Ice You Can‘t See Out The Window
Extra awareness is needed to detect the onset of induction system icing
Some pilots may harbor the mistaken notion that if the airplane they're flying doesn't have a carburetor, they don't need to be concerned about the possibility of induction system icing. During initial pilot training, so much emphasis goes into the proper use of carburetor heat on the engines typically used in trainer aircraft that the possibility of ice blockage affecting an engine without a carburetor may be overlooked. With induction system icing, unless your aircraft is equipped with a detector, you won't have a clue that there's an issue until things actually start to go wrong. Induction system icing is a frequently recurring theme in NTSB accident reports, such as this one involving a Beech Bonanza.
A private pilot and his four passengers were killed when the single-engine Beech B36TC they were in crashed near Yellow Pine, Idaho. The personal Part 91 flight was operating in day/instrument conditions on an IFR flight plan from Baker City Airport (BKE) in Oregon, to Butte, Mont. It was VFR when the flight took off from BKE. A few minutes after takeoff, at 12:41 p.m., the pilot radioed a controller at the Salt Lake City Air Route Traffic Control Center asking to have his IFR flight plan opened. The controller radar identified the aircraft as being 13 miles south of the BKE VOR and cleared the flight direct to Butte. The controller told the pilot to climb and maintain 13,000 feet. The pilot acknowledged the clearance. Then, the controller issued a slight amendment, telling the pilot to proceed first to the Donnelly VOR in McCall, Idaho. The controller advised the pilot that the McCall Municipal Airport's (MYL) altimeter setting was 30.04. The pilot didn't acknowledge until about one minute and six seconds later. Then, the controller asked the pilot to confirm he was holding 13,000 feet and was established on victor airway 121. The pilot responded, "Negative, three six mike lima's picking up too much ice, uh, we'd like to divert to, uh, Salmon." Lemhi County Airport at Salmon, Idaho, has a 5,150x60-foot runway and is at 4,043 feet MSL.
In the next transmission, the pilot asked for a descent to 11,000 feet. The controller said he'd have to maintain 12,000 feet for terrain clearance. The controller confirmed with the pilot that he wanted to divert to Salmon, cleared the flight direct to the Salmon VOR and instructed the pilot to maintain 12,000 feet for terrain clearance.
At 1 p.m., the controller asked the pilot to confirm he was still on Victor airway 121, and the pilot responded with, "Establishing Victor 121." The controller directed the pilot to turn right 15 degrees to join the airway, and the pilot acknowledged. The controller then issued a low-altitude alert, instructing the pilot to immediately check altitude and maintain 12,000 feet.
At 1:01:04, the pilot radioed that he was having engine problems. Just 15 seconds later, the pilot radioed that he needed to "go to three uniform two immediately." That airport, Johnson Creek at Yellow Pine, Idaho (3U2), was in the vicinity of the airplane's location. It has a 3,400x150-foot turf runway. When the controller asked the pilot to verify altitude, he responded 10,000 feet and advised he couldn't climb. The controller told the pilot that McCall Airport was 24 miles behind him and asked if he wanted to divert. The pilot said he did and asked for guidance. The controller suggested a heading of 253 degrees.
At 1:02, the pilot reported that "36ML just lost its engine," and asked the controller to "say again heading for six mike lima." The controller said MYL was at six o'clock and suggested a heading of 253 degrees, adding that 3U2 was right below their position. The controller stated that he had information for the airport and asked the pilot if he'd like that information. There was no further radio contact. The controller asked the pilots of American Airlines Flight 2269 to try to raise the Bonanza on the radio. They tried several times, but there was no response. In addition, they listened for an ELT signal, but didn't hear one. The controller radioed airport information for 3U2 in the blind and asked other aircraft to try to make radio contact.
Search and rescue operations were conducted throughout the area where radar contact was lost. More than a month later, the wreckage was located by a ground search team about 1.5 miles east of the last recorded radar target.
The 51-year-old pilot held a private pilot certificate with airplane single-engine land and instrument airplane ratings. His second-class medical certificate required corrective lenses. The pilot's logbook wasn't located.
The Pilot Operating Handbook for the airplane stated, in part, "Flight in icing conditions prohibited." Because the airplane had a fuel-injected engine, there was no carburetor. The airplane did have a system for dealing with blockage of the engine air induction system. A section of the handbook headed "Induction System Icing" explains, "... the possibility of induction system icing is reduced by the non-icing characteristics of the Bonanza's fuel-injected engine and an automatic alternate induction air source. Under certain conditions, however, impact ice can form at several points in the induction system. If the air intake or filter becomes clogged with ice, a spring-loaded door in the air intake duct will open automatically, and the induction system will operate on alternate air. If the alternate air source door becomes frozen in the closed position, a pull-and-release T-handle is provided to force the door open. With a completely blocked induction filter, full power can be expected up to approximately 13,000 feet" [with alternate air open].
A review of weather information found that the freezing level was identified at 8,455 feet and the upper air soundings indicating a relative humidity greater than 80% from 8,000 to 12,000 feet, with temperatures below freezing that supported icing conditions.
At the accident airplane's cruising level of 13,000 feet, the wind was from 300 degrees at 53 knots with a temperature of minus 8.5 degrees C, with a potential for some rime icing. The icing potential increased at lower levels and reached a 77% probability of moderate clear icing at 8,500 feet. Between 12,000 and 11,000 feet, there was an approximately 80% probability of light clear icing with the relative humidity at 95% and temperatures between minus 7 degrees and minus 5 degrees C.
Satellite images from about the time of the accident showed bands of low, mid and high clouds over the area. Cloud tops extended to about 35,000 feet. Several lenticular cloud bands were also observed supporting mountain wave activity over Oregon, southern Idaho and northern Nevada, with overcast clouds over the accident site.
AIRMETs for moderate icing, turbulence and low-level wind shear were current for the time period surrounding the accident and the intended route of flight. There was no indication of the pilot having used FAA resources for a weather briefing, despite having logged onto DUATS to file his IFR flight plan.
Investigators traveled to the wreckage site, which was about 1.6 miles northwest of the Johnson Creek Airport, at an altitude of about 7,619 feet MSL. All major structural components of the airplane were located within the wreckage debris path. There was fire damage. Continuity was established from the cockpit engine controls forward to the engine.
The wreckage was recovered to a secure location for further examination. The induction system from the induction air filter aft was intact and fire damaged. Heat generated by a fire, of course, could melt impacted ice. The alternate induction air door remained attached and exhibited thermal damage. The spring was intact and remained attached to its respective mounts. The control cable was thermally damaged and remained attached to the actuator arm; however, it was separated about an inch from the attach point. The remaining portion of the cable was continuous to the cockpit area. The actuator arm functioned by hand, and the alternate induction air door opened and closed freely.
No evidence of any preexisting mechanical malfunctions was observed with the recovered airframe and engine that would have precluded normal operation.
Data from a GPS unit showed that after the flight leveled off at about 13,000 feet, its ground speeds fluctuated between 200 and 228 knots. At 12:44 p.m., the GPS showed a decrease in ground speed with the altitude at about 12,900 feet. Over the following six minutes, the recorded ground speed decayed to 165 knots. It then increased to 200 knots while the recorded altitude fluctuated between 12,800 and 13,000 feet.
At 12:53, a decrease in ground speed started, as did a descent that continued for about four minutes. Over the following two minutes, the flight climbed to 12,014 feet and showed a ground speed of 139 knots. Then, there was a descent with fluctuating ground speeds between 123 knots and 179 knots. The last recorded GPS position was about 0.19 miles west of the accident site at an altitude of 7,989 feet and a ground speed of 123 knots.
The NTSB determined that the probable cause of this accident was the pilot's continued flight into known light-to-moderate icing conditions over mountainous terrain. Contributing to the accident was the loss of engine power due to induction icing.
The NTSB didn't provide a conclusion as to whether the alternate induction air door had opened automatically, manually or not at all during the accident flight.
Peter Katz is editor and publisher of NTSB Reporter, an independent monthly update on aircraft accident investigations and other news concerning the National Transportation Safety Board. To subscribe, write to: NTSB Reporter, Subscription Dept., P.O. Box 831, White Plains, N.Y. 10602-0831.
Subscribe to Our Newsletter
Get the latest Plane & Pilot Magazine stories delivered directly to your inbox