5 things I learnt from a recent vacuum pump failure

The instrument rating training covers wide areas about how instruments work, how failures manifest and how to work without certain instruments. A recent experience demonstrated how different an actual failure can be.

It happened two weeks ago . We were flying on an IFR currency flight in visual conditions at night with a fellow safety pilot. I flew under the hood and we just came out of a hold to fly back to AUR VOR in Guatemala city. I usually hand-fly currency flights without using the autopilot, but decided to turn on the autopilot to brief the approach I was given. We were cruising at 12,000 ft. and ATC gave us a new QNH which changed the indicated altitude to 12,100 ft. I disabled the altitude mode on my autopilot and moved the wheel to a descent. The artificial horizon still showed a climb. So I tried a more aggressive descent without success and until I reached the end of the wheel’s margin. That has never happened to me before! Right then I knew something was not right, but it took me a couple of seconds longer to process the information from the cross-check instruments. My brain was stuck for a little while before being able conclude what had just happened. I stayed under the hood, disconnected the autopilot and looked for my stand-by vacuum pump. Yeah, I was looking for it. As for an emergency instrument I had looked it up with POH but never had done anything else to it. While the horizon came back, I realized how difficult it would be to fly straight and level under this conditions. The artificial horizon showed a 5-10° climb now and a slight turn to the left. After maybe 2 minutes the horizon came back and the rest of the flight was uneventful. There was no danger or panic at any point. I stayed under the hood, but would this have been the same in a real IMC situation? I made several notes to myself to learn from this event. I have to admit that I have read about most, maybe all in the past, but having experienced it first hand is an eye-opener.

  1. Understand how your autopilot works

    My Autopilot is connected to the artificial horizon. There is no warning if the vacuum goes out nor does it automatically disconnect. The autopilot will follow what the artificial horizon is doing until reaching its limits and then disconnect. This might take a while and get you in a pretty bad situation. Read about your autopilot on your POH and understand what happens if the vacuum pump stops working (or any other instrument). This understanding will be helpful or even life-saving when this happens to you. In my case I have to disconnect the autopilot immediately when a vacuum pump failure is detected.

  2. Detecting a vacuum pump failure is a problem

    The vacuum pump drives your main instrument in the old steam gauge cockpit. You would expect that a big warning signal or something rings when it doesn’t work any more. Contrary to that, it just slowly stops working. Now, you are supposed to cross-check your instruments and detect discrepancies. But this is something you train once a year, if you are a very responsible pilot, otherwise maybe a couple of times in your life, at best. How good are you supposed to be at detecting it? Another thing is the suction instrument, it should be part of your scan. So imagine you fly 5 years without any problem. So 5 years long that little instrument has not given you any useful information. It only does when something bad happens. That is a mental challenge I believe. In addition, the suction gauge is often, definitely in my case, outside the instrument scan area, making it harder to include it in a natural scan. Sorry, no solution to this problem yet, just the realization that detecting a vacuum pump failure can be pretty hard. Hard translates into time (to detect) and depending on your phase of flight that might be something you do not have. Just saying, this topic requires attention.

  3. Artificial Horizon – better none than incorrect

    As part of your instrument training, you fly without an artificial horizon, kind of standard. It is covered up and you use the other instruments to fly the plane. The difference is that in a vacuum failure scenario, the instrument is not covered up. Instead it is right there and gives your wrong information. It makes it much harder to process as your central instrument now has to be ignored, which really is much harder than you would think. For the future, I decided to keep one of those instrument covers reachable from the pilot seat to cover the artificial horizon in a vacuum failure scenario. It will make it much easier!

  4. Backup instrument could save your life

    In the past, I bought a Dynon D2 and have since installed it whenever flying in instrument, marginal or declining conditions. This is not an FAA certified device. So officially this is not a backup instrument. Flying compass heading and rely on heading indicator for altitude changes is ok for backup though. Well, I decided to have both and I now confirm its worth it. This device is 100% independent and will run on battery for an hour or so. It gives you attitude information plus some GPS information. Who cares if it is certified if it can save your life? Oh and as a little bonus I have found the G meter functionality useful for afraid passengers. Telling them how many Gs would be necessary to get just in the yellow zone and then showing them how little Gs even in mild turbulence are being pulled, can help tremendously.

  5. No Gyro should be trained regularly

    Again, this forms part of instrument training. Compass heading and altitude indicator as primary indicators… hard to fly a straight line. I have decided that in one of my next instrument currency flights with a safety pilot I will fly a big portion without artificial horizon. Another time I will try to fly with the Dynon D2 and see how that feels. I am sure this will be a great exercise and confidence builder.

    Here is another interesting question. Could you fly with you Garmin 795 3D view and get out of an IMC condition with only that? Interesting question we had discussed with some other pilot friends. My take would be that probably you could, especially at higher speeds the GPS should give you pretty accurate information. The more wind comes into play and the slower you fly, maybe in uncoordinated fashion, this might get tricky. Well, this might be a project we could evaluate with other pilots on a visual day. More to come…


Oh and here is the picture of the broken vacuum pump. By the way I realized in the last flights, that it took longer than usual to stabilize. Maybe an early indicator. Definitely something I will be watching out for much more closely in the future.

Wish you everlasting vacuum pumps!

Broken Vacuum Pump

Broken Vacuum Pump