to spinning awhile ago." The controller attempted to vector him back to the Texarkana Airport, but noted the pilot was having difficulty holding a heading. At 1114, the pilot confirmed he was declaring an emergency. A 1127, another pilot who had been listening to the radio conversation suggested that he use the turn and bank indicator for attitude information. The pilot replied that it "does not work." Shortly thereafter, the pilot said, "I need to get above these clouds. I'm fixin' to get vertigo." He was cleared to 9,000 feet, and acknowledged the clearance at 1138. This was the last known radio communication with the pilot. PERSONNEL INFORMATION Flight times found in this report are based on a photostatic copy of a portion of the pilot's most recent log book. This document contained entries from December 28, 1989, to December 3, 1992, and indicated the pilot had obtained instrument competency checks on March 22, 1992, and March 11, 1993. According to the instructor who administered these checks, partial panel procedures were reviewed and the pilot's performance had been satisfactory. WRECKAGE AND IMPACT INFORMATION Wreckage was strewn through a heavily wooded area for about 1/4 mile on a magnetic heading of 005 degrees. The left wing outboard panel, with the aileron missing, was found on top of a shed in the rear of a residence. Across the street, in the back yard of another residence, was the left wing tip. The left wing inboard panel, with the landing gear still attached, was located in the woods behind this house. The inverted fuselage, with the right wing attached by aileron cables, was found about 1/4 mile away. The separated stabilator, vertical stabilizer, and rudder were in close proximity to the fuselage. The tops of several trees lay on the ground nearby. The left wing spar was twisted up and aft; the right wing spar was bent slightly down and aft. The front spar of the vertical stabilizer was bent aft. TESTS AND RESEARCH The vacuum pump, low vacuum annunciator light, vacuum gauge, and the filter from the gyroscopic artificial horizon vacuum were examined at NTSB's metallurgical laboratory. According to the metallurgist's factual report, the vacuum pump rotor was broken in three pieces "consistent with overstress breaks is castings." The vanes were intact but worn, measuring between 0.58 and 0.62 inch. The vacuum pump drive shaft coupling was fractured. On the engine (drive) side, the fracture was on helical planes "indicative of a brittle torsional separation...There was no evidence of ductility." On the pump (driven) side, there was "post fracture rotational damage...and the resulting shape was conical." The low vacuum annunciator light filament was stretched, "indicative of the filament being hot" on impact. The filter in the gyroscopic artificial horizon filter had two small tears and contained amounts of carbon, oxygen, silicon, sulfur, aluminum, potassium, calcium, and iron deposits. The vacuum gauge bore no evidence of needle indicator marks on its face. The turn coordinator was examined at AMR Combs, Denver, Colorado. Despite impact damage to the instrument, the turn needle and gyroscope were free to turn, but the slip tube glass was broken and the ball was missing. An electrical impedance test disclosed a 10,000 ohm resistance. It could not be determined if this resistance was a preexisting condition or the result of impact. ADDITIONAL INFORMATION The wreckage was released on August 7, 1993, and the turn coordinator, filter from the gyroscopic artificial horizon, and the vacuum pump, gauge, and annunciator light were released on January 26, 1994, to the John Ashford Company, Sherman, Texas.