Engine Failure

[Fuji Abound]

Is deploying the parachute in the case of an engine failure in a Cirrus a recommended procedure? The Cirrus SR20 flight manual I have does not leave me with that impression, indeed, somewhat to the contrary...

I think the view has always been pretty clear.

You would have to be very certain of a successful forced landing not to deploy the chute.

Here is the rational.

With the chute you will land vertically at around 20 mph. The u/c and compression panels beneath the seats are all designed to absorb energy and the energy that is transmitted to the pax is in the vertical plane - the precise plane the spine and the rest of the body is best able to absorb the residual energy.

Compare this with a conventional forced landing. If you are lucky you might impact with a forward velocity of around 60 mph (it could range by maybe -10mph but could equally exceed +30mph) which at best means you are carrying nearly ten times as much energy and the vector almost certainly will be in the wrong plane when it comes. The energy might dissipate slowly if you are lucky, but then again if you are unlucky the nose might dig in and the aircraft flip, with all sorts of other possibilities.

Which option would you take?

For these reasons the current thinking is you would have to have a good reason for not using the chute. The insurance companies share that view and favour the use of the chute if the engine quits.

In reality if you review the majority of chute deployments the aircrafts final resting place has been further cushioned by trees, shrubs and other elements of the terrain. While usually the damage to the aircraft is in fact relatively light and repairable, usually the aircraft is not repaired. The same would be true of an aircraft involved in a FL unless there was no serious structural damage. You would be lucky to land a Cirrus in most fields and find the u/c intact.

[Chuck Ellsworth]

Chuck:

in an earlier post you posted that you had a double engine failure in a Piper Navajo, due to the fuel lines being blocked. What casued the blockage ?

Ice crystals.

[SkyHawk-N]

kala87 said ...

3. By all means slip the a/c to lose height on final, but be careful with the flap limiting speed. On one PFL I had flap 40 selected on a C172, tried a quite aggresive slip and quickly found the speed was 10 kts above the flap limiting speed

What does the POH for the 172 in question say about slipping the aircraft with 40 degree flaps set?

[24Carrot]

Chuck, you are a one-man statistical data source!

As a rough estimate, how many times did you decide not to fly at all because of pre-flight doubts about the engine?

[Chuck Ellsworth]

Chuck, you are a one-man statistical data source!

All I can do is relate my own experiences with flying.

As a rough estimate, how many times did you decide not to fly at all because of pre-flight doubts about the engine?

Just off the top of my head I can think of three engine problems I had that were enough concern to persuade me not to fly them until the cause was found.

The first time was with an airplane that had had an engine failure and the crew had to use METO ( Maximum except take off. ) power to get back to the airport.

They changed the failed engine and I did the next trip on the airplane, on descent the other engine started to backfire as power was reduced at the destination and also when returning to the departure airport ( Goose Bay ).

The decision was made to change the AMC in the carburetor ( Automatic mixture control unit. ).

On the test flight it had the same problem and we started to look deeper into the issue, I happened to be coming back from getting some stuff for the job and they were running it up and listening to the sound from a distance I had a hunch what was wrong, cracked cylinder ? 's.

We did a leak down test and there were four cylinders below specs.

We changed four cylinders and on the test flight I was looking at the engine as the power was reduced from climb power to cruise on the down wind leg...as I watched a cylinder let go and went right up through the cowling, I just hit the feather button and then followed the feather shut down check list.

The engine as I had suspected was junk and my decision not to fly it because of the way it was acting was correct.

That was the first, should I use more bandwidth and relate the other two?

[douglas.lindsay]

Always interesting to hear the tales - particularly the inexperienced of us like me... go on...

[Big Pistons Forever]

Originally Posted by bjornhall

The usual error of trying to draw conclusions from statistics using only mean values. One needs higher order moments as well in order to draw any kind of meaningful conclusions from the data itself (at the very least some variance measurement). If one were to do that, I guess one would find that the variation among pilots is so high that the mean values are quite pointless.

The other big problem is that the sample sizes are too large. People try to create large sample sizes by grouping together all sorts of pilots undertaking all sorts of operations in some belief that large sample sizes make the statistics better. What one ends up with is a fairly accurate value of something nobody is interested in measuring. If I want to know, say, my probabilities of having an engine failure, I want a sample consisting of pilots similar to me doing similar things in similar planes. The proper sample size is one individual, and the result will not be available until my career is over!

The third rather obvious point is that engine failures are not random events. An engine fails for a reason. If one has a perfectly good, healthy engine that is handled properly, the chances of it failing is close to zero. If there is a fatigue crack in a rod with a remaining life of some 3000 cycles, the chances of the engine failing in the next five minutes is about 100%. Since one does not know if one belongs to the former or the latter cathegory, one fundamentally has no idea of the probability of having an engine failure during that flight. Statistics is useful for all sorts of things, but not for making predictions of individual events.

I agree that one has to be carefull not to attribute too much accuracy to a statistical value but I think the accident trends provide usefull information. The 80% number came from light aircraft that had an engine failure and had to execute a off airport forced landing or were damanged. I think it is safe to say that the majority of light aircraft hours flown are either as training or as a private flight. The fact is that the majority of crumpled airplanes sitting in a field got there because they ran out of gas or let carb ice stop the engine. Very few had an engine failure because of a "fatigue crack in the connecting rod", or any other catastrophic internal engine failure. To reduce risk one must be able to assign an appropriate level of risk to possible events. I think the engine failure statistic is of value in reminding all pilots that they are in direct control of the major risk factors that are the most likely to cause a pilot to experience an engine failure.

I suppose if you are convinced that you will always takeoff with sufficent uncontaminaded gas with the fuel selector always correctly set and never let carb ice develop then the accident statistics that show many pilots have in the past failed to take those steps and suffered the resultant engine failure...... you would feel that the information is of no value to you. Obviously each pilot must decide for themselves how much weight to put on the statistics and whether or not the trends that they may reveal would prompt a reexamination of how they operate their aircraft.

BTW of the private light aircraft engine failure accidents that I have personal knowledge of and which resulted in off airport landings with the aircraft substantially damaged/destroyed; 2 were "ran out of gas"(C150,C172), one was the result of leaving the fuel selector on one tank untill it ran dry, even though there was fuel in the other tank (C172) 1 one was carb ice (C150)and the last one ironically was a fatigue failure of a connecting rod (C172).

[24Carrot]

That was the first, should I use more bandwidth and relate the other two?

Yes please.

Always interesting to hear the tales - particularly the inexperienced of us like me... go on...

For the same reason!

[Contacttower]

What does the POH for the 172 in question say about slipping the aircraft with 40 degree flaps set?

I only have the C182's in front of me but to quote (and this would probably be the same for the C172) "Steep slips with flaps settings greater than 20 degrees can cause a slight tendency for the elevator to oscillate under certain combinations of airspeed, sideslip angle, and center of gravity loadings".

[Chuck Ellsworth]

The second time I had reason not to fly an airplane with an engine problem that I felt unsure about was in Cape Town where I was giving type ratings to the owner and also a good friend of his Glen Dell who is now with the Red Bull air Racing Team.

We had previously been doing training in J/berg but due to the altitude we decided to go to Cape Town to finish the water work portion of the training as Cape Town is at sea level.

The problem first showed up one morning when I noticed the oil temperature started to climb and very soon went into the red zone which resulted in us returning to Cape Town airport.

After two weeks of trouble shooting which included not only changing the oil cooler and all the oil lines we even switched oil coolers from one side to another we decided to start the ferry flight to Oshkosh and hope the problem did not get worse...the temperatures were not high enough to have the over temp manifest itself and the owner wanted the airplane to get to Oshkosh if possible for the air show.

I told him I was almost certain that once I got in the higher temperature areas the problem would return.....I also put forward the possibility that the problem was related to a partially plugged oil galley in the engine caused by changing from mineral oil to detergent oil after 450 hours of running on mineral oil.

The engineers in Africa and the engine overhaul shop were doubtful about my diagnosis so I agreed to do my best to get it to Oshkosh.

All went well until we landed in Djibouti where the temperature was 45 c when we landed and the oil temp started to rise on the approach and landing. After a couple of days going through the usual problems associated with trying to get anything done in that part of Africa we finally taxied out for take off just as the sun was rising and the temperature was still low...for Djibouti, around 35 C.

I managed to finally get the airplane nursed up to nine thousand feet on the good engine and using the sick one at partial power to control the oil temperature where the air was cool enough to use cruise power and still have a safe oil temperature.

Our next fuel stop was Jeddah where the trip ended because we had to shut it down just after take off due to a run away oil temp problem.

Seven months and two trips back to Saudi Arabia due to running out of time on our visas we finally finished the engine change and ferried the airplane to England where it stayed for two years and was used flying for Miramax before I finally ferried it to Suffolk Virginia where it is now owned by this guy.

There is a good story on the history of that PBY in his web site for you keeners who like reading about airplanes.

Fighter Factory - WWII Aircraft Recovery and Restoration

By the way when the engine was opened and examined during the rebuild I had been right about the cause...there was a partially plugged oil galley leading into the nose gear reduction case that caused the oil temp to run away in hot weather.

Will relate the third engine I was reluctant to fly, which also turned into a real expensive problem for the company who owned the airplane when I get some more time to peck away on this keyboard.

[Pilot DAR]

the vertical plane - the precise plane the spine and the rest of the body is best able to absorb the residual energy.

I am not a medical professional, and will certainly defer to such a person on this, however, until then, I very much disagree with this idea. I accept that there are crush panels in the Cirrus, which will absorb a vertical load, but even so, I would far rather ondergo a 60 MPH deceleration across the ground over many feet, perhaps hundreds if I do it right, than a vertical deceleration at 20 MPH in no more than about 4? feet! (unless a tree cooperates).

A five, four, or at worst three point harness will distribute the decelertion forces widely across the torso if it is a forward or nearly so force. Add to that, an injury resulting from deceleration in that direction tends to be less severe, relative to the "G" force. A vertical deceleration up the spine concentrates all the force up the spine only, and no muscles or tendons can play a part in absorbing that force, as they would were it to be forward into a harness. It's entirely weight on the spine. And, if you do an injury up the spine, it will be much more serious. I expect that experts on ejection from aircraft would have a lot to contribute on this.

Until then, when I'm flying a Cirrus, deployment of the 'chute will firstly be in accordance with the flight manual only, and thereafter, only if I deem the situation life threatening. If an engine failure in this type of aircraft is life threatening, I must have already pushed the limits really far somehow!

I have force landed four times, and (very lucky) not only have I not damaged an aircraft doing it, each time, I was able to fly it out very safely, once the problem was resolved. Though at times I have been required to wear a 'chute, I have never wanted to have one, or considered using it!

[IO540]

I would far rather ondergo a 60 MPH deceleration across the ground over many feet, perhaps hundreds if I do it right, than a vertical deceleration at 20 MPH in no more than about 4? feet!

I agree.

Experiments done by the USAF with rocket powered sleds in the 1950s involved a forward deceleration of 80+G. At the top end, the subject ended up with a detached retina but no broken bones IIRC.

A vertical acceleration is just going to b*gger up your spine so you end up in a wheelchair, and an excessive value is going to snap your pelvis, which is not great either.

I am not a doctor but give me a field to plough any day.

I would still like a chute for

- engine loss over forest
- engine loss over mountains
- structural failure

[VMC-on-top]

I'm no expert but would imagine that the G's from any Martin Baker ejection seat get the pilot out a hell of a lot faster than 20mph - and I understand the maximum number of times a pilot can eject is twice because of compression damage to the spine. So I'm pretty sure that landing vertically at 20mph is going to be considerably less damaging to the spine than (a) ejecting from a MB seat and (b) hitting something solid at 60mph.

So, given the the option, I'd be reaching for the chute every time!

[24Carrot]

I would far rather ondergo a 60 MPH deceleration across the ground over many feet, perhaps hundreds if I do it right, than a vertical deceleration at 20 MPH in no more than about 4? feet!

We can compare them. The vertical 20mph to zero in 4 feet is 3.36G. The horizontal 60mph to zero at 3.36G takes 36 feet. 60mph to zero in 4ft (stone wall scenario) is 30G.

speed squared = 2 times 32 times G times distance

e.g. (20x5280/3600)sqd = 2x32xGx4

As said already, tolerance of vertical and horizontal G will vary.

[Fuji Abound]

I am also no expert but I have taken an interest in this matter.

There are a number of factors that will determine the outcome of a forced landing. However, reading the accident reports, it strikes me that with conventional forced landings the single commonality is the lack of predictability. That makes sense to me. I suspect we all think about forced landings in the relatively sterile training environment. In reality there is every chance the engine will probably fail when we are least expecting it to do so.

From my limited experience, discussions with those who have had engines fail, and from reading the accident reports a lot of engine failures are at relatively low levels. Statistically this is not surprising because so many pilots usually operate between 1,500 and 3,000 feet. Catastrophic engine failure is often, not surprisingly, associated with some sole searching. For example, a pilot I know had oil all over the screen and almost no forward visibility. There are a host of reasons why in advance of a failure and for a time after the failure, it is all too easy for the pilot’s vision to have tunnelled, such that he has lost some of his situational awareness. He must now prepare for the landing and select the best landing site as well as maintain control of the aircraft. Again this is barely representative of the normal sterile training environment in which the pilot is expecting a PFL, has thought about wind, has rehearsed the vital pre-landing actions and, as luck has it, is almost certainly over a area that will present some reasonable landing opportunities.

To put it simply you are lucky if a forced landing occurs in ideal circumstances and you are truly current if you are able to totally avoid tunnel vision and correctly prioritise the vital tasks. I would suggest that is not representative of the average GA pilot flying less than 50 hours a year, and may not be representative of many pilots flying a lot more hours.

Moving on, the wind has been assessed, although perhaps not that well or without adequate thought to the variation in direction or strength at the bottom of the valley with which we are confronted, and a field has been selected.

Everything now depends on maintaining control and arriving where we intend. Over the years I have enjoyed taking part in spot landing competitions. At some you can stand with the other onlookers almost in the box. It is interesting to see how many pilots new to the game fail to even get there wheels in the box – and that is with the availability of power. Inevitably some types are much more difficult to spot land accurately – in fact the very types many pilots commonly fly.

We can all recall the tips we were given about field selection – and they are in the main sound. However find me a field that turns out the way you would have wished form 2,000, by which many pilots will have committed to the landing site. I think we all realise that there are many hazards which we may not have anticipated.

As to the landing, yes of course in ideal circumstances we slide across Bowling Green grass in an exactly horizontal plane until eventually the aircraft comes to rest of its own accord. Yeah, right. The reality is likely to be different, and it is likely to involve some degree of impact damage involving both horizontal and torsional forces some of which may be high. Both the aircraft and its occupants are ill prepared for these forces. We all understand that very little thought was given to the design of the vast majority of light aircraft to energy absorption. How many light aircraft have air bags, crumple zones, roll bars and don’t have all sorts of sharp and protruding panel work almost designed to impale or prevent your egress?

A reasonably sound resume of the effect of deceleration on the human bodies appears here:

http://ftp.rta.nato.int/public/PubFullText/RTO/EN/RTO-EN-HFM-113/EN-HFM-113-03.pdf

and is worth reading. If anyone is interested I have more detailed material. We believe the body is better able to withstand axial loads than horizontal rotational loads which are more than likely in a typical forced landing (other than the greaser along the apocryphal Bowling Green). An analysis of rotary accidents suggests spinal injury is no worse than in fixed wing accidents, surprising in itself given that in the typical rotary accident the vertical rate of descent and the absence of energy absorbing technology results in significantly greater loads being transmitted to the pax.

IO I am aware of the tests to which you refer. However with respect I cant imagine they are representaive of the impact loads in a typical forced landing because the aircraft is unlikley to be running along the ground in a single plane but more likely to be subjected to some "tumbling". Moreover how many pilots find themselves tightly strapped to their seats by means of a "proper" five point ratchet harness.

So, to sum up, for the average pilot, who at best is likely to find himself dealing with his worst nightmare while lacking in the currency to achieve better than a pedestrian forced landing an engine failure is a real threat to his and his passengers. For the pilot who considers himself current, he may still need a healthy dose of luck or live to regret as he descends through IMC into the mountain valley painted on his GPS, that the terrain at the bottom of the valley was at least as good as he hoped and the winds aloft had not turned into a stonking 20mph tail wind.

I don’t want this to sound like a Cirrus PR campaign, because it isn’t. However it does reflect my thoughts on why ballistic parachutes and cockpit design are vital as a means of improving on average the survivability of a forced landing. Yes, with luck the outcome of many forced landings is very good. However, I would rather we take luck from the equation. By doing so that means we give the average pilot experiencing an engine failure above the sort of terrain and in the type of conditions we could reasonable expect him to be operating the best chances of survival and the least injury.

I also think for an aircraft with a BP I am comforted by the predictable arrival this should ensure. I know with almost certainty that I will land with a vertical velocity of less than 20 mph. I know that the energy absorbing material will do its job. I know that it is very unlikely any of the forward facing elements of the cockpit will impact on me, or hinder my egress, I know there will not be any surprises that I hadn’t spotted when I turn final, importantly I know that I have got maximum time during the descent to organise myself and my pax to best improve their chances. For all of these reasons I think a chute is almost always the best option. I think the injury record in every successful Cirrus deployment speaks for itself.

Please persuade me otherwise – I would be truly interested.

At the moment I will take the predictable 3G any time.

[VMC-on-top]

We can compare them. The vertical 20mph to zero in 4 feet is 3.36G. The horizontal 60mph to zero at 3.36G takes 36 feet. 60mph to zero in 4ft (stone wall scenario) is 30G.

speed squared = 2 times 32 times G times distance

e.g. (20x5280/3600)sqd = 2x32xGx4

As said already, tolerance of vertical and horizontal G will vary.

Ejection seat - Wikipedia, the free encyclopedia

"The pilot typically experiences an acceleration of about 12–14 g (117–137 m/s²)"

- and the purpose is for them to walk away from the scene, so 3g is more like the forces experienced on a half decent rollercoaster!

[172driver]

At the moment I will take the predictable 3G any time.

The chap in this thread would certainly agree!

On a different note: I once read (cannot find the reference any more) that about 70-80% of engine failures - or least the EFATO variety - occur at the first change in power setting after t/o. This comes from some FAA statistic and it certainly impressed me enough to avoid changes to the power setting until at about 1000' (when and where possible). Can someone confirm?

[24Carrot]

Fuji, thank you for a very informative post.

One very tiny quibble, the R22 helicopter does have energy absorbing bendable skids and crushable seat bottoms.

[Fuji Abound]

24carrot - no thank you for taking the time to read my ramblings.

I wasnt aware that was the case although you would expect it to be so with more "modern" civil designs. I dont recall the report to which I gave the link mentions the specific types included in their study.

[Big Pistons Forever]

Originally Posted by 172driver

The chap in this thread would certainly agree!

On a different note: I once read (cannot find the reference any more) that about 70-80% of engine failures - or least the EFATO variety - occur at the first change in power setting after t/o. This comes from some FAA statistic and it certainly impressed me enough to avoid changes to the power setting until at about 1000' (when and where possible). Can someone confirm?

This notion came from a study of EFATO failures of large radial engines. It has to do with the change in loadings of the nose case reduction gear and or valve train. The theory being that the change of power and especially RPM alters the forces inside the engine which could trigger a harmfull vibration or imbalance in an allready failing engine and lead to the engine self destruction. Therefor it became a realtively common practice to not make early power reduction in these engines. However since almost all light aircraft have simple direct drive lightly loaded engines the factors that are an issue for the big radials do not exist. I read a short note on this issue from one of the engine makers (lycoming I believe) which stated that if the engine made full power and was running smoothly at the start of the takeoff run it was very unlikely to suffer an internal mechanical failure at the first power reduction.

However in a take off of a single engine aircraft, altitude is your friend so I think a full power climb to a 1000 feet at Vy is a good idea to quickly get some altitude and therefore have more options in the event of an EFATO