Back          Home

In his own words…

Milt Brewer, the pilot-in-command of the stricken C-5, recounts the catastrophic engine failure and describes in detail his and the crew's actions that day.

We were a 75th MAS crew of 14 out of Travis on a routine channel mission in the Pacific.  Our Lockheed C-5A Galaxy , 90020, was a keeper.  We had flown direct Hickam, direct Clark and made one shuttle down to Diego Garcia and back prior to the mishap.  Suspicion was already present concerning the performance of our No. 3 engine.  Health checks during the 11.2 hour Hickam—Clark leg revealed abnormal vibrations on No. 3 at one point not long out of Hawaii, but we had a “Lockheed Fix” for the remainder of flight.  “Lockheed Fix” was a term we commonly used whenever the plane seemingly fixed itself for no apparent reason.  However, in view of the fact our next leg was down to Diego, I wrote it up.  A DSC-100 check revealed no problems and the next leg was uneventful.  MADAR made no vibration printouts.  But, on the return leg to Clark, vibrations came back to haunt us.  MADAR printouts occurred on a few occasions.  Health checks indicated abnormal, but not excessive vibrations on No. 3.  It was not bad enough to warrant a precautionary engine shutdown, but there was no doubt in my mind I would be making another maintenance write-up.

To our surprise the next morning, the 0245 local alert was made with the airplane in-commission.  Another DSC-l00 check had revealed no problems with our No. 3 engine.  At any rate, lacking 100% confidence in the reliability of our MADAR and not overly skeptical of the assurances provided by the DSC-100 check, we continued to press for another MAC on-time departure. 

Our mission was to Yokota via a quick turn at Cubi Point to pick up some cargo so we were to tanker 130,000 pounds of fuel out of Clark with no cargo.  However, we had a full load of 73 passengers in the back, all destined for Yokota.  This put our brake release weight right at an even 500,000 pounds.  That is lightweight for you non-C-5 types.  Also, to our advantage was the fact takeoff was in daylight VFR conditions.   Making our 0700 local scheduled takeoff time on June 7, 1982 was no problem.  Tower cleared us for takeoff on runway 20 at 0655.  Runway 20 is 10,500’ long and field elevation is 478’.

Eight long minutes would now transpire until we began evacuation of the airplane.  The tower and /or CDPIR tapes you are listening to are of these next eight minutes.  I will try to fill in the gaps.  

The tower controller's takeoff clearance had included the phrase “...change to departure control frequency….”  In the not too distant past, it was a very common procedure to be directed over to departure frequency prior to commencing takeoff roll.  Nonetheless, current worldwide procedure was to remain with tower until safely airborne.  I would have countermanded tower if it had not been for another consideration; Cope Thunder was in operation and activity was already beginning to become heavy.  

A flight of four F-15s had just arrived at Number 1, just ahead of us.  It was obvious they were trying to beat us off, but they could not complete their final checks in time.  Tower cleared us by them.  The controller's tone implied he did not need us on his frequency any longer due to the congestion, and in light of this recently changed controller procedure seemingly not rigidly being followed nor yet highly endorsed by our Standardization/Evaluation folks, we quietly consented to switch frequencies.  However, as the tower tape reveals, the controller said this phrase unconsciously because as we were completing our lineup, he directed us to taxi up another 500’ prior to commencing our takeoff roll.  Presumably, this was to avoid hindrance to the Remote Supervisory Unit (RSU) located just to our right.  But, it is a transmission we never heard, having already switched to departure control.

Precisely at "GO" speed, several things began to happen simultaneously.  On his own, my scanner still a bit skeptical of the DSC-l00 check, had set up MADAR to interrogate the No. 3 engine.  When he pushed the execute button, it was as if it was a detonator.  It was funny, but a pure coincidence he did this at the exact instant of engine disintegration.  MADAR began to retrieve data as the RPMs were decaying shortly after liftoff.  As my copilot said "GO," I hesitated on rotation for a brief moment.  I had heard the closely spaced muffled noise of the engine coming apart, but at that instant I did not yet know I was losing an engine.  I did think I was either losing an engine or perhaps blowing tires.  A yaw to the right was increasing during my moment of hesitation, but a glance at the engine instruments did not reveal any problems.  I had plenty of runway left in which to reject the takeoff, and although I was considering it, I did not feel I had enough information to go on.  In retrospect, it was perhaps best I decided to take it into the air because it gave ground crews critically needed response time to fight flames which, upon landing, rode up the pylon and were lapping above our right wing.  Additionally, as was discovered later, we were only seconds away from magnesium or titanium components igniting, which are extremely difficult to extinguish. 

A moment after rotation, my jump seat pilot noticed the first internal indication we were losing No. 3.  The low oil pressure light illuminated.  As it turns out, this was due to the oil sump being one of the first components to depart the aircraft.  This light was soon followed by rapid decay of the other engine parameters. 

I concentrated on flying the airplane.  Knowing I did not have blown tires or the complications of an outboard engine failure, I commanded the gear up.  My scanner moved quickly to the copilot’s window and gave us our first indication the engine was on fire.  The fire light in the T-Handle never illuminated until somewhere in the final turn to landing.  Presumably, this was due to the majority of the fire loop being left on the runway with other large sections of cowling.  As I directed the T-handle to be pulled and the agent to be discharged, my jump seat pilot and copilot made a race for it.  The jump pulled the T-handle and the copilot discharged the agent.  My engineer began shooting nitrogen into the area and kept it up through the landing phase.  The airplane continued to climb out well.

Tower had made another unheard transmission to us regarding a fire on our right side and approved a closed pattern.  The RSU made a clear transmission on guard frequency, which we did hear.  Closed approved or not, I was already in a left turn at about 400’ AGL.  Jump took the radios to free my copilot for checklist duties and got back in contact with tower.  In the turn to downwind, the sense of urgency to get this thing back on the ground was increasing, as you can tell from my scanner's voice. 

While still in the turn, I decided to accelerate to minimum flap retraction airspeed.  I had already told my copilot to delay discharge of the alternate fire bottle.  My priority was to get the slats up to improve the effectiveness of the engineer's nitrogen discharges.  After the flaps and slats were up, I directed the copilot to discharge the alternate fire bottle.  Cowling intact or not, I had nothing to lose.  I would have fired it regardless of my scanner's recommendation not to bother since there was nothing left aft of the translating cowl.  At any rate, my copilot had pushed the button just before the recommendation.  I know he pushed the button.  I glanced over at his hand just as he did it.  But, there was no bottle out light.  We did not try again.  Later, as it turns out, this was the only C-5 in the fleet that had the No. 3 and No. 4 engine alternate fire bottles wired backwards!  The No. 4 fire button would have been required to be punched through its T-handle in order to provide agent to the No. 3 engine.  But, not knowing this at the time, I would not have even considered it.

I leveled the airplane on downwind at about 600’ AGL.  There was no need to continue climbing to standard pattern altitude.  I felt it would further delay our return, and besides; I was comfortable at circling altitude, a familiar picture seen many times during training.  On downwind, my copilot and I got our command markers reset to the emergency return approach speed.  Tower wanted us to extend our downwind until they could pick up some of the Foreign Object Damage (FOD), but we made it clear we were still on fire and going to land.  

No time to brief, but I knew I still had checklists to run.  As I began to reconfigure the aircraft, I rea1ized cockpit distractions and the sense of urgency to get back on the ground had allowed my pattern to become a bit tight.  I started my turn to final and selected full flaps.  Initially, I planned for a normal rollout point.  MADAR later revealed I never went over 30 degrees of bank, but I already knew I was pressing this limit most of the way around.  I overshot final slightly, but once I saw the size of the engine FOD on the runway I knew I would have to land long anyway.  It was no problem getting lined up before 200’ AGL. 

My copilot and I finished the "Before Landing" and "Emergency Engine Shutdown” checklists in the final turn.  The No. 3 throttle was jammed at the position it was in during takeoff.  As I rolled out on final, the flight of four F-15s was still sitting Number 1.  I could see large sections of my engine strewn between 6,000 and 7,000 feet remaining on the right side.  They say I touched down about 5,500’ remaining.  We came to a complete stop about 1,200’ remaining.  I could have stopped quicker, but I did not want to compound the emergency with possible hot brakes. 

As we stopped, I elected to T-handle the remaining engines and set off the bailout alarm as we proceeded to evacuate the aircraft on the left side.  In retro­spect, I could have done an "Engine Fire on the Ground" checklist which might have been a bit more efficient, but my general ground emergency actions were just as effective. 

This is where the tapes end.

My aft flight deck loadmaster proceeded to the troop compartment to help get the passengers off.  The rest of my flight deck crew exited via the crew entrance door and proceeded between the No. 1 and No. 2 engines toward the escape slides to assist with the passenger evacuation.  We did not deploy the flight deck's No. 5 escape slide.  For us folks up front, it was much faster to go down the crew entrance ladder. 

Adrenaline was really pumping as we hit the ground running toward the back of the airplane.   The whole belly of the aircraft was engulfed in gray-white smoke and molten parts of the engine could be seen dripping to the ground.  Residual fuel trapped between the wing shutoff valve and the engine was still feeding the fire.  Flames from the engine were lapping above the wing, as later reported to me by one of the F-15 pilots who turned out to be an Undergraduate Pilot Training classmate of mine!  We had not run across each other since graduation nine years earlier. 

The aft slide at 3L did not deploy properly.  It came out nearly perpendicular to the fuselage and cantilevered under its own weight, suspended approximately ten feet above the runway.  Some of the taller guys had to jump up and knock it down into position. We had about a third of the passengers off by the time the fire department rolled up near the tail section.  My crew chief helped direct them into position.  Nearly everyone was off by the time the fire trucks, after three or four wild shots, zeroed in on the burning engine. 

Others of my crew helped guide the passengers off through the high grass about a 1,000’ to the left of the runway.  We had to pick a few people up as they came off the slides and also turn a few disoriented ones around that started heading toward the fire.  It was a bit chaotic.  Some were reluctant to leave their possessions behind which we were batting off the slides.  And, we had to scream at a few of them to get them started down that steep 51 degree angle.  Some of the heavier folks were overtaking the lighter ones in the chute.  The kids seemed to enjoy it and wanted to go up and do it again.  

I can remember stating in my pre-mission briefing a standard phrase to assemble upwind at 1,000’ in front of the nose.  I never briefed the same ever since.  For just crew all seated forward, this would have worked fine.  Meanwhile, as the passengers came off the left escape slides, we found it only logical to keep them going the same direction.  Although there was nothing wrong with the left side of the aircraft, the wing and engines presented a formidable barrier to be avoided.  Additionally, it became obvious a perpendicular path was the quickest way to gain distance from the burning aircraft.

My loadmasters were last off the airplane.  One carried an infant down in his arms.  We all stayed at the base of the slides until everyone was off.  Ironically, I literally came within about a step and a half of landing directly on a neurotoxic snake coiled in the high grass as we ran toward the assembly area!  I directed another head count to confirm we had everybody.  The only injuries involved about five of my passengers who suffered minor friction burns from the slides.

After we all did our toxicology testing and wrote our statements for Safety, we found ourselves legal again the next morning at 0245 local for the same mission.  As I later realized, this situation warranted declaring additional crew rest in the interest of flight safety.  Some of my crew had not quite recovered from the psychological trauma of the incident.  However, it all became academic, as our scheduled alert never occurred anyhow.  A C-5 at Altus AFB lost an engine in a very similar manner on the very same day as us.  Until HQ MAC got to the bottom of this coincidence, only the highest of priority C-5 missions would continue to move through the system for about a week.  As it turns out, the engine failures were unrelated.  The cause of our -A engine failure was due to a weakened combustion chamber which literally exploded.  Stress cracks at the 4 O’clock position developed around the igniter plugs over a period of time and resulted in the weakened chamber. 

Prior to our incident, periodic borescoping inspections of -A engines, a procedure not 100% effective, had not spotted this type of defect.  Afterwards, however, several -A engines in the fleet were identified with similar problems.  The current -C engines, with their beefed up combustion chambers have pretty much eliminated this problem today.  The 1982 dollar damage to our engine was approximately $330,000.

Our mission continued about a week later, staging another C-5 crew and their airplane.  We went direct to Yokota.  On takeoff leg out of Yokota, headed toward Travis, our hearts all skipped another beat as we heard a muffled explosion from the back.  Fortunately, it was only the air conditioning ducting separating from one of the packs.  Following unsuccessfully trying to pressurize, we air aborted back into Yokota.  The next day, we performed the final homebound leg uneventfully.

For our timely actions, we received the MAC “Excellence in Airmanship” award and the first ever, Air Force Association’s Lieutenant General William H. Tunner Award for “Best Overall Aircrew in MAC” for 1982.  Our trophy, with all our names engraved, is kept in our former squadron's archives.

We were also submitted for air medals, but for reasons unknown to us, they were not awarded.  I did not want myself put in for the Distinguished Flying Cross as I did not feel I deserved any greater recognition than the rest of my crew.  And afterwards, I would not resubmit anyone for a lesser award as I felt anything less would be an insult. 

The amazing part about the whole affair was how little direction I needed to provide my crew.  Everyone seemed to do what they were trained to do and fell in where they were needed most. 

Incidentally, this was my first mission in command as a brand new aircraft commander.

Milton M. Brewer, Major (Then a Captain)
USAF, Retired

                          
Sign Guest Book         Browse Guest Book