Partial panel/loss of control in IMC

[BorealOne]

Not a Mooney incident, but a chilling read - 

https://www.tsb.gc.ca/eng/enquetes-investigations/aviation/2019/a19w0015/a19w0015.html

Pilots on board a King Air 200 depart with 1 working AI. The working AI rolls in descent in IMC. Pilots lose control, plane enters spiral dive. Plane impacts terrain.

What stands out for me is that the pilot flying had a perfectly functional iPad Mini on his yoke, running Foreflight with synthetic vision. The pilot failed to make use of it.

 

My take away - and recommendation for Mooney pilots similarly equipped - is to practice partial panel before you have to do it for real, and to be familiar with the use of the standby AHRS system on Foreflight.

 

[GeeBee]

So let me get this straight. You depart in severe IFR conditions with one AI because you believe the other will "eventually come up". Is that what I am reading here?

iPads are nice, but the MEMs modules are not reliable either. How many times do you pull out your IOs device and it will not align to earth?

 

 

[N201MKTurbo]

A friend of mine flew freight at night for a year or so. Just out of pure boredom he flew a flight one night with just the iPad. He had the iPad propped up blocking the instruments. It was real IFR with icing and all. He said it was way easier then the crap instruments in the plane.

BTW, my little Mooney is way better equipped than this C402 freight dog. except for the boots and spare engine.

[N201MKTurbo]

I remember when partial panel was flying with an altimeter, wet compass and turn coordinator. 

Now partial panel is flying with an iPad with a fully functioning AHRS.

[Ibra]

Maybe not just the lack of partial panel currency, (just guessing) they were probably just flying too fast on that decent just near the edge and were simply not ready for AP disengage (on full panel let alone fly on partial panel), it would have been good to know what was their decent speed & power setting? my feeling AP disconnect on high power with fast cruise decent and out of trim condition are tough to sort out (let alone after the only gyro has failed)

It will be good to know what speed/power setting Mooney pilots use for cruise decent in IMC from say from FL200 to 2kft? which speeds on ASI/VSI? with AP or hand fly? how many will chicken out and throttle back in IMC (usually it gets bumpy as one goes down )? and who keeps it near VNE until the IAF?

Also, not related, for approach on partial panel does ATC in Canada offer "no-gyro" vectors for ILS/RNAV? 
(well they lost it in cruise descent rather than approach to minima on partial panel)

Edited May 5, 2020 by Ibra

[Hank]

Just now, Ibra said:

It will be good to know what speed/power setting Mooney pilots use for cruise decent in IMC from say from FL200 to 2kft? which speeds on ASI/VSI? with AP or hand fly? how many will chicken out and throttle back in IMC (usually it gets bumpy as one goes down )? and who keeps it near VNE until the IAF?

My little C won't reach FL200, but I almost always descend at 500 fpm and maintain cruise MP and EGT all the way down (unless turbulence gets too bad, so far not often). This leaves me near the top of the green (165-170 mph) but well above Va (132 mph at gross, decreasing as weight goes down). 

So far, my longest descent like this was from 10,000 msl cruise to land near sea level (<200 msl). IMC, VMC, AP, Hand flying, all same same--power on, 500 fpm when able. 

[GeeBee]

Did your friend pull out and reference his iPad after the airplane had rolled past 60 degrees of bank? Do this experiment, bank over to 60 degrees then activate your panel app on your iPad and see what happens.

 

 

[BorealOne]

1 hour ago, Ibra said:

Maybe not just the lack of partial panel currency, (just guessing) they were probably just flying too fast on that decent just near the edge and were simply not ready for AP disengage (on full panel let alone fly on partial panel), it would have been good to know what was their decent speed & power setting? my feeling AP disconnect on high power with fast cruise decent and out of trim condition are tough to sort out (let alone after the only gyro has failed)

It will be good to know what speed/power setting Mooney pilots use for cruise decent in IMC from say from FL200 to 2kft? which speeds on ASI/VSI? with AP or hand fly? how many will chicken out and throttle back in IMC (usually it gets bumpy as one goes down )? and who keeps it near VNE until the IAF?

Also, not related, for approach on partial panel does ATC in Canada offer "no-gyro" vectors for ILS/RNAV? 
(well they lost it in cruise descent rather than approach to minima on partial panel)

That seems to have been the case here. Having rolled into a spiral dive, they were going downhill fast. There is a pretty good chance that the AP was already following the failing gyro into the roll before it disconnected. as the TSB notes: "During the descent, the aircraft was calculated to have reached a maximum airspeed of over 400 knots calibrated airspeed (KCAS) just before impact, which exceeds the maximum operating speed of the aircraft by approximately 141 knots. The aircraft also achieved a maximum descent rate of 35 637 fpm". 

The no-gyro radar vectors option is avail where we have radar coverage, but that can be limited in Canada (and particularly here in the North). They wouldn't have been in radar range in this instance.

 

[Ibra]

4 hours ago, N201MKTurbo said:

Now partial panel is flying with an iPad with a fully functioning AHRS.

You can take it to the next level, without AHRS you get "iPad partial panel": GPS derived ground speed, GPS derived vertical speed and HSI on GPS track   

You can still fly based on that, surely not "IFR flying" (e.g. instrument flying/landing inside ATC system or following a PAR) but it may help to stay alive for 10min while "cloud flying" until you find VMC?

The value of any backup is simply what's one expects from it it terms of mission A capability and plan B

[RobertGary1]

Not sure setting up an ipad happens quick enough to be useful. For decades most of all always departed on 1 attitude indicator. That was all that Mooney provided us. Today most of us have backups. I have 3 in panel attitude indicators (1 vacuum, 2 with their own AHRS units). Never understood why people were so excited to remove their vacuum system. Its a very cheap way to have an extra attitude indicator.

 

-Robert

[BorealOne]

20 minutes ago, RobertGary1 said:

Not sure setting up an ipad happens quick enough to be useful. For decades most of all always departed on 1 attitude indicator. That was all that Mooney provided us. Today most of us have backups. I have 3 in panel attitude indicators (1 vacuum, 2 with their own AHRS units). Never understood why people were so excited to remove their vacuum system. Its a very cheap way to have an extra attitude indicator.

 

-Robert

Agree. If you are pulling it out an iPad the middle of an upset in IMC, you are doing it wrong. But if you have it up and running already on your yoke, running a moving map display, or in the case of the lost King Air crew - on both yokes - there's no reason not to use it. 

I'm with you - the more redundancy I have, the happier I am. Having lost my vacuum AI twice now, both in IMC, I've always appreciated having a backup. Or two.

[ArtVandelay]

My vacuum system and electric backup pump weighed 20lbs, and the pumps failed regularly.

[Vance Harral]

Strongly suggest those of you relying on an iPad for emergency attitude information cover up your traditional instruments and actually spend some time flying with it during training.  I'm sure some of you have, but for those that haven't... when I gave it a shot, I found it surprisingly difficult.  Partly this is due to the AHARS built in to our GTX-345, which turns out to be marginally OK for keeping the greasy side down, but rather poor for precision control (frequently shows a slight bank when straight and level, and lags actual attitude just enough to make fine control difficult).  Partly this is due to the Foreflight display, which I found distracting and "swimmy" for various reasons that are probably just personal.  The point here is not to say it can't be done, it most certainly can.  But it's just different enough that I sure wouldn't want to be trying it for the first time in an actual failure scenario.

My casual poll of a locals who are excited about the emergency backup capability of their iPad suggest about half of them have never actually used it in real or even simulated IMC.  Like any other tool, it shouldn't be relied on unless you've trained with it.

[gsxrpilot]

1 hour ago, Vance Harral said:

Strongly suggest those of you relying on an iPad for emergency attitude information cover up your traditional instruments and actually spend some time flying with it during training.  I'm sure some of you have, but for those that haven't... when I gave it a shot, I found it surprisingly difficult.  Partly this is due to the AHARS built in to our GTX-345, which turns out to be marginally OK for keeping the greasy side down, but rather poor for precision control (frequently shows a slight bank when straight and level, and lags actual attitude just enough to make fine control difficult).  Partly this is due to the Foreflight display, which I found distracting and "swimmy" for various reasons that are probably just personal.  The point here is not to say it can't be done, it most certainly can.  But it's just different enough that I sure wouldn't want to be trying it for the first time in an actual failure scenario.

My casual poll of a locals who are excited about the emergency backup capability of their iPad suggest about half of them have never actually used it in real or even simulated IMC.  Like any other tool, it shouldn't be relied on unless you've trained with it.

We should go try it out sometime. We'll put the foggles on and see if we can fly off the iPad in my 252.

[Ibra]

2 hours ago, Vance Harral said:

My casual poll of a locals who are excited about the emergency backup capability of their iPad suggest about half of them have never actually used it in real or even simulated IMC.  Like any other tool, it shouldn't be relied on unless you've trained with it.

Yes, iPad precision & lag will perhaps not help as backup for cruising at 75% power, that will require lot of accuracy: less than 5deg bank and pitch to keep things in shape without going wild in split seconds but I guess there are better speeds & power settings where an aircraft is perfectly controllable by reference to an iPad? maybe slow it down to approach speeds and try unusual attitudes recovery on iPad with full control deflection? after all you just need to know if turning left/right and maintain height & speed to stay alive (when flying at 90kts level cruise nothing bad happens when the nose/wings are 45deg away from where they should for at least 10 seconds, it should be enough), of course, I will go for the turn indicator first (less ideal for a precision approach) but I also take iPad as serious backup to get out of IMC and find VMC but I am not expecting lot from it (not for 2h of cross-country at 170kts and surely not for landing in 200ft ceiling & 800m vis, to some extent the same applies to flying on turn indicator for backup as well)

Edited May 5, 2020 by Ibra

[steingar]

I ever get to do the IR thing I'll use the iPad with foggles.  I doubt I will ever have the coin to loose the vacuum.  Too much money.  Gotta have some kind of backup.  Guys better than me have died when their pump broke.

[Davidv]

On a side note, quick question for those familiar with the inside workings of a gyro.  My KG-258 will intermittently slowly stop working on a flight to the point where it's at a 45 degree angle in level flight.  Fortunately, this gives me a lot of practice with partial panel even though I have an electric backup on the bottom of my panel.

I flew a 2:30 flight from FXE to ECP on Saturday and it failed about an hour into the flight (which meant that I had to hand fly the rest of the way since its connected to the AP).  I flew at low altitude (4500 feet) due to winds and it was fairly warm and humid.  Last night I flew from Rusk County Texas to FXE at 17,000 ft and it took roughly 4:30.  During this flight my gyro worked perfectly.  I've also noticed on other higher flights it works better.  My theory is that it has something to do with the humidity and air density being fed into the vaccum system.  Does that make sense to anyone or any other ideas?  I'm probably not far off from replacing the unit with something digital but going to get some more use out of it for now.

[Eduleo]

4 hours ago, Davidv said:

On a side note, quick question for those familiar with the inside workings of a gyro.  My KG-258 will intermittently slowly stop working on a flight to the point where it's at a 45 degree angle in level flight.  Fortunately, this gives me a lot of practice with partial panel even though I have an electric backup on the bottom of my panel.

I flew a 2:30 flight from FXE to ECP on Saturday and it failed about an hour into the flight (which meant that I had to hand fly the rest of the way since its connected to the AP).  I flew at low altitude (4500 feet) due to winds and it was fairly warm and humid.  Last night I flew from Rusk County Texas to FXE at 17,000 ft and it took roughly 4:30.  During this flight my gyro worked perfectly.  I've also noticed on other higher flights it works better.  My theory is that it has something to do with the humidity and air density being fed into the vaccum system.  Does that make sense to anyone or any other ideas?  I'm probably not far off from replacing the unit with something digital but going to get some more use out of it for now.

Like this?

My KG258 failed twice before I knew I had to replace it with an overhauled unit for around 3 AMUs. Once I got the unit from Mid-Atlantic it was a short install and autopilot calibration.

This video is from the second time it failed, and by then I had the FS210 AHRS on the iPad all of the time. Still annoying to not have the KAP150 autopilot but no problem keeping the shiny side up even at night. After this flight, even with the replaced unit, I always have the iPad on the FS210 and the iPhone with a Stratux velcroed to the glareshield any time I'm in IMC or at night. At least until I get the two GI275s in and lose the vacuum, standby vacuum, DG, and AI, hopefully by August. 

[Davidv]

Yes, just like that! I find that even if it's off by 20 degrees or so the altitude works and I can "trick" the autopilot in heading mode by selecting a heading further to the right.  At 45 degrees the heading bug is no good...

[Guest]

Vacuum pumps cost less than $1.00 per hour based on a nominal 500 hour life.  If you don’t let the drive shaft seal leak oil into the pump, change the filters regularly and install a cooling shroud they last quite long in my experience.

Clarence

[Vance Harral]

On 5/5/2020 at 5:08 PM, Ibra said:

but I guess there are better speeds & power settings where an aircraft is perfectly controllable by reference to an iPad? maybe slow it down to approach speeds and try unusual attitudes recovery on iPad with full control deflection? after all you just need to know if turning left/right and maintain height & speed to stay alive (when flying at 90kts level cruise nothing bad happens when the nose/wings are 45deg away from where they should for at least 10 seconds, it should be enough), of course, I will go for the turn indicator first (less ideal for a precision approach) but I also take iPad as serious backup to get out of IMC and find VMC but I am not expecting lot from it (not for 2h of cross-country at 170kts and surely not for landing in 200ft ceiling & 800m vis, to some extent the same applies to flying on turn indicator for backup as well)

The problems I described with my experiment using the iPad as an emergency backup occurred while flying level approach at 90 knots.  I disagree with the suggestion they can be alleviated by slowing down.  Or, if they do vary with speed, I actually think you have it backward: I think they're less of an issue in cruise than they are on an approach.

Let me elaborate on a couple of things.  First, it's not that I had trouble keeping the aircraft upright or that I was ever anywhere close to an unusual attitude.  Rather, the errors and lag in the displayed attitude resulted in my chasing heading/track and altitude excursions.  e.g. I'd set a straight and level attitude on the display, but see track and/or altitude slowly changing.  So I'd input a small bank and/or pitch correction to fix that - which might or might not be immediately reflected in the attitude display.  Then I'd try to re-frame my brain, like "to hold a constant heading, I think the display needs to show a 2-degree left bank"; only to discover the error didn't seem to be consistent.  The net result was that my performance wasn't really any better with the backup iPad attitude display than with traditional partial panel technique.  I mostly blame this on the GTX-345 AHRS that generates the attitude information.  I've seen several reports that it's simply not a high-quality attitude reference, and my own experience agrees: OK for emergency backup, but just not accurate enough for the single-digit bank angles and half-bar pitch changes associated with precision instrument flying.  My guess is most of the AHRSs people are thinking of as emergency backups are similar: Stratus 2/3, Sentry, GDL 52, etc.  I freely admit I only have experience with the GTX-345.  But my technical reading on the matter leads me to believe there's a good reason why certifies AHRS systems are expensive, and inexpensive AHRS systems aren't certified.

The second issue for me wasn't a problem with the hardware/software, but rather a lack of experience with high-end, digital attitude depiction.  I have essentially no time behind a G1000, Aspen Evolution, etc; and therefore no experience with synthetic vision depictions.  By default, the Foreflight attitude display is backed by synthetic vision information: it includes both terrain depictions, and augmented-reality pointers that show nearby airports.  What I didn't appreciate until I tried it is that if you hold a fixed pitch/roll attitude on this sort of display, the depiction changes even though your attitude is not changing.  Perhaps this is intuitively obvious in a turn.  But even when flying straight and level, terrain and airport pointers slowly slide down and to the side of the display as your position moves forward.  Not being distracted by this requires an adjustment in thinking, even on a huge display with an infinite pixel count and frame rate.  On a small display with limited pixel count and limited refresh rate, it's even more distracting to a pilot used to looking at an old iron gyro, or even a digital presentation with a simple sky/ground background depiction.  I have no doubt I can get used to this, and I'll continue working on it.  But it's another great example of how a very nice tool can get a pilot in trouble if they're trying to use it for the first time in a high stress situation.

[N201MKTurbo]

I don't see those delays while using the Stratus and the  IPad. Just don't re-position the Stratus.

[Ibra]

58 minutes ago, Vance Harral said:

Or, if they do vary with speed, I actually think you have it backward: I think they're less of an issue in cruise than they are on an approach

It is the opposite: the aircraft that can be flown easily and safely at those speeds handoff in without lot of precision even with distractions or disorientations, at 90kts cruise I can do anything prop, gear, flaps (even mixture ) and apply full control deflection of yoke & rudder with 45deg pitch and rolls and really nothing bad happens for about 10 seconds in VMC/IMC

If the iPad has less than 20deg roll/pitch accuracy or less than 20seconds lag, it can safely use to fly at that 90kts cruise speed, note that all piston aircrafts tend to fly stable & hand off at about 1.3*VS0 cruise (with about 40% power and 85kts in mine), this is so deep and stable inside flight envelope with lot of margins against degradation of flying precision and flight envelope corners, anywhere else (faster or slower) will need split second attention and 1deg roll/pitch flying precisions...

At 160kts Rate1 turn is 25deg and 3deg pitch gradient is 800fpm, hard to stay on top of aircraft in IMC if gyro has more than 5deg error, while at 90kts Rate1 turn is acceptable 15deg (can be achieved with rudder only ) and the 3deg pitch gradient is 500fpm, so probably a more ideal speed for IMC flying on partial panel?

Of course, consult POH to find V-speeds but you need to look really hard to find a piston aircraft that can't fly safely at 1.3*VS0 with minimal effort and attention (there are good reasons why this is the default "approach speed" for flying circuit and landing in any config, it also sits close to max rate climb Vy, best glide Vbg and min radius turn at 45deg, always bellow Vne, Vno, Va, Vfe, Vle, Vlo, Vb...and always above Vs0, Vs1, Vx, Vs45, Vmc, Vxse...)

58 minutes ago, Vance Harral said:

But it's another great example of how a very nice tool can get a pilot in trouble if they're trying to use it for the first time in a high stress situation.

Yes that is the big issue familiarity, if you switch to iPad screen in anger, how do you know it has all wrong as well?
But to stay alive on it as backup you only need few reliable outputs from it and filter out the rest

I have abused non-certified AHRS+GPS as backup when gliding but I was young and had parachutes & airbrakes,
I don't have that luxury in the Mooney but I am only planning to use it to find a safe way out to VMC

Edited May 7, 2020 by Ibra

[N201MKTurbo]

45 minutes ago, Ibra said:

I have abused non-certified AHRS+GPS as backup when gliding but I was young.

Huh, there was no GPS and AHRS when I was young... Or even when I started flying....

[Vance Harral]

1 hour ago, Ibra said:

It is the opposite: the aircraft that can be flown easily and safely at those speeds handoff in without lot of precision even with distractions or disorientations, at 90kts cruise I can do anything prop, gear, flaps (even mixture ) and apply full control deflection of yoke & rudder with 45deg pitch and rolls and really nothing bad happens for about 10 seconds in VMC/IMC

I understand now that you and I are talking about two different things.

You are talking about not losing gross control of the aircraft: getting upside down, ripping the wings off, etc.  I agree almost any "low end" backup attitude indicator is helpful in reducing that risk, including the imprecise AHRS in the GTX-345.  I also agree the risk of complete loss of control is reduced at slower speeds.

But I'm talking about executing an instrument approach to minimums, well enough not to get off course resulting in CFIT, even though the airplane is still flying at normal attitudes.  My odds of being able to do that traditional partial panel were "decent" because I practiced it... at least until a recent 2+ month hiatus due to the combination of a complicated annual and the COVID pandemic.  But even at my most proficient, there has always been room for improvement in my partial panel skills.  Accordingly, I chose not to fly my airplane over widespread low IMC.  I was considering changing that strategy after installing a GTX-345, with the rationale that I now have attitude backup if my vacuum AI rolls over and dies.  Based on what I've seen actually flying with the iPad and GTX-345 though, I've flipped back to my original mindset.  I reserve the right to adjust my risk profile yet again with further practice, but I'm not there today.

Having said all that, my original point stands: I've met pilots who changed their risk management profile based on having backup attitude information they have not actually practiced with.  That's a sketchy idea even with a high-end, certified attitude backup solution.  It's an even worse idea with a portable, uncertified solution.  Go practice with your backup, then adjust your risk management profile based on your experience.  Make the decision based on real flying, not armchair flying.