Thursday, October 29, 2009

Flying the Embraer Phenom 100

It’s been about a year and a half since my factory visit to Embraer in Brazil. Since then, the Phenom 100 has been certified, about 40 have shipped, and I’ve taken delivery of my Cessna Citation Mustang.

Today I finally got the opportunity to fly the real plane, N73DB, the current East Coast demo plane.

Since I’ve flown the Mustang 30 hrs in the last 20 days, I was pretty comfortable with the flows required for a G1000 / P&WC 600 light jet.

Henry Yandle, the Embraer rep was very accommodating as usual. I flew with two Embraer demo pilots. As a last minute addition, my friend Philip, his wife and infant joined us to make a full cabin. With these additional passengers I didn’t attempt to exercise the stick pusher, but did try most of the other maneuvers I wanted although no climb to altitude and cruise.

Pre-flight


The P100 is an easier plane to pre-flight than the Mustang as there are fewer fluid gauges to check. The Mustang requires manually checking the hydraulic pressure, the nitrogen gauges, and the fire bottle gauge. All of this is done automatically for you in the P100. One Mustang annoyance is the oil filter bypass especially on the left engine as it is hard to see without bumping your head. Embraer solved this problem by a physical extension which can be easily determined by touch.

A downside of the P100 is the gear pins that must be added and removed manually. The tow point is also non-standard.

There are three independent pitot static systems, instead of just two on the Mustang.

The door stair is significantly sturdier (and more expensive) than the Mustang. It has a much better ramp presence, although I do imagine it will be hard to maintain.

Cockpit


The Mustang cockpit is more ergonomic and less cluttered than the P100. There are more switches in more places on the P100, like the side panels, and the light panel above the windshield. There is less storage space for cups or paper.

The P100 has the G1000 in a 3 x 12” configuration. The Mustang has the G1000 in a 2 x 10”, plus 15” MFD configuration. The P100 / Prodigy uses more screen real estate with less empty space. For example CAS messages appear on the PFDs. The MFD has gear position and synoptic views which take up more real estate. Between more information and a smaller MFD, I found myself squinting more than in the Mustang.

Embraer has not yet certified Synthetic Vision and Jeppesen Charts (ChartView), both of which have been used in every one of my Mustang flights so far.

Startup and Taxi


Since the P100 has two batteries, startup is typically done by battery, not GPU. The startup is much nicer than the Mustang for several reasons. The FADEC is newer and smarter than the one currently in the Mustang. It has better hot start prevention. There is only a single switch instead of a starter switch and then moving the throttles above the detent to idle. There is a nice schematic view showing actual real-time battery and generator loads, which is filtered on the Mustang.

Several common items on every flight are better automated on the P100. For example, if the Mustang pitot static heat is operated for more than two minutes on the ground, it could damage the AoA indicator. The P100 has an automatic pitot static switch. The beacon is automatic. All in all, as with the pre-flight, a few steps are saved every operation compared to the Mustang.

Taxiing is easy although not quite as easy as the Mustang. The forward view is better although it isn’t as easy to see either wing tip from either pilot location. It takes more power to start rolling, but the rest of taxiing is at ground idle.

The electric brakes were a bit of a disappointment. They are “grabby” with a short travel length. The brakes feel more like a binary on/off switch than a smooth pedal motion. While it is possible to use differential braking and power for a tight turn, the turn radius is not as tight as the Mustang.

As with the Mustang, V speeds must be calculated manually or looked up in a book. With six souls on board, and moderate fuel we were over 10,000 lbs so V1 and Vr were about 10-15 kts more than the Mustang.

Takeoff, Climb and Maneuvers


We departed KBED in rain, ceilings to near minimums and light cross-wind gusts. The plane felt solid and a bit stiffer than the Mustang. The workload was near identical: gear up, autopilot on, reduce from T/O thrust and hope ATC clears you up fast enough. We quickly broke out at 6,000’ and asked for a block of airspace for some airwork.

Northwest of the CON VOR at 8,000’, we did steep turns and slow flight. Since the demo schedule was a bit behind and I would not be immediately trading my Mustang for a P100, we didn’t have as long as I would have liked in the P100.

Steep turns were more challenging than the Mustang but not especially difficult. It was tricky to find just the right pitch until well until into the second turn. In slow flight, the plane was quite stable and responsive, even with 30 degrees of bank. The power needed some adjustments as it was quick to drop below the “donut”.

All in all, the plane was solid, sturdy and a little heavier than the Mustang. The Mustang feels a little more responsive, especially in turns.

As I mentioned earlier, with a very young passenger in the back, I did not attempt the stick pusher and imminent stall. Doing full gentle stalls in the Mustang, without a pusher or shaker, was a major selling point to a low time jet pilot like me.

I tried out the silver painted boots. As you may recall, Embraer painted the boots silver so it would give a “hot wing” look, even though these are really boots not bleed air. Since silver does not contrast as well with ice, a certification engineer had them put this funny looking black circle near the wing tip in order to see icing more easily. The whole thing is silly – the boots should be all black.

Return and approach


ATC gracefully gave us vectors back to KBED for the ILS 11 approach to near minimums. Slowing the plane takes a bit longer than the Mustang. There is no current option for speed brakes. Gear deployment speed is lower so they cannot be easily extended as drag either (Mustang gear can be extended at Vmo). ATC asked to slow first to 190 kts, and then again to 170 kts for spacing and each time required reducing power to idle and then waiting, whereas in the Mustang more options are available.

Adding drag is not purely academic. In the case of an emergency descent, it’s very easy to add drag to the Mustang via gear and speed brakes and get down in a hurry, at more than 8,000’/min. It would be very interesting to see how quickly the P100 can descend in an emergency. Also many large airports require keeping the speed up as late as possible. It would be equally interesting to see how much distance is required to slow from say 160 kt to 100 kt Vref in both aircraft.

When we received our final vector and were cleared for the approach, the P100 in combination with the G1000 and GFC 700 autopilot, did a terrific job to stay stabilized while we were bounced around in shifting winds and light turbulence. We stayed at flaps 2 as flaps 3 and full are not yet available due to a software issue. Apparently, when this software issue is resolved, Embraer recommends adding flaps full after the Final Approach Fix, which is very unusual. Typically, a jet is fully configured for landing at the FAF and then stable all the way down to the Missed Approach Point. When I disconnected the autopilot 100’ above minimums, there was a bit of jumping around but things stabilized quickly.

There is very little flair in the P100 – it lands even flatter than the Mustang. Again, there were no speed brakes to slow us down. With the grabby brakes, I agreed with the demo pilot’s suggestion to hold off on the brakes and roll out using more runway. When I did finally use the brakes, we were too close to the last taxiway (G) and ended up using the full runway length of 7,000’ even though I was right on Vref over the numbers. While we could have braked more aggressively, there’s no doubt the P100 soaks up more runway than the Mustang.

Taxiing back was easy. On the crowded and chaotic Jet Aviation ramp, we were able to fully utilize the minimum turn radius of the P100.

Conclusion


The P100 is a solid competitor to the Mustang. It has better automation and synoptic views making life a little easier for the pilot. It’s faster, harder to slow and less forgiving than the Mustang so a more experienced pilot or crew of two is a better fit. The cabin is better appointed and comfortable than the Mustang, especially with the door stair and rear lav. If you pay more, you get more.

I was not able to evaluate many other aspects of the P100 such as high altitude climb performance, fuel burn in cruise, emergency descent and many other critical aspects, so I feel this is an incomplete picture of a fine aircraft.

Overall, I’m happy with the Mustang for business and personal use and feel that it was the appropriate choice for a first time jet pilot. If I were more experienced, or had paying passengers, and wanted a little extra luxury, the P100 would have been the more appropriate choice.

Mustang Delivery Experience

I picked up my much anticipated Mustang in Wichita, KS on October 5-8, 2009. This was a demo plane that was fully paid for (financially delivered) in March 2009. So for six months, Cessna has had my plane and my money, while paying me a modest monthly fee. Needless to say, I was eager to get back one or the other.

Since I had never taken delivery of a jet before, and was unsure of wear and tear from demo usage, I hired Cyrus Sigari of JetAviva to help with the acceptance. JetAviva is primarily a broker for light jets including Mustang, CJs, and Embraer Phenoms. Cyrus previously worked as and engineer and has a huge amount of experience with all of these light jets, though had never accepted a demo plane prior. We reviewed his very detailed pre-acceptance plan over breakfast that morning and added a few items (0 g maneuver, satphone, etc.)

Most deliveries are performed at the union-free factory in Independence, KS. Cyrus suggested that we take delivery there rather than at the Service Center in Wichita, KS because of higher likelihood of parts availability. Cessna wanted the plane to stay in Wichita where the post-demo refurbishment was done. Given what happened over the next few days, Cessna should have followed Cyrus’s suggestion.

The morning at Cessna started off well enough. Cessna had some kind gifts, including a scale model of the real plane with accurate colors and tail number, leather jacket, etc. This was a very nice gesture but I was ready to get to business with the plane inspection and test flight.

The plane was in immaculate condition. It was so clean, even deep into the wheel wells, that you would never know it had about 300 hours and 400 landings. The appearance was truly factory new.

There were only a few minor squawks, such the torque seals on the gears and tension on the outer door latch. The headliner had a tiny scratch in it and the pilot visor was a little loose. We were ready to go on the test flight.

Cyrus was in the pilot seat. A Cessna test pilot was in the right seat, while I was in back. The test flight proceeded smoothly and again only minor items were found, such as the standby attitude indicator was 3 degrees off pitch compared to the PFDs. Cyrus purposefully does a few 60 degree bank 360 degree turns to load up 2g of force, as his experience has shown that this will shake loose some marginal items. For the same reason, the zero g push over was interesting. In the cabin, we kept everything strapped down while the two tray tables started to levitate.

After the flight, Cyrus and I examined the plane literally from bottom to top, using a creeper to check for any sign of leaks or deformation.

When we returned, we asked for two additional inspections from Cessna. We wanted a full FADEC download history. The FADEC download showed that one engine had had an ITT exceedance soon after manufacture prior to entry in demo service, but was still within tolerance (less than 5 seconds between 830 and 865 dC – it was only for 1 second). Also, since Cessna themselves had not followed P&WC conservative compressor wash schedule during the demo period, I wanted to borescope both engines to check for corrosion.

The borescope turned out to be a huge problem in terms of logistics. Even with all of the Cessna Wichita resources, they could not find a decent borescope as the lone good one was in repair. P&WC’s local representative had a good borescope but insisted that I sign an onerous agreement disavowing them of any responsibility whatsoever. Since this agreement was only between me and P&WC and did not involve any responsibility on Cessna’s part, I refused to sign as Cessna had operational control during the demo period.

The borescope easily cost us ½ day of wasted effort. By the time extensive discussions and meetings occurred, the good Cessna borescope was repaired and we no longer had to wait for P&WC. Fortunately the borescope showed no detectable corrosion but that’s not to say there isn’t any lurking.

Neil Singer, my mentor pilot arrived from Boston and joined us in the waiting game. Neil and I took our Mustang initial training together at Flight Safety Orlando a few months ago.

With the borescope out of the way, we planned for an afternoon departure to Los Angeles to fly Cyrus back home. Taking into account potential fatigue, we set a 4pm wheels up cut off time. Cessna went off to do a final engine run up and check.

The final run up failed. The FADEC has three redundant inputs to determine whether the plane is on the ground or not. One of the three did not agree. After various continuity checks and board swapping, Cessna determined, with P&WC disagreement, that the FADEC was bad. FADECs are highly reliable pieces of equipment and fail very rarely. Consequently, they are not a stock item, even in a large Service Center like Wichita. Had we been in Independence, another would have been easily available at least to swap in and try. So now a FADEC had to be ordered from Michigan and would not arrive until 1pm the following day (Wednesday). This problem caused us to lose yet another day.

Cyrus had to return to Los Angeles, so he flew back commercially on Wednesday. Neil and I waited until the FADEC arrived Wednesday afternoon. Of course when it was swapped out, the problem did not go away. So Cessna spent the evening swapping out boards until a combination of new fuel controllers resolved the issue. Around 11pm Wednesday night we received notification that all the issues were resolved and we would be good to go the next day.

As a precaution, the subsequent overnight shift also re-tested the plane and did additional run ups. They gave their blessing as well.

We planned for a departure Thursday morning. Unfortunately, the weather did not cooperate as there were low ceilings, thunderstorms, and icing up to the top of the clouds around 25,000’.

We started up and did as many ground checks as we could do, while waiting for a final lightning cell to pass over the airport. The engine run up was fine. The FADECs worked well. Everything in plane seemed perfect. With the Mustang RADAR scanning into the clouds, the passage looked clear, so we took off.

Climbing steadily through 15,000’ I shook Neil’s hand and congratulated ourselves on our successful departure out of Wichita, a day and a half late. Unfortunately our satisfaction was short lived.

Climbing through 18,000’, a CABIN DOOR message appeared on the Crew Alerting System. This was just like the training in the simulators at Flight Safety - low ceilings, a new CAS message, and shooting approaches. The checklist action is “Land as soon as possible” so we diverted back to Wichita and our favorite Service Center.

Soon afterwards, another CAS message appeared that both PFDs were on Air Data Computer 2. Though the checklist suggested simply switching back, we did not want to risk losing both ADCs in the clouds.

Air Traffic Control asked we wanted to declare an emergency but this was not necessary, especially as we descended quickly and pressurization was holding well.

The Service Center got right on the problem. When they adjusted the stiff door on Monday, they had put the door out of specification, so the CAS message was valid. It had only manifested at high altitude not during any of the ground checks. The ADC problem was a known G1000 transient issue.

After another set of ground checks, we departed once again. This time, the plane worked flawlessly and a 140 kt tail wind carried us home non-stop in 3:15.

Since then, the plane has been a flawless joy and wonder. It’s been an amazing machine during subsequent trips to Montreal, Teterboro, Las Vegas, and Orlando. I’ve flown it over 30 hours in just this first month.


Thanks to Bill and Charlotte, the Cessna customer service reps who gave their personal best and tried whatever they could to resolve these issues as quickly as possible.