I just earned my FAA instrument rating and a considerable amount of time was spent training partial panel. This meant flying with no vacuum gyro instruments - attitude indicator and directional gyro. After my check ride the examiner said "you fly better without those instruments" which I considered a nice compliment to both myself and my instructor.

In reflecting upon this, I started thinking, what if I had an additional failure of something - how would I handle it? I launched X-Plane and set up an LNAV+V GPS approach on a windy, bumpy low IMC day and gave it a shot. Unsurprisingly, this was HARD.

I used the instruments as follows:

  • Altimeter / VSI and on the final approach the VNAV glide slope for altitude control. Not too bad.
  • GPS track and a bit of magnetic compass checking, along with the CDI needle. This is a very poor way to navigate. The magnetic compass obviously has lagging/leading errors and the GPS track is what I'd call a rate instrument - it lags a little.

The main problems I faced were that you can't time turns with no turn coordinator and the bumpy/windy air made constant corrections required. There appears to be no way to identify wings level. You can tell when you're no longer turning but you might be starting a turn the other way.

My strategy as I worked through this was to slow the plane (C172) down with 10 degrees of flaps throughout the procedure to make heading changes have the least effect. In the end, I tried to keep the track as close to the DTK and the CDI as centered as possible - effectively chasing two needles. This resulted in a sinusoidal course when I looked back at it on the X-Plane instructor console.

(By the way, I did land the plane safely but I'm not sure I'd be able to do this in the real world - the workload I experienced in my comfy chair was EXTREME.)

I know the scenario is unlikely but it's certainly possible. There appears to be no guidance that I can find on the Internet about it.

My question: is there a better way to do this? If so, what's the procedure?

  • 2
    $\begingroup$ "Slow the plane [to] have the least effect." Don't planes have higher rate of heading change the slower they are? Or did I misinterpret what you meant? Anyway nice question you have there and congrats on the IR. $\endgroup$
    – user14897
    Commented Feb 24, 2018 at 19:21
  • 3
    $\begingroup$ "is there a better way to do this?" You mean besides contacting ATC and declaring an emergency because you have lost a bunch of flight-critical instruments? $\endgroup$
    – user
    Commented Feb 24, 2018 at 19:49
  • 2
    $\begingroup$ @MichaelKjörling, well, if the ATC does not have a precision approach radar, they won't be able to help you all that much. And most don't. $\endgroup$
    – Jan Hudec
    Commented Feb 24, 2018 at 21:07
  • 1
    $\begingroup$ Good points above. I hadn't given more thought to the broader picture of this, but let's assume that in my scenario, I'm assuming no precision radar, ATC offers this is the "best" airport and approach for weather and my available fuel. The approach had both vertical and horizontal guidance so that's probably reasonable. As for speed, my thought was that if I was weaving back and forth, I wouldn't go off course too far if I was slow. Again, not too realistic but the maneuvers were what I was interested in here more than the scenario. $\endgroup$
    – Hawkman
    Commented Feb 24, 2018 at 23:05
  • 1
    $\begingroup$ In reality, my airplane has a GTN750 with the FlightStream 210. This gives a backup AI on my iPad along with a synthetic turn coordinator so I do actually have this redundancy in the real world. This was more to try out a really tough scenario than anything else. A "bad day in the office" if you will... $\endgroup$
    – Hawkman
    Commented Feb 24, 2018 at 23:07

2 Answers 2


This isn't exactly your scenario but at my last instrument checkride the DPE asked me how I would descend from VFR-on-top through a solid cloud layer following complete electrical and vacuum failure (in a C172 or something similar). No avionics meant no help from ATC (vectors, or even a PAR approach) and we also assumed that I had no tablet, phone or other external systems that could help me.

The question came at the end of a discussion about aircraft systems and failures and the DPE didn't expect me to have a ready answer because it isn't something you train for. But he suggested this procedure:

  1. Set up the aircraft on an east or west heading to minimize compass turning errors
  2. Use the rudder only to fly a constant heading on the compass; once you have the aircraft going straight, do not move your feet again
  3. Set power and trim for a stable descent at a little above Va and a little less than your usual descent rate, e.g. 300fpm
  4. Sit back, cross your arms and wait to break out. Resist the temptation to touch the ailerons or power.

His reasoning was that by avoiding using the ailerons and power you can avoid going into a graveyard spiral quickly. You may still do it, but you'll do it a lot more gradually and hopefully by that point you'll have broken out anyway. I've never tried it myself, but I'll have to do it sometime with a safety pilot.

In real life, if you can't get to VMC easily then a tablet app with a portable AHRS unit would be a good backup. Although personally, if I ever lost the complete panel my first thought would be to get to VMC, not to shoot an approach. I suspect that for most people, landing in a field is likely to be the much better choice. But like all 'what if' scenarios, it depends.

  • 2
    $\begingroup$ Now this is an interesting story about the DPE suggesting to use the rudder only (and I am not questioning it a bit). On one of my 135 checkrides an examiner said that he would fail me if he noticed that I 'steer with the rudder only' on a localizer. Aerodynamically it seems using the rudder too much would get you into a spin rather than a spiral. Which, incidentally, is a rather stable way of getting through a layer as long as you are sure there is at least 1000' clear at the bottom (they used to do this in the old days). Spinning a twin though ... $\endgroup$
    – alexsh
    Commented Mar 2, 2018 at 2:14
  • $\begingroup$ East and west are not the best headings at least for flying at mid-latitudes. See my answer. $\endgroup$ Commented Oct 17, 2018 at 15:19
  • $\begingroup$ @quietflyer I assume my DPE believed it was best to minimize the turning errors: his point was that you should resist the urge to move the controls anyway. If the compass swings a lot due to turning error then the temptation would be to overcorrect what's actually a minor turn, or turn in completely the wrong direction. Whether or not it works in practice, I don't know (I still haven't tried it!). $\endgroup$
    – Pondlife
    Commented Oct 18, 2018 at 1:44
  • $\begingroup$ @alexsh: Then again, planes back then were a lot easier to get out of a spin (even a fully-developed flat spin) than they are nowadays - lots lighter with lots less angular momentum to null out. $\endgroup$
    – Vikki
    Commented Dec 14, 2021 at 23:05

It is possible to use the wet magnetic compass to descend through even a rather thick cloud layer if the air is not too turbulent. In mid-latitudes in the northern (magnetic) hemisphere it is important to choose a course close to due S-- not further than 45 degrees off and preferably less. On a southerly heading, the compass will over-react to a slight bank which is extremely helpful. If your heading wanders to north of magnetic east or west the compass will tend to turn backwards of the appropriate direction for the turn-- this is deadly-- you will likely end up in a steep-banked turn with the compass spinning wildly, reversing direction, etc.

In an aircraft with good yaw (slip)- roll coupling, steer with the rudder pedals rather than the yoke. This makes the compass over-sensitive to your inputs which tends to prevent over-controlling.

I've seen video of a Cessna 120 descending through several thousand feet of cloud in this manner, with tape covering all gyro instruments. It was very clear that the pilot was actively making steering inputs, not just relying on the aircraft's inherent stability, which would not have been sufficient to keep the aircraft wings-level indefinitely.

This technique may not work near the magnetic equator to the lack of the "lead" effect-- the over-reaction to a slight bank-- in this area. Likewise this technique will not work at very high (magnetic) latitudes.

Then there's always the recourse of a hand-held GPS, set to the compass-like "heading" display (actually direction of ground track). In my experience there's too much lag for this to be reliable, and the wet magnetic compass offers superior guidance. This not be the case with higher-performance GPS units, especially those that simulate aircraft-style displays.

  • $\begingroup$ missing word ( this MAY not be the case) in last sentence $\endgroup$ Commented Oct 18, 2018 at 7:46
  • 1
    $\begingroup$ Note that in most aircraft the wet magnetic compass-- or alternatively a vertical card compass which still suffers from many of the same errors --- is mounted in such a location that it is almost impossible to practice flying "under the hood" by reference to this device. This may partly account for the diversity of opinions on this topic. $\endgroup$ Commented Oct 18, 2018 at 7:50

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .