“How on earth did I get here?” IMC, 36 kg of fuel in the main tank before minimums and facing a lengthy period of radar vectoring to get lined up for an approach. I had planned in detail, applied Threat and Error Management (TEM) and been supported by the rest of the company along the way but I still nearly faced the dire situation of being in Instrument Metrological Conditions (IMC) with no fuel left. Let me rewind a bit.

The mission

I was notified in plenty of time that I would be moving an EC135 from Gloucestershire Airport to a private site in Glasgow at the weekend in mid-October. I would complete the task following some simulator training in the morning and then complete further simulator training completed the following day back at Staverton. My immediate thoughts ran along these lines. It would be a VFR transit due to lack of radar services at the weekend, high terrain and low temperatures at the time of year and limited endurance.

Whilst theoretically possible with a trailing wind, it would be a minimum of 2 legs to achieve the distance VFR. I had the Manchester corridor, the Lake District and Galloway to negotiate on the way up so I knew I needed to keep some extra fuel on board over the normal 5% contingency on top of trip fuel. I had convenient mid-journey company bases where I could quickly refuel. The transit plan had just written itself, along with a rough time plan (4 hrs from departure to arrival) and therefore this led to a required departure time to achieve the sortie in daylight. This allowed some travel south in a car the same day to facilitate the following days training.

The plan

Weather requirements were fairly obvious with a tail wind desirable and adequate cloud and visibility to fly VFR. IFR might be possible dependent on the altitude of the zero degree layer for the second leg, so a few minutes with a PLOG and EuroFPL produced a flight plan for that option. Applying TEM, I identified the following threats: weather, engineering delays, simulator unserviceability, fuel availability and fall back plans for training.

With the help of the operations staff, I mitigated most of these as far as possible. Operations queried whether the partial drive back following the flight was possible within FTL rules. I worked out it was possible but unpleasant so I refused the offer of a change to the roster. And so vanished my first opportunity to prevent the incident.

Getting closer to the event, I discovered the simulator sortie in the morning had been cancelled. This was good news and reinforced that my plan would actually work just fine. With a week to go the weather forecast was also looking good. Full steam ahead.

Plan for change

In the run up to the task, my benign week of simulator training took on a slightly more interesting tone. Simulator training was cancelled, it was time for some front line duty. Come in early, prepare an aircraft, deploy to a base, cover the duty and remain there for several days whilst the base aircraft was repaired. Forget the Glasgow run, you’ll still be on shift. Phew.

Except the base aircraft was repaired and I returned home after just one full 12 hr duty day. Glasgow is now back on but the aircraft needs some further preparation before it can fly. This married nicely with my need to come in late following the long shift so I was content. The later start put my timeline exactly back to where it would have been if I had done the originally planned early simulator trip on the positioning day. No problem.

Start to the day

On arriving at work, I packed my belongings in the aircraft and made my final planning preparations. The aircraft needed to be topped off with fuel and needed some last minute tweaks to its paperwork before being ready. I ordered fuel and re-assessed the timings against the weather and daylight to come up with a hard no-later than departure time. I did not want to be negotiating the forecast weather deterioration at dusk in the mountains south of Glasgow.  Remember that sentiment!

Time rolled on so I completed some unrelated work and I kept an eye on the weather. The forecast deterioration in weather was moving right in time so delaying did not seem to be a problem. I could always take the coastal route round Galloway as it was in my fuel plan, although that would use all of my discretionary fuel.

till more time passed and my fixed abort time was approaching. There was not an urgent requirement to move the aircraft so there was no issue with calling off the flight. The aircraft paperwork issue was resolved as my hard time approached but departure would be after the time.  The hard time I said I was not going to break.  A quick review of the weather and the cloud base at Glasgow was going to be at 1500 ft on arrival in accordance with the TAF. No problem, I could always recheck later and abort at the halfway point. And so vanished another opportunity to avoid the ending.

Bad decisions during fuel stop

My chosen mid-point was Manchester Barton, where fuelling is convenient and self-service. The flight up was in lovely VFR conditions with very light traffic. A quick stop at another base on the way up to collect some items left on the previous days 12 hour shift and I was at Barton in good time.

The local pilots assisted with the refuel so I was on the ground for a very short time. I reviewed the weather and reasoned that we did not need to labour the refuel to the absolute full point, leaving me about 40 kg short of full (this was of course against the sage advice of the base pilots). I had a 1 hr 20 min transit direct with a likely 30 min diversion for some cloud enroute which left a little more in hand. Another opportunity to stop lost.

The second leg started well; more solid VFR with a following wind. As Windermere approached, some wisps of clouds were tucked round some of the hills and visibility dropped a little but I still had a fairly straight run. As I got to Galloway and now talking to Scottish information, the cloud base dropped. This was pretty much as my contingency plan predicted so I switched on the terrain overlay on the EFB navigation app and Garmin GTN and started trying to try to find a way through. The Nith valley north of Dumfries looked promising so I decided to have one go at it before taking the coastal route (my spare 10 mins of fuel).

Trying to find a way through

Unsurprisingly, it didn’t work out, so I asked Scottish to let Prestwick know I would be routing along the coast. The Castle Kennedy ranges were cold so would not cause an issue. The path along the coast was not as simple as I would have liked, but some furious, rapid calculations on the EFB revealed I would reach Prestwick with only enough fuel for a direct run to Glasgow.

Prestwick was as forecast with a 1500 ft cloud base so I requested Scottish to book me fuel at Prestwick. I planned to drop in, get the fuel and finish the coastal run to the mouth of the Clyde before running along the river to Glasgow. All low ground with decent forecast cloudbases. Sound plan.

Bright idea – Bad decision

Passing Prestwick, I had a bright idea. Why not push on direct for Glasgow? I had the fuel, the forecast was good enough and if I got halfway, I could always go back to Prestwick and complete the booked refuel. Glasgow agreed to a direct arrival and I pushed on past Prestwick.

There are two valleys that breach the higher terrain south of Glasgow. The western one is aligned with the runway so I discounted that as traffic was inbound (visible to me thanks to the Garmin GTN Traffic display and also a Sky Echo ADS-B receiver showing data on EFB). Pushing up the eastern valley, my heart sank. The cloud made it impassable.

Ignoring yet another of my carefully thought-out abort options, I asked if I could try the western valley now that the other traffic had landed. Glasgow agreed and I noted I had fuel to give it one try before going to Prestwick. As locals will know, this valley is furnished with several lakes and tends to get clouded up before the one I had already tried. I could see I had just 5 miles to the heliport so worth a try.

Pull up before you are pushed down

I maintained a constant watch on my escape to Prestwick. Having successfully completed several VFR turnbacks already I was confident I could get out of any problem. Then I suddenly I knew I could not. I had insufficient visibility forward to carry on, insufficient visibility down to land or side wards to turn around. I noted 36 kg in the tanks, my position underneath the ILS beam and reducing options. Time to accept defeat and conduct an orderly abort before going inadvertent IMC. Announcing my intent and pulling up, I found the AFCS upper modes really helped. Glasgow ATC were outstanding. Calmly passing a climb to 3000 ft they waited expectantly for my intentions.

I assessed my fuel. I really did not want to go into my reserve. Hang on, there’s a phrase I am supposed to use on the R/T for that. Instead of just saying somewhat close to the right words, I spent what felt like several minutes of precious time racking my thoughts for the right words. “MINIMUM FUEL”, I blurted and ATC smoothly gave me a 360 turn to intercept the localiser at mid-way down the approach.

Worst ILS intercept ever

And then to my shame, I realised I had failed to plan for this eventuality. The ILS frequency was not set and my course bar was way off. With the upper modes completing a descent to 1600 ft, in accordance with ATC instructions, while heading mode completed the turn, I stabbed merrily at the Garmin GTN to get the ILS frequency loaded (I since learnt many ways of doing this very quickly).  Read up on my ILS article I wrote after this event to learn how an ILS really works https://rotarywinggeek.com/practical-ils-explanation-for-pilots-the-surprising-way-they-really-work/

I got sucked into trying to automate the localiser intercept with APP mode but unsurprisingly the Garmin GTN tried to go around in a wide arc to intercept at the right distance. Fortunately, mental capacity returned after a failed intercept and I reverted to manual flying. Well above the localizer, I switched to using localizer-only minima and at 1500 ft (as forecast all along), the airfield appeared. A short transit and I landed with fuel still in the main tank. The worst ILS intercept in history

I could not believe I had made so many contingency plans and then painted myself into a corner I foresaw weeks before. And I had not learned a thing of course. With plenty of adrenaline in my system, I decide to drive halfway back to Gloucestershire as planned; a final mistake which could have bitten hard had I succumbed to a post-adrenalin drop in attention and it was entirely unnecessary.

Conclusions

So, what did I learn from that? A technique I use in many of debriefs when instructing is “What went well and what can we work on next time?”

Successes

a. My planning had been sound, with mitigations of the threats in place and robust abort options.

b. The terrain overlay on the EFB and GTN were excellent in poor weather. 

c. Familiarity with the GTN was crucial to rapidly getting the cockpit setup.

d. ATC will help if asked.

e. The call “MINIMUM FUEL” works.

f. Aborting rather than trying to push on was the right option.

g. Using the Route function on EFB was excellent for on-the-fly fuel planning.

Errors and Mistakes

a. Ignoring the advice of colleagues (change the roster, take more fuel, change the plan)

b. Ignoring my own escape options.

c. Assuming that weather at the destination applied beyond the airfield boundary.

d. Failing to apply a margin to VFR minima when there is no practical IFR option

e. Failing to prepare for the IMC abort.

f. Trying to go up the level of automation when more basic modes were doing what I wanted.

g. Driving after the incident.

• Lights, helipad, action! The problem with new helicopter pad lights
• Helicopter on Fire – Could accident investigators have learned more?
• The Ultimate Medical Helicopter – Selecting the right machine for HEMS
• Aerial teamwork – Practical tips for working together by staying apart in emergency aviation
• HEMS Landing Sites – Reliable places to drop your medics