Sunday, November 13, 2016

IFR Operations: Two-way radio communications failure.

That's AIM 6-4-1 (which quotes FAR 91.185 and adds).  As I was mulling over the consequences and procedures for electrical failure, I wondered how I would get the airplane back on the ground at KAUS, which is a Class C airspace that requires two-way communications.  

If the conditions are VFR, you maintain VFR and land as soon as is practicable.  This allows the pilot discretion beyond "as soon as possible" to select a suitable airport or continue to the intended destination if it's close.  The idea in general, though, is that if ATC can't contact you and/or you can't respond, you're the jerk screwing up the airspace for everyone else.

To land VFR at a non-towered field is a simple matter of good judgment.  Fly over the field to assess winds and traffic, then fit yourself into the traffic safely and land.

If, however, the conditions are IFR, it's a little more complicated.  Assume it's all clouds.  What do you do?  First let's think about just radio failure, not full-on electrical failure (so you still have your transponder, which it says to set to Mode A/3 squawking 7600; does Mode A/3 broadcast altitude?).  The AIM section says:

  • Route
    • route assigned in last clearance
    • if being vectored, go direct to the next fix/route/airway in the clearance
    • route ATC said to expect
    • route filed in flight plan
  • Altitude - highest for current route segment
    • altitude assigned in last clearance
    • minimum altitude for IFR operations
    • altitude ATC said to expect
The ideas here are to be predictable to ATC.  There is likely to be confusion when you don't acknowledge or heed instructions (note to self: this is one reason it's so important for me to respond right away even if I have to ask to repeat), but if you move along in a predictable fashion, they can plan around you.  Again, if you get to VFR conditions, land.

What if the failure happens before your approach but it makes sense to try the approach?  First, this seems pretty risky to me since you have no way of knowing whether the airspace is clear.  Second, however, you've been acting very predictably so perhaps you can go for it.  The AIM instructs to start the descent/approach as near as possible to the expect further clearance time (if received) or the expected arrival time, keeping you predictable.

Assuming it's just a comm radio problem, consider using the nav radio to contact FSS on a VOR that can receive.  AIM 6.4.2 talks briefly about having them relay situation reports and clearances.

FSS gives in-flight weather, too.  Imagine being in IMC with no radios, headed straight into some nasty weather?  (And the Stratus/SXAR1 ran out of battery. Or the iPad died.)  Egad, man.

Suppose we were to lose comm while still inside KAUS airspace.  I believe we would have to maintain the last clearance (which generally is sending us out of the Class C) and stay out.  There are procedures for rocking wings and flashing lights to communicate with Class D towers, and they use a light gun where certain colors/patterns mean different things (solid green means cleared to land, flashing red means unsafe airport, etc).  If we were outside the airspace, we couldn't come back in:  Class C airspace requires two-way communications and Mode C transponder.  (I would think losing transponder in their airspace but maintaining communications could be dealt with to allow us to land.)

So, that's when radios fail when flying IFR in IMC or VMC, but ATC can still see you on the radar.  What about full electrical failure?  No comm and no transponder, so ATC doesn't know you exist, although hopefully they're concerned that you disappeared.  I'm getting sweaty palms just thinking about it.

I'll take a guess before I look it up.  If it's VMC, land.  If it's IMC, get to VMC safely ASAP.  So where is VMC, and how do I get there safely?  Having briefed the route, we should have a good idea of what the weather systems are like along the route (and ForeFlight + external device(s) is indispensable in a situation like this).  Turn back?  Proceed with a deviation?  Could descending to an MEA (or even a MOCA?) or climbing to an altitude (VFR, I assume) appropriate for our heading get us to VFR?  Do any of these options have a high likelihood of success with a low chance of compromising separation?  Do we pick areas away from airways or routes between waypoints (harder with GPS waypoints all over the place)?

In IMC, this is all risky business; who knows who else is out there!  Intermission now for dinner...

Saturday, November 12, 2016

California flying, day 2

(I started this post Tuesday evening and have been busy working, flying and mothering so it hasn't been finished until now!  Perhaps tomorrow I'll get to the afternoon flight...)  

We got started bright and early today but again had no actual IFR to play in.  The missions for the day went ahead as planned anyway.  I filed KSQL OSI SAPID SANTY for the LOC RWY 2 approach into KWVI (Watsonville).

Upon startup, the display for COM1 was blank except for a flickering vertical line separating the com from nav frequencies.  Hello, N24AF, anyone?  :)  Brightness controls altered the glow around the knobs but didn't change the critical display.  Combined with yesterday's flakiness on the #1 CDI (fed by NAV1), we were suspicious but had other troubleshooting ideas as well.  Jason thought it was potentially a sunlight-on-sensors issue (it's auto-dimming), and the sun was lighting us up from the back in the parking area.  By the time we taxied most of the way to the run-up area (into the sun), the display was behaving properly.  There was still some wariness about the CDI, but we were willing to give it another chance today to see how it went.  (Spoiler: it was not as flaky, but never agreed with #2 CDI, which was still solid.  Was within the 4-degree tolerance, but, you know, you'd like to have complete confidence in the instrument.)

CIGARS.  Lights, camera, action.  Pre-takeoff briefing (verify runway, last check of wind, takeoff abort conditions, emergency procedures before and after rotation).  We talked on each flight about how we should be starting a timer when cleared to depart and checking in after that 5 minutes if the expected higher altitude hasn't been cleared.  In the bay area, in particular, there are plenty of nearby MEAs above 5000', so in actual IMC it would be pretty important to follow up with ATC.  We forgot on every flight to set a timer, and Jason related how even when he starts a timer, there's enough going on as the flight starts that it's usually way later when he notices how far past 5 minutes the timer has gone.  Countdown timers with audible or haptic alerts would be best for reminders like this.  Had I not forgotten my Apple watch charger, the watch-based ForeFlight timers may have been able to fulfill this role.  I think I'll try to incorporate starting the timer as part of the pre-takeoff briefing as we roll onto the runway, so at least step 1 has a deterministic trigger point.

Clearance:

We followed the departure clearance and were soon vectored southeast then cut loose to resume own navigation.  I was quite busy aviating and navigating and didn't get ahead of the plane by the time we were at Woodside VOR ~6nm away, so I wasn't ready with my 5Ts.  My living autopilot took over in the right seat while I went into executive mode, assessing and directing the current phase and planning the next phase (verify SAPID with GPS waypoint and SJC radial).  Here's an example of what the 5Ts look like for SAPID from OSI:

Turn: none
Time: none
Twist: none (already twisted #2 CDI for SJC R-210, verify)
Throttle: none
Talk: none

With that 30 seconds of planning, I had caught up with the plane and was now six minutes ahead!  That was really pretty sweet.  I had lots of opportunities to do configuration change flow/checks (mag/DG sync, flight instruments, engine instruments, breakers/switches, throttle/mixture, fuel controls; checklist).  By getting ahead, more time is available for the management of the current phase of flight and being prepared.  An autopilot is a magnificent tool for facilitating this.

Next was SANTY (6+ minutes ahead once we passed SAPID).  SANTY is an IAF for the LOC RWY 2 into KWVI, so the workload would start increasing there.  Here's what SANTY's 5Ts look like; you can see that there's more to work on:

Turn:  left to track the SNS R-293
Time:  none
Twist:  #2 CDI for identifying SNS R-293; #1 CDI for the I-AYN localizer; these courses intersect at NALLS, the IF for turning outbound for the procedure turn and also the FAF inbound.
Throttle: none
Talk: none

Before getting to SANTY it was also time to start the 5As:
ATIS: tuned and received.
Altimeter:  Make note of it from ATIS, but don't enter into the altimeter until cleared for the approach.
Approach briefing:  this is the attention-consuming part.  Pull up the plate (or paper), check that it's the right approach at the right airport.  Read everything on the top, out loud is best.  If it's appropriate to load any frequencies at this time, go ahead and do so (see the 5Ts above).  Read restrictions.  Memorize the first part of the missed approach (climbing right turn to 5000).  Orient on the plan view, visualize how you'll fly it and how the instruments will guide you.  Check the profile view for altitude changes and where.  Get the BOTTOM LINE:  How low?  How far?  What's next?  700', at the MDA or 4:12 from NALLS, climbing right turn to 5000'.  That last bit is the missed approach trigger.  If you get down to the minimum descent altitude of 700' or go 4:12 (at 90 kts for the 172) past the FAF and can't see the runway or aren't oriented to use normal maneuvers to land, execute the missed approach by making a climbing right turn to 5000' (after which time you'll head southeast to the SNS VOR for a hold, but first is just to get up and clear of the mountainside).  The FAA has cleared the area and guarantees obstacle and terrain clearance if you follow the instructions, so plan to do it.
Avionics: This all ties together, right?  Redundancy built in, lots of chances to double-check yourself and catch any tuning errors.  Gather the avionics items from the approach plate, either tune it now or note it for the 5Ts for the upcoming trigger.
Airplane:  Flow/checks.  Descent checklists.  DG sync.  GUMPS.

With the LOC RWY 2 approach, you turn outbound at NALLS and perform a procedure turn within ten miles of NALLS.  90 kts is 1.5 nm per minute, so flying outbound for two minutes, a 45-degree right turn for one minute, and a left standard-rate turn is well within the 10 nm requirement.  This was done, and then it was a matter of tracking the localizer inbound, maintaining airspeed, and reducing altitude at the appropriate times.  2200' at NALLS, the FAF, 6.3 nm from the runway threshold, down to 700'.

I forgot to start a timer.  In practical terms, the GPS was also telling me where the runway threshold was, but it's sloppy to not utilize the correct methods for a non-precision approach.  And here's why the timer is important:  Imagine you're in actual IMC and pass NALLS.  You can safely descend to the MDA of 700'.  At some point, however, 700' is no longer safe and you need to execute a missed approach if you can't see the airport environment.  You can only identify that point by some kind of nav (GPS fix, a cross-radial from a VOR, etc) or by time since the FAF.  Being forward-acting pilots, our 5Ts and 5As have ensured that we're situated to identify that point, but what if....  What if the GPS RAIM check fails?  What if the VOR turns out to be unusable below 1000'?  There are enough reasons to have that timer going.

We executed the missed approach just before the airport and noted that they were using 20 for landing this calm day.  Climbing right turn to 5000' and heading to SNS VOR.  Once established in the climb (with flow/checks complete), a quick consult of the plate filled in the details for the rest of the missed approach procedure.  Sometime in there we had also switched to CTAF for WVI, so needed to communicate with NorCal approach again for holding instructions at SNS.  We would be using the published hold, so the controller just told us when to expect further clearance, the current time, and to advise if we wanted to leave the hold earlier (since we were on a training flight in VMC).

I looked at the plate for inbound holding course, mentally mapped it on the DG, and made my best guess for how we'd enter the pattern -- direct, parallel, or teardrop.  Then I used the thumb tool as a redundant check and found that my guess was wrong!  (Psst, the system works.)  It was obvious that the 5Ts would be happening frequently and in quick succession in the hold, so I started working it out.  The goal is to have the inbound leg be 1 minute long, so depending on winds, the outbound leg and turns may be longer/shorter and abnormally shaped.  Note to self: review the airspace extents; my recollection is 4 nm on the inbound (non-holding?) side, 8 nm on the outbound (holding?) side.

The first outbound leg was one minute, and the resulting inbound leg was well over 2 minutes!  The T for time would need adjusting on the next outbound leg.  Just when does that outbound leg start, anyway?  The holding fix is at the end of the inbound leg, then there's a standard rate turn, at the end of which you should be on the reciprocal course.  The outbound timer starts when you are abeam the fix; in this case, when the CDI flipped from "to" to "from."

The second outbound leg was only 25 seconds, and the turn to the inbound leg was finished before intercepting the radial.  In this case, the inbound time starts when wings are level or when the course is intercepted, whichever comes first.  Wings level, start the timer.  The inbound course is the resting time.  I was flying the hold, using the heading-only autopilot as much as possible, and trying to figure out when I'd have time to brief the RNAV approach back into KWVI.  The outbound legs were getting shorter, and the inbound leg was shortened by going through the 5Ts each time -- necessary to be prepared.

The third outbound leg was more like 15 seconds, and resulted in an inbound leg of close to one minute.  I started getting through the approach briefing, and by the time we were on the fourth circuit and the controller asked if we wanted to wait, I was prepared to ask for a clearance to the RNAV (GPS) RWY 2 approach back into KWVI.  This is what that set of holds looked like, though it felt like we were doing some kind of spirograph design.  The pattern was getting tighter as I shortened the outbound leg to standardize the inbound leg.



We exited the hold on a heading of 260, IIRC, being vectored to the IAF waypoint RISPE.  While enroute, more flow/checks and 5Ts to get ready for the approach.  As we headed out over the Pacific Ocean, it was clear that we would be north of RISPE and got a DTK from the GPS to take us there.  The controller noticed the change and asked us what we were doing.  My memory here is fuzzy, but I believe we had been cleared for the approach.  If not, surely at least Jason and I would have had a conversation about remaining on the vector even if it was off-course v. pinging the controller.  He did not object, however, when we told him we were correcting course to go to RISPE.  Perhaps because he knew we were training and had already proven to be a nice controller :)

These approaches are pretty simple.  They'd be even simpler if this aircraft had a WAAS-enabled GPS, which would provide vertical guidance as well as horizontal and allow use of a glidepath down to lower (LPV) minimums.  Without WAAS, however, we'd be managing our own altitude in accordance with the LNAV row.  NBD (LOL, unnecessary acronym -- no big deal), we'd fly it just like the localizer, but recognize waypoints and distances with the GPS instead.

The LPV DA is 458'.  This means that as you're descending on the glidepath, when that path intersects an altitude of 458', that's the MAP.  At that point, you're 288' above the touch down zone elevation (TDZE).  This DA happens short of the runway, allowing time to see the airport environment and runway lighting and use normal visual maneuvers to complete the descent.

The LOC MDA is 700' and the LNAV MDA is 740'.  But where?  It seems reasonable that the LOC MDA is closer to the runway threshold since it's lower, but.....  The only things I know for sure are that (1) the LOC approach specifies 4:12 after crossing NALLS as a (somewhat) definitive marker for when to execute the missed; we'll forego the imprecision for now; and (2) the LNAV has no time, but the rule is that the MAP is the last named waypoint; in this case, RW02, the threshold of runway 02.  At RW02, however, you still have 588' to descend!  Note this MAP explanation for future reference.

The other little complication to this approach was that Jason wasn't confident the non-WAAS GPS would properly manage the linear scaling.  When using the LNAV portion, the linear scaling can be 5 miles when enroute (beyond 30 nm), 1 mile in the terminal area (30 nm to the FAF), and 0.3 miles on approach (from FAF to runway).  This is to increase the course deviation sensitivity, ensuring your best chance at alignment with the runway in the clouds as you approach the airport.  Ideally, the GPS is smart enough to do it for you when you cross 30nm and the FAF, but Jason didn't trust it so he manually switched it.  People fail checkrides for that little detail, and it is an important detail.



As we neared the runway, we opted to convert to more of a circling approach since 20 was in use and it was almost lunchtime.  On final for 02, we turned right to offset from the runway and enter a downwind for 20.  Landing, taxi, and time for lunch and a debrief!  (BTW, the restaurant at Watsonville is a tasty Italian place, highly recommend.  :) )



That final "final" check

As we sat in the run-up area before departing Watsonville (KWVI), I called NorCal on the radio to get the IFR clearance over to KSNS for the ILS approach.  KWVI is a non-towered field, meaning no one controls the airspace and that pilots use the common traffic advisory frequency (CTAF) to announce positions and intentions.  We're all working together in that environment.  It's crowd-sourced tower ops.

It's all great, with one caveat:  no radio is required in this airspace.  So that means that not every plane and pilot may be participating.  This makes it über-important to use those VFR traffic scanning skills and be responsible for your own safety and separation.

This was brought into very clear focus after we received the clearance, checked for traffic, announced our intentions and rolled across the ILS threshold believing everything was safe and that we had the right-of-way.  Luckily the run-up area has an odd shape, and as I glanced up the final approach before entering the runway, sure enough, there was a little Cub descending to land!  Brakes!  We stopped short of the runway threshold and watched him land.  He was probably cursing us; I hope he would have been able to perform a go-around had we taken the runway.

Let's look for explanations.

First and most obvious, is that by a stroke of (un)luck, when we visually checked the pattern for traffic, that plane must have been behind the high right wing of our Cessna.  We couldn't see him.

Second, we verified that we were talking and listening on the appropriate CTAF, not still accidentally tuned to NorCal or another frequency by mistake.  Confirmed.  If the other pilot was radio-equipped, he was not talking on this frequency.  He was wearing the unmistakeable mint-green David Clarks, so he at least was equipped for comm, whether his plane was or not.

Jason relayed a conversation that he had had with an "old timer" who always incorporated a 360-degree turn at the end of his run-up before taking the runway.  It seemed like overkill, but such an operation would have revealed the other traffic that surprised us.

Thankfully it was not even a near accident.  Just a learning experience.

Friday, November 11, 2016

Homework: Ammeter discharge, ALT FLD breaker out

From the 172R POH, Section 3: Emergency Procedures, Electrical Power Supply System Malfunctions:

-----------------------------------------------------
INSUFFICIENT RATE OF CHARGE 
NOTE The low voltage annunciator (VOLTS) may come on and ammeter discharge indications may occur during low RPM conditions with an electrical load on the system, such as during a low RPM taxi. Under these conditions, the annuciator will go off at higher RPM. 

If the overvoltage sensor should shut down the alternator and trip the alternator circuit breaker (ALT FLD), or if the alternator output is low, a discharge rate will be shown on the ammeter followed by illumination of the low voltage annunciator (VOLTS). Since this may be a "nuisance" trip out, an attempt should be made to reactivate the alternator system. To reactivate, set the avionics master switch to the OFF position, check that the alternator circuit breaker (ALT FLD) is in, then set both sides of the master switch to the OFF position and then to the ON position. If the problem no longer exists, normal alternator charging will resume and the low voltage annunciator (VOLTS) will go off. The avionics master switch may then be returned to the ON position. If the annunciator illuminates again, a malfunction is confirmed. In this event, the flight should be terminated and/or the current drain on the battery minimized because the battery can supply the electrical system for only a limited period of time. Battery power must be conserved for later operation of the wing flaps and, if the emergency occurs at night, for possible use of the landing lights during landing.
-----------------------------------------------------

During our Thursday morning flight, the low voltage annunciator did not illuminate, although the entire annunciator inset flashed once before the ammeter started reading low.  The discharge was not present just before takeoff, and takeoff and climb were performed at full throttle.  I don't think the "NOTE" applies.

The ammeter was showing a discharge.  The ALT FLD breaker was tripped.  No failures or abnormalities were observed on any instruments or avionics using electrical power.  I did not think to check the voltage on the voltage/OAT/timer instrument.

The mentions of discharging ammeter and tripped ALT FLD breaker in the POH all indicate that the Low Voltage annunciator should be on, flashing at first, then solid on until the problem is resolved, but it wasn't.  The annunciator came on during preflight test, so the annunciator itself is capable of lighting.  

How I wish I had had the POH on my iPad!  A quick search would have brought me to the above section and proper procedure for dealing with what in all likelihood is a "nuisance" trip out.  The proper procedure is basically to shut it all down and start from scratch:

1. Kill the avionics (the big power users).

2. Reset the breaker.

3. Kill both battery and alternator with the master switches.  
---At this point, the circuits are open and electricity is not being delivered.  

4. Turn the master (both sides) back on to connect battery and alternator.
--- Incrementally bring things back to life.  Check for neither charge nor discharge and verify no annunciators and the breaker is in.  At this point, what is receiving power?  The turn coordinator, the voltage/OAT/timer, panel lighting according to knob setting, beacon/strobe/etc as per the switches, flaps when actuated, fuel pump when activated, and that's all I can think of.

5. Turn on the avionics master.
--- It could be a good idea to start with some of the items off.  For Class C flying at KAUS, transponder and radio are required.  If being conservative, I might turn off the GPS and COM/NAV2 and leave the autopilot off.  (Note to self: investigate how to get back to KAUS without radio or transponder.  An IFR clearance gets you your cleared route even in the event of lost comms.  At this point in my training, I'd land at Smithville or Lockhart and call home for advice!)

This forum has some interesting information about the circuits (as well as a lot of hot air and attitude, sorry).  And discussion of whether resetting a breaker in flight is a good idea.  Resetting it repeatedly is obviously ill-advised, but the POH itself indicates this particular issue can be a nuisance rather than an emergency.

Takeaways:
- Be more thoughtful before rationalizing an abnormal indication.
- Have the POH in searchable form!!!!!
- Continue the flow/checks that identified the abnormal indication so promptly.

As for the fog in the turn coordinator, there are a number of reports of this in forums across the web and the recommendations are (1) do nothing, it'll be fine when the humidity goes down, or (2) moisture in an electrical instrument is not good, so repair or replace.  It's not my plane, so I have to make a determination of whether I feel comfortable as PIC to continue using it.  For now I think I will continue to fly 2MA, but will keep an eye on it.

Crosswinds, foggled unusual attitudes and stalls, mild in-flight failure

Yesterday morning Mark and I had an early flight.  In the interest of getting caught up with the posts and sleep a bit (I still need to write up the last of the California flying), I'm going to keep this one a little shorter.

My preflight was more efficient.  I think I'm getting more fluid and less hypersensitive to every little ding these "well used" old trainer planes have.  :)

Getting set up inside was fine, comfortable.  Clearance was fine, and luckily the controller monitoring ground was also handling clearance (oops).  VFR clearance is easy; tower just cares about keeping you safe and separated while getting you out of their airspace.

We were departing 35R, which was a first for me.  Every single other flight has been from 17L.  No big deal, especially with ForeFlight and a georeferenced airport diagram.  We taxied down alpha to mike for CIGARS and the run-up.  At this time, everything was normal, and that's an important note.

There was a slight crosswind from the right, so while taxiing and during takeoff I was using some aileron and rudder correction.  I don't have a great feel for it yet, so we ended up drifting into the wind and not maintaining alignment over the runway.  We weren't off by much and it was corrected quickly enough, and not long after they had us turning eastward (into the wind) to head over to 84R (Smithville).

We cruised along for a few minutes, and upon leveling at 3000' I ran the flow/check.  Mag/DG, engine instruments, flight instruments, breakers/switches, throttle, mixture, fuel.  But at engine instruments I noted the ammeter discharging.  Mark asked me if that was okay, and I STUPIDLY rationalized it with some hurried thought about sure it's discharging, it's supporting a big electrical load.  He didn't say anything at the moment, so I went about my business.  Moments later I did another flow check and found the ALT FLD breaker popped.

Hmm.  Ammeter discharging and an alternator breaker tripped.  I handed the controls over to Mark so I could look it up.  The checklist didn't have any troubleshooting tips for an abnormal electrical situation (just electrical fire).  Next I wanted to look through the POH -- still not sure why the searchable POH wasn't in my documents in ForeFlight -- but flipping pages in the paper POH was taking too long and requiring too much inattention to the airplane.  What if I had been flying alone?  What would a good pilot do?  Should I reduce the load?  Was the fog in the (electric) turn coordinator related?  Would it be safe to land?  What all would be affected?

So my homework (another post) is to research what conditions cause these two symptoms and what is an appropriate way to deal with it.  For that time, Mark had me reset the breaker and keep an eye on it.  Short story, no more symptoms.

We got to Smithville and had a more difficult than expected time at finding the runway.  Thank you, again, ForeFlight, for the situational awareness.  We basically entered the pattern on an extended base, watched to be clear of the water towers and power lines, and went in for my first crosswind landing in ten years.

After practicing the exercise at KHAF with Jason on Monday where we lined up on one side of the runway and cross-controlled to drift across the runway while keeping the nose aligned, it was fresh in my mind of *what* to do.  Doing it well will obviously take more practice.  I think I did ok as far as staying near the middle of the runway, but at touchdown I had a bit of a crab in so there was side-loading and a sensation of skidding to the left as the airplane straightened out.  I kept right aileron in into the wind as we rolled out to the taxiway, but I still had this incredibly unsettling feeling that we were tipping to the left.  Mark assured me we were fully on the ground, but (I think?) he said I could use more aileron correction.

Discussion, recheck of checklists and materials concerning the electrical situation, then back to it. The windsock was showing winds inconsistent in both direction and strength, and my heart sank a little.  I had been excited to get solid crosswind practice, but this was going to require way more change of inputs in response to constantly changing winds, which all in all is a good thing to be able to do but I'm already weak on crosswinds and wanted something nice and stable to work on! We kept our eyes and ears open while back-taxiing down the runway to depart.  After takeoff, I started a turn off to the right to make the pattern and do it again, when Mark asked where I was going.  Oh, dagnabit!  Left traffic!  Thank goodness we were alone in that airspace.  Left turn.  Self-briefing failure.

The wind variability was very noticeable on final.  Constant control changes.  I landed well left of center and didn't straighten out as quickly as I should have, but still got back to center, lifted the flaps and throttled up for a touch-and-go.  C'est la vie.

On the way back to KAUS, Mark gave me some unusual attitudes and I did power-on and power-off stalls under the foggles.  These were fine.  It takes a fair bit of effort to get to the stall point, and that's not where I like to be so Mark had to keep telling me to pull back, pull back, pull back.  Lol.  I just hope that in reality I stay on top of my pitch and airspeed well enough to only have these recoveries in practice.  Part of the unusual attitude bit was interesting.  While hand-flying, he had me look completely down at my lap, as though reading or checking the map or something, all the while trying to maintain altitude and heading.  After a minute, he asked how I thought I was doing (I felt like I had probably climbed but stayed relatively straight) and then told me to recover.  I was banking left and (I think?) descending at a very slow rate.  Ha!  Point made that you can't trust your sensations.

Getting back into KAUS was fine, although with the haze it was very hard to find until we were fairly close.  The KLN89B (in-panel GPS) and ForeFlight were helpful.  Another landing with variable light winds, not well done and hard with the stall horn squealing (on landing that's ok, but I was a foot or two high).  Live and learn.

Oh, my other homework is to go over takeoff emergency checklists.  Mark was unimpressed with my pre-takeoff briefing about emergency procedures should we need to abort or if we experienced a failure before or after rotation.

Except for &$*%# still saying "for 2MA" in response to some ATC comms, my radio work was ok.  Improving.  More room to improve.

Monday, November 07, 2016

Learning is hard work, part 2!

Oh my word.  I'm sitting in the hotel at 7:17 pm trying to scrape together the energy to continue sitting in this chair, typing.  

After lunch, we departed VFR from 30, this time turning left to go toward Half Moon Bay, being wary to stay clear of the SFO Class B surface ring.  Climb over the mountains.

Another enlightening exercise on our way to the coast was to use a dry-erase marker and mark the horizon on the side of the windshield, with an X for a reference point.  The point here was that the sight picture doesn't have to be over the nose, and for me I frequently can't see over the nose, and for all of us there are times when our pitch doesn't allow it, so having a good idea of what to expect from a diagonal view is useful, too.  After flying a bit straight-and-level using just that reference, we changed it up with some slow flight using that reference.  I hate the stall horn.  Or rather, I hate the precipice the stall horn is warning about.  Does anyone like it?  Does anyone like hanging out in a near stall?  (For those of you that don't know, this is what it sounds like.)  Changing sight picture, but you zero in on it and maintain.  Then we pressed the exercise one step further by attempting to maintain altitude and heading using just that reference while going full-power to recover.  If the X stays in the same place, the rudder pressure is correct.


 Hello down there!

Jason did a steep turn so I could snap a few pictures of the coast, then we headed to Half Moon Bay for some landings and to play with slips since I have 80+ hours and only three crosswind landings (all on my solo!).  First approach was straight-in and was way high so we just did a go around (right traffic).  The number and order of patterns is a little vague to me right now, but he demonstrated the crosswind technique as well as options during a power-off landing to change the geometry of the approach and use a forward slip if necessary.  At the key point in the pattern, turning base, in the 172, he likes to be at 20 degrees of flaps and 75 kts, which allows time and space to make choices, like turning in early when you're low or making a late turn to final if too high.  

Half Moon Bay airport is just across that half-moon-shaped bay.

We did not have a crosswind, but practiced the required cross-control by lining up on one side of the runway during a low approach and using aileron to cross the runway while using rudder to keep the nose aligned down the runway.  Jason demonstrated.  Back and forth, weaving across the runway.  This is a small move, smaller than what I was doing when I tried it on the next approach.  I understand it, I know why you use the controls you use, but I don't like it.  (Thanks, solo.) We did it a few more times and I felt like I was kinda starting to feel it. but really need practice.

Next it was DME arcs!  I need to blog this in detail when I'm better rested, but let's just say that I was finally nailing it by the time we finished.  The constant tweaking of the OBS, the checking of distance on the GPS, the heading coordination, the altitude monitoring, the heading changes to move to an arc of a different distance.  It's a high workload operation.  With no wind.  At the end of the exercise, still arcing, Jason had me descend 1000' and use my checklist.  Boom, fell apart.  That was too much workload.  I did not do the checklist and descended through the altitude (but stayed on the arc).


We pretty much came back after that.  I think.  I'm getting fuzzy brained.  :)

Learning is hard work!

4.9 hours over two flights today, and I am both enthused and exhausted!  It was a pretty amazing day, crammed full of good information and good experiences.  If only I could have 100% retention!

This morning we filed KSQL SJC V334 SUNOL V195 ECA to Stockton, planning to do the ILS 29R approach into Stockton.  After taxi and run-up and everything, we waited for probably at least 20 minutes for release because KSQL was the alternate for someone who was approaching KPAO (7nm SE) and probably going missed there.  Didn't happen, but we sat around burning fuel and had time to talk about things like the option to depart VFR and pick up clearance in the air.  

This is a good segway to mention a couple things.  First, using a taxi diagram to, um, you know, taxi.  Jason says this is a bullet point that examiners are looking for, so get in the habit of doing it all the time, even at the familiar airports.  (Pro-tip:  ForeFlight makes this easy with aeronautical charts, georef'd diagrams and plates-on-map.  Just sayin'.)  Second, in addition to the walkaround, preflight, another walkaround (gives a chance to catch any mistakes done during preflighting like missing a tie-down or forgetting to tighten a fuel cap), startup checks, taxi checks, CIGARS and run-up, do a pre-takeoff briefing that includes verifying the runway you're at is the one you're intending to depart, checking the windsock, establishing abort criteria, and reviewing the pre- and post-rotation failure procedures.  The pre-takeoff briefing was done during this time.

In our chats on the ground yesterday, we talked about standard operating procedures becoming a ritual.  There are so many things that we ritualize and it helps us to do things the same way every time and usually for a good reason.  Things like get in the car, fasten seatbelt, crank the engine.  Shower routine.  Getting ready for school routine.  Routines are successful (when followed), and considering it a ritual takes it one step farther.  SOPs for commercial operators have been established and revised in response to multitudes of occurrences that identify weaknesses in the process, and they become safety rituals.  Do it from memory, back it up with a quick run through the checklist for missed items.  And avoid complacency or lack of attention; match up each item on the read-aloud checklist to a specific action you just took.

Ok, back in the run-up area we finally were cleared for takeoff runway 30.  And by the way, this is what the clearance looked like and I did it all by myself! :)  Basically the clearance was along the lines of "Cleared to Stockton airport, after takeoff maintain runway heading until the diamond-shaped waterway, turn right to heading 120 within 2nm of the airport, radar vectors to sierra juliet charlie, victor 334 to SUNOL, victor 195 to Manteca, direct.  Maintain VFR at or below 1100 until the Oakland 165 radial, then climb and maintain 2100, expect 5000 in 5 minutes.  Contact NorCal departure on 135.65, squawk 4503."  Other than saying "runway vectors" in my readback before correcting to radar vectors (and trying to quell my desire to giggle at my own goofiness for saying that), I was pleased.  It's certainly the most complicated thing I've dealt with.  But also credit to Steve at FlightChops and Jason for having shown this clearance in videos -- I may have briefed it before coming to CA... 


Normal takeoff.  Airspeed was alive and over the 70% rotation speed (38kts) well before halfway, and we were up and off.  We had barely crossed the OAK R-165 (which marks the threshold of 30) when they vectored us to the east.  We got to pass through just a bit of wispy clouds before being VFR-on-top crossing the bay.  ATC wanted to vector us straight to KSCK, but since this was a training flight, Jason pushed back and asked if we could go to SJC and fly the airways as filed.  The somewhat grumpy controller gave us a big fat negative and set us toward Stockton.  Soon we changed frequencies and asked the new controller for the airways, and he vectored us toward V334 short of SUNOL.  We got there and turned onto the airway and got our wrists slapped by the controller who deliberately was sending us PAST it and had not cleared us ONTO the airway.  Ok, fine, that was a learning experience, too.  But given the hassle and the lack of IMC in the area, we just canceled IFR and went about our business on the airways as we wanted.  :)  

About that time, we were at the SUNOL intersection so we dialed the 229 radial of the ECA VOR/DME into NAV1 (after identifying) and the OBS and started flying V195.  Well, trying to, anyway.  It was pretty bad.  At the beginning, I thought it was just wind and started bracketing to find the right amount of wind correction.  But it just wasn't working.  The CDI needle would be centered, then would suddenly swing way to one side (usually left) and fluctuate.  We tried all the way to ECA, while also trying to utilize the cruise flow (mag to DG sync, flight instruments, engine instruments, breakers and switches, throttle and mixture, fuel flow controls; back up with checklist readout) periodically, plan for the next action point (5Ts and 5As), and brief the ILS 29R approach.  We had largely given up on the approach briefing as we got to ECA and were still working just on tracking and being suspicious of the CDI, so rather than turn northwest toward the airport (ILS was tuned and waiting), we continued on to work ahead to have time to brief and compare the NAV2 behavior against NAV1; while the NAV2 CDI did have little fluctuations, it was way more stable than the primary so Jason just turned off the primary radio.  It got way better after that.  The proof is in the track:


Starting from the lower left at San Carlos (just below the big label for San Mateo), we flew east, did a little circle when picking up V334 and then getting off of it and then canceling IFR, then did all the wavy tracking to the northeast, first on the lower leg and then on the upper leg to the right.  Coming back from east to west was nav by GPS and NAV2.  The lesson there is that if you're in actual IMC and it's that hard to stay on track, simplify what's going on to give as much attention as possible to get quickly to a reliable state of equipment.

Another lesson:  We had a goal of finishing the approach briefing before ECA, which would have been great since at ECA we should have been turning to start the approach.  Bottom-line items like DA/MDA at what point and the first few steps of the missed approach procedure, but big-picture items like the plan and profile, and detail items like frequencies and published restrictions or adjustments to the standard procedure.  Full briefing didn't happen, and we talked again about creating time to accomplish things and get/stay ahead:  slow down.  Seriously, slow the plane down.  Create time before the next action point.  Groundspeed deviations on an IFR flight plan require reporting when the change is 10 kts or 5%, whichever comes first.

On the way back to ECA, we did a few hold and hold entry visualizations using the heading indicator.  Then at ECA I did my first hold!  Entry has lots of room for improvement, but once established in the hold it was good.  So as long as all of my holds are in no-wind conditions, I'm good to go.  There was even time on the 1-minute legs to do the cruise flow/checklist and plan the 5Ts for the next action point.

On the way back, I had some foggle time to do constant airspeed AND vertical speed climbs and descents, which were ok and with practice I'll be able to nail the power/pitch balance more quickly.  Once established, VSI is the primary instrument for pitch and the airspeed indicator is the primary instrument for power.  I think we may have briefly done a DME arc somewhere in there, but I totally remember my brain slowing down and not feeling on top of things, so we stopped that exercise and ate some snacks (slow brain is a sign of low blood sugar for me for sure).

At various times during the flight, too, Jason covered some instruments for partial-panel work.  One illustrative exercise was to establish straight and level flight, then cover the altimeter and VSI and try to hold altitude.  This was to bring awareness to how very very small the changes in the attitude indicator's sight picture are.  It took less than two seconds to become hypersensitive to that.

I took a minute to snap a few pictures of the beautiful area, too.  The hilly bits would be a nightmare for a forced landing, but pretty to look at.  :)  



And here we are, ready for lunch but feeling like it was a productive flight.


Somewhere in there we failed the DG and flew by the mag compass, which is super screwy but with diligent study and experience would be fine.  There's ANDS (accelerate north, decelerate south) that explains what the compass does on east-west heading when ac/decelerating.  And there's COSUN (compass overshoots south, undershoots north) that explains that on a turn to the south the compass will lead your actual heading by up to 30 degrees and on a turn to the north it will lag by up to 30 degrees.  Oh, and the numbers turn the opposite way from the DG, so there's that complication.  (Increasing or decreasing is what matters...)

Back to KSQL for right base to 30.  Final was so very much like final for 31 at KJGG -- over some swamp-like water with its own monster.  But landing was fine, taxi was fine, and then it was lunchtime!

Sunday, November 06, 2016

It's not a crash course...

... but it is, in a good way :)
 VFR-on-top leaving Austin this morning, but IFR was required to get to it.  But AUS-SFO is IFR anyway since passenger planes use Class A airspace for long trips.

 A preview for our Thanksgiving trip.

Crossing the bay to land at SFO on the airliner.  See how easy it would be to get lost out here?

Today I started my intensive 2+ days of training with Jason Miller.  Not my Jason Miller, the other Jason Miller, of The Finer Points of Flying and CFII out of San Carlos, CA, who helps students like me develop skills and understanding to stay well ahead of the airplane and create a wide margin of safety even in the lowest of conditions.

I've been watching some of his videos alongside the base IFR training videos (from the <chuckle> Kings, Cleared for Approach).  We went over a lot of that material today and some new stuff.  Seeing things in the educational/information-delivery King format, the practical flight lesson King videos, and the safety/ritual perspective in Jason's videos was great, and talking through stuff today will help it all to get glued into place.

The biggest concept that applies to all phases of a flight is to know and be on top of what you're doing now, and know what's next.  If you have the ability to think ahead another step, even better.  Staying ahead means fewer surprises and fewer corrections.

Here are some of the topics we covered:

- Redundancy
Redundancy is hard to argue as a safety feature, yet hard to create in the cockpit, especially when you're flying solo.  Anytime there's a commercial incident, ALL of the commercial companies analyze it, learn from it, and implement redundant operations to ensure it will never happen again.  For a single pilot in a small aircraft, redundancy can be a straight-forward as doing a checklist from memory (by a standard flow) and then following up by reading down the actual checklist to ensure nothing was missed.

- Flow
While we're on the topic, Jason introduced me to a new acronym, because you know aviation needs just one more.  ;)  This one I think will actually be extremely beneficial to me, since one of my areas that is the least redundant (read: always done line-by-line from the checklist) is the pre-takeoff checklist and run-up.  The mnemonic is CIGARS:  Controls, Instruments, Gas (pump/selector/shutoff/quantity), Annunciators/Autopilot/Attitude trim, Radios/run-up, Seats/Seatbelts/Security.  Pair that with the last step of Lights, Camera, Action, or integrate the two, and I'd say it's covered.  Lights could be included either in "Instruments (and switches)" or as switches in the S section, kinda like the S in GUMPS for pre-landing check.  Camera is avionics which would go with instruments.  Action is the gas portion.

- Pre-takeoff briefing
When was the last time you thought about aborting a takeoff?  Before the situation?  Do you consider it every time?  I don't.  I think about it now and then when I start to get uptight about not having refreshed myself on emergency procedures lately.  This is a "learn from others' mistakes" opportunity.   Brief it every time, as a ritual, and it will always be fresh.  Takeoff is all action so there's no time to grope for a checklist.

  •  Before taking off, identify again that you're on the correct runway.  
  • Check the winds one last time.  
  • Determine what 1000' above the field is (it's roughly 1500' at home, but here on the coast it's 1000').  
  • Understand the abort plan -- if you haven't hit 70% of the rotation speed by the halfway point on the runway, abort (and know the actions).  In the Cessnas I fly, that's a whopping 38kts, which basically means the airspeed indicator hasn't even come alive!  What if the oil pressure isn't in the green?  Consider your other reasons to abort.
  • Review the emergency procedures, for both before and after rotation.


Oh man, we talked about so much stuff.  I didn't take notes on all of it.  We spent a good bit of time talking about holding patterns and performing holds at intersections.  It turns out that my difficulty came in the nomenclature, and using a VOR as the hold fix makes it clear why it was unclear.  The holding course is the inbound course on an outbound radial, so you use the reciprocal heading.  The outbound course is parallel to the outbound radial on the outbound heading.  But all of that aside, remember only these two things:

  1. Holding courses are always given in the clearance as INBOUND TO THE FIX.
  2. ALWAYS dial the OBS TO the fix.
We did a few exercises for how to visualize a hold on the DG and choose the appropriate of the three entry methods (straight-in, teardrop, or parallel), then double-checked with the thumb/sector tool for redundancy.

We talked about DME arcs, which I understand (on paper, at least).

We talked about the scan methods.  Radial for cruise and inverted V for turns.  For a radial scan (which is what I have been working on with Mark), Jason recommends the cadence of a certain well-known 70s song.  I try moving my eyes that way here just sitting in the hotel room and it hurts and I can't keep up.  It'll take time to get a scan going that fast that can actually meaningfully interpret what is shown on each instrument.  The inverted V uses the turn coordinator (electric), the attitude indicator (vacuum), and the VSI (static).  It inherently includes a chance to recognize early symptoms of failure on any of those systems.

So much we talked about.  I have homework tonight, too.  There's an Air Safety Institute video about a particular fatal accident because of running out of fuel from not being able to land at an airport or several alternates.  I will review a lesson from a "classic" AIM publication covering climbs and descents of certain characteristics.

We'll meet up at the airport at 7 am, file and fly a specific flight out to Stockton, then do the Oscar pattern (a rectangle with 3 minute sides, where the middle two minutes are standard rate 500 fpm climbing or descending turns to the left or right.  The Oscar pattern is an exercise that evolves as the pilot's capabilities expand: required use of flow+checklists on every change, starting headings that make corners harder to calculate, shorter lead-in/out legs, and additional workload layered on top.

Phew.  I'm tired, but no where near as tired as I'll be this time tomorrow!