A customer called to say his R/H generator went offline mid-flight. He happened to have an F90, but this squawk could happen with any King Air. For troubleshooting purposes, I suggested he swap the generator control units to see if the problem went to the other side, and it did. I had an exchange GCU drop-shipped to him and it was installed on the right side, putting the GCU that came off the left side back in its original position.
Everything worked fine on the ground, but going through 12,000 feet on the next flight the right generator dropped offline again. He flew to my shop so I could take a further look. We took the usual troubleshooting steps such as ringing out all the wiring between the GCU and the generator to confirm continuity. We found no obvious smoking gun, but we tweaked a few things, installed another exchange GCU and it ran great on the ground.
Frustration
On the next flight, going through 12,000 feet, the right generator dropped offline yet again. By then I was frustrated. As it turned out, this was only the beginning of a long and difficult journey that came to be known around my shop as “The Great GCU Saga.” I will spare you some of the excruciating details and most of the expletives that proliferated throughout this aggravating experience.
My partner in this battle was John, an avionics tech who worked part-time for me. Although sidelined by a bad back, his brain fired on all cylinders. He was a fantastic “sparky” – my term for avionics technicians. I swear those guys are wired differently from the rest of us, pun intended. His approach to troubleshooting meshed with mine, and between the two of us we had unraveled quite a few mysteries. But this GCU issue had us stumped, and we were not happy.
In our fight to resolve this conundrum, we went through at least five GCUs, though I think one was an out-of-box failure. We sent the starter generator out for repair and found the field was out of specs; we were hoping that would do the trick. But in the subsequent flight, after about 20-25 minutes while going through 12,000 feet, the same problem reared its ugly head.
Exasperation
We megger tested all the wires going in and out of the GCU. A megger tester will tell you if there is a short, then it’s up to you to find it. Everything checked out OK, indicating no shorts in the wiring. All systems ran great on the ground and the R/H generator was fine on the subsequent flight home. Hooray! On the next flight … you guessed it, the R/H generator went offline.
The exasperation multiplied exponentially, and the expletives followed suit accordingly. The core charge billbacks from the GCU cores were piling up. John and I scrutinized the teardown reports on those cores and found a possible clue to the problem. Two different shops squawked blown transistors at Q6 and Q7. This was like manna from heaven to John because the King Air maintenance manual includes a breakdown of the GCU. We saw that Q6 and Q7 run the line contactor, so we used a power supply on the line contactor to see if it was drawing the correct amperage – it was. We even left it going for four hours to see if it would break down from the buildup of heat – it did not. Just when we thought we were on to something, we hit another dead end. We were crushed.
Despair
At this stage, we had even exchanged the starter generator. I was beyond frantic with the extravagant parts costs mounting up for my customer. I gave no thought to what this was costing me in labor. I had no intention of charging my customer for that. Fortunately, this customer trusted me implicitly and had the patience of Job the entire time. I, however, was in agony over the exchange GCUs, the core billbacks, the generator repair and replacement and the downtime for this King Air. Add to that the costs incurred by flying this King Air back and forth to my shop. I was borderline despondent.
My wife attempted to placate me with logical discussions such as: “Troubleshooting is like peeling an onion; you can only go layer by layer, and you can’t always get to the crux of the matter on the first try.” This was of no help at all.
John and I discussed this situation ad nauseam, then soldiered on. I rented a GCU breakout box and teed it into the system. We flew the aircraft and confirmed that the GCU was operating normally. Then we wondered if we had a chafed wire problem triggered by pressurization. Going through 12,000 feet measured around 3.0 psi in the cabin.
We borescoped the wire bundles, starting at the GCU and worked our way out. We looked for something inside the pressure vessel but found nothing. Totally deflated, we labored on and removed the leading edge yet again. We examined every inch of the wire bundle with magnifying glasses. There was no evidence of arcing on any portion of the airframe. Parts of the wire bundle were encased in spiral wrap, showing no traces of arcing, but we removed it anyway. At long last we found a sign: a small black smudge on four little wires – and one of them went from the line contactor to the GCU. Eureka!
Success at last
We repaired that wire. We replaced the Q6 and Q7 transistors in the GCU for the last time and put everything back together. That R/H generator stayed faithfully online from then on. Hot diggity dog!
The saga was over, and we’d won the war. It’s a great feeling when everything finally works. My only lingering problem was understanding how replacing the transistors at Q6 and Q7 translated into $1,000 core billbacks each time I returned a GCU. When John did the job it was $50 in parts and an hour of labor – go figure that one!
This seemingly simple squawk turned into quite the wild goose chase. It’s been years since it happened, but the memory of it still boggles my mind. It was peculiar for several reasons. First, chafed wires are not the norm in a King Air. Then there was the location on the leading edge – an area where there’s not much expansion/contraction – which is a mystery. Also, why did it short out consistently when going through 12,000 feet? And why didn’t the megger testing indicate a short in the leading edge wire bundle? That wire bundle was properly tied up, had protective spiral wrap added and appeared pristine in every way. The leading edge was removed several times during this saga, and each time everything looked normal.
Keeping the craft in Beechcraft
King Airs do not typically get chafed wires. What happened with this F90 is rare for any King Air. I know this from experience. After leaving Beechcraft, I ran a corporate flight department for many years. They had King Airs when I started but later acquired Cessna Citations. Suddenly, chafed wires became an everyday problem. Those Citations were the bane of my existence. Oh, how I missed the King Airs!
When I worked for Beechcraft, I made many visits to the factory. In those days, great care was taken when the wire bundles went in, to allow for the expansion and contraction brought on by fluctuations in altitude, temperature and pressure. I remember this vividly.
Working on the Beechcraft product line has been a source of great pride for me my entire career. When owners of non-Beechcraft airplanes asked me to help them out, I used to jokingly say, “Sorry, but my alphabet stops at ‘B.’”
I hope your association with the Beechcraft King Air, whether through ownership, flying, fixing or all of the above, has given you the same sense of pride and satisfaction that it has for me.
Dean Benedict is a certified A&P, AI with 50 years of experience in King Air maintenance. He was an inaugural inductee to the King Air Hall of Fame. He owned and ran Honest Air Inc., a Beechcraft maintenance boutique with a strong following of King Airs, for 15 years. Currently, with BeechMedic LLC, Dean and his wife, Lisa, consult with owners, pilots and mechanics on King Air maintenance issues, troubleshooting and pre-buys. Dean performs expert witness work on request. He can be reached at 702-524-4378 or via email at dr.dean@beechmedic.com.