SEA04FA022 · Beech S35

At 0347, the pilot received a standard weather briefing, which indicated instrument flight rules would prevail along the entire route of flight. The pilot was advised of an advisory for mountains to be obscured, an Airmet for occasional moderate turbulence below 15,000 feet msl, icing, occasional moderate rime or mixed in clouds and precipitation, with a freezing level to 16,000 feet msl. At 0540, radar contact was established with the flight. At 0635, while at 11,000 feet msl the pilot requested a higher altitude, and at 0639 the pilot reported that he was at 13,000 feet msl and clear of rime ice. At a time which could not be determined the pilot advised air traffic control that he had just lost his suction gage, and shortly thereafter radioed, "Mayday, Mayday, Mayday." At 0650:52, the airplane was on a south-southwesterly heading at 13,100 feet msl, and at 0650:59 it was at 12,900 feet msl on a southeasterly heading. From 0651:04 until 0651:28 the airplane was between 12,800 feet msl and 13,100 feet msl turning left from a south-southeasterly heading to an east-southeasterly heading. Between 0651:28 and 0651:52 the airplane descended on a northeasterly heading to an altitute of 10,700 feet. The airplane then made a right turn of approximately 60 degrees to a south-southeastly heading, and seven seconds later at 0651:59 was observed at 10,500 feet msl, indicating its descent had been arrested. Six seconds later, at 0652:05, the flight had descended to an altitude of 6,800 feet msl on the same heading. The last radar hit was at 0652:17 when the airplane was on a northwest heading at 6,400 feet msl. The engine and fuselage were located 1 mile northwest of Warren, Oregon, while both wings, the tail section and the left stabilizer were located approximately 1 mile northwest of the main wreckage. Separation signatures were consistent with the left stabilizer separating in an upward and aft direction, while both wings had separated in an upward and positive direction. Examination of the vacuum pump and its history indicates that it is probable that the pump failed due to vane failure as a result of highly worn vanes being subjected to impact bending loads from the uneven interior surface of the housing, with a possible contribution from contaminating material between the vanes and the housing.

Verbatim NTSB analysis and probable cause from the NTSB dataset