another Cirrus engine failure. Another chute saves

[GRG55]

A buddy of mine had an engine failure in a Cirrus a few years ago. The engine-driven fuel pump failed. Cirrus said it was the first such failure in their records (or at least that's what I seem to recall). When it happened, my buddy turned on the electric fuel pump, but no restart. IIRC, he found out later that the electric fuel pump will not supply enough fuel for the engine to run without the engine-driven pump. This was news to my friend. So much for redundancy. Oh well, it's got a chute .

...

If this was actually the case (and I am highly sceptical), what then would be the purpose of having an electric fuel pump at all in the Cirrus?

 

[flydigital]

In recent Cirrus training I was taught the same: that the electric fuel pump is not sufficient to keep fuel flow adequate in the event of engine-driven fuel pump failure. Also peculiar, the engine-driven fuel pump needs the electric "boost" to fight vapor lock. Was told stories of engine stopping or sputtering when the boost pump went off in warmer conditions. Here in the Socal desert during summer, we leave the pump on at all times, and in the winter, are instructed to turn it off only after 15 minutes of cruise.

 

[Scott@KTYR]

Never mind. Here ya go:

"Federal Aviation Administration studies indicate that piston engines in aircraft have a failure rate, on average, of one every 3,200 flight hours while turbine engines have a failure rate of one per 375,000 flight hours. Accordingly, for every turbine engine experiencing a failure, 117 piston engines will have failed."

Yep that is the reason my friend made the switch to a Turbine.

 

[Hank S]

In recent Cirrus training I was taught the same: that the electric fuel pump is not sufficient to keep fuel flow adequate in the event of engine-driven fuel pump failure. Also peculiar, the engine-driven fuel pump needs the electric "boost" to fight vapor lock. Was told stories of engine stopping or sputtering when the boost pump went off in warmer conditions. Here in the Socal desert during summer, we leave the pump on at all times, and in the winter, are instructed to turn it off only after 15 minutes of cruise.

Seems like a good reason to skip the Cirri family. Pull the cord for fuel pump failure? YGBSM!!

 

[Acrodustertoo]

Seems like a good reason to skip the Cirri family. Pull the cord for fuel pump failure? YGBSM!!

My thoughts as well.
Akin to having a spare tire for your car that you cannot drive on, sure it will hold your car up but won't get you anywhere.

 

[GRG55]

In recent Cirrus training I was taught the same: that the electric fuel pump is not sufficient to keep fuel flow adequate in the event of engine-driven fuel pump failure. Also peculiar, the engine-driven fuel pump needs the electric "boost" to fight vapor lock. Was told stories of engine stopping or sputtering when the boost pump went off in warmer conditions. Here in the Socal desert during summer, we leave the pump on at all times, and in the winter, are instructed to turn it off only after 15 minutes of cruise.

There is only one engine driven fuel pump. Given it's a mechanical device (cam driven diaphram pump I expect?). Magnetos are also mechanical devices, prone to occasional failure in service. But unlike a magneto, which is normally paired for redundancy, I am to understand the mechanical fuel pump on a Cirrus does not have an adequate backup sufficient to keep an otherwise perfectly good airplane in the air? That would seem a rather disturbing deficiency. Good thing its got an airframe parachute...

 

[FastEddieB]

When I found out the boost pump could not keep the engine running in a Cirrus, I too thought it was a poor design.

The rationale is that the design of the engine-driven pump is such that total failure was highly unlikely.

Or so I was told.

 

[hindsight2020]

Boy I can feel the bern around he'a!

 

[Jay Honeck]

When I found out the boost pump could not keep the engine running in a Cirrus, I too thought it was a poor design.

The rationale is that the design of the engine-driven pump is such that total failure was highly unlikely.

Or so I was told.

None of this makes sense.

Why even install an electric fuel pump if its output is insufficient to keep the engine running? Surely this is baloney...

 

[FastEddieB]

Why even install an electric fuel pump if its output is insufficient to keep the engine running?

From the SR22T POH: "The electric fuel pump is provided for engine priming and vapor suppression."

 

[hindsight2020]

Reading these responses is like a real-time illustration of the 12 stages of grief ... we're due for anger any minute now.... LOL

 

[Dennis McKim]

Reading these responses is like a real-time illustration of the 12 stages of grief ... we're due for anger any minute now.... LOL

Not that it matters in this context, but anger is the second stage of five, in Dr. Kubler-Ross' stages of grieving. DABDA. Denial, anger, bargaining, depression, and acceptance.

 

[PaulS]

When I found out the boost pump could not keep the engine running in a Cirrus, I too thought it was a poor design.

The rationale is that the design of the engine-driven pump is such that total failure was highly unlikely.

Or so I was told.

Those are famous last words... everything fails sooner or later, sometimes at the most inopportune moment.

 

[GRG55]

Those are famous last words... everything fails sooner or later, sometimes at the most inopportune moment.

Like somewhere between rotation and minimum 'shoot' deployment altitude

 

[FastEddieB]

Those are famous last words... everything fails sooner or later, sometimes at the most inopportune moment.

True. And stipulated redundancy is desirable.

Yet, to be fair, many components of aircraft and aircraft engines soldier on without redundancy.

For instance, oil pump, carburetor, prop governor, and on and on.

Not defending Cirrus in its design choice - just pointing out it's hardly unique.

 

[PaulS]

True. And stipulated redundancy is desirable.

Yet, to be fair, many components of aircraft and aircraft engines soldier on without redundancy.

For instance, oil pump, carburetor, prop governor, and on and on.

Not defending Cirrus in its design choice - just pointing out it's hardly unique.

True, but the fuel pump, like the magneto, is critical and can cause instant loss of power when they fail. The other items you listed generally don't cause instant power loss on failure. I'm just surprised the pump isn't big enough to take over if the main pump fails.

 

[pburger]

Several have asked why have an electric fuel pump if it isn't sufficient to keep the engine running. The obvious answer is to get fuel to the engine to start it. Once it is running, it can rely on the engine-driven fuel pump (or so you'd think). However, you'd sure think that if you're going to put in a separate fuel pump, you would design it to keep the engine running if the engine-driven pump failed. Hello? I ASSUME that the fuel pump in my Arrow will keep the engine running if the engine driven pump fails. I don't plan to test that theory, but I have always assumed that, and I think they state that in the literature. My Cirrus buddy assumed the same for his Cirrus - but he found out after the fact that it won't keep the engine running.

I am very surprised to read that the engine-driven pump may need the electric pump to STAY RUNNING at times. That creeps me out. I guess the answer to everything is that "it has a chute."

 

[denverpilot]

I ran a quickie analysis (just searched for "Engine" in the Probable Cause) of Cirrus engine failures. I omitted those that were pure pilot error (fuel exhaustion/starvation), and trimmed a PC or two to eliminate information not strictly related to the engine.

I'll try run a similar analysis on another high-performance type later.

Ron Wanttaja

A good number of these is failure to properly torque through bolts on the engine. Pretty scary when we are now faced with FAA ADs that require arguably unnecessary engine disassembly. It will be interesting to see how many of these engines result in more failures than the AD was supposed to be addressing. It goes from a single cylinder failure to a catastrophic complete engine loss.

I was also noticing how many mechanics the FAA says don't know how to run a torque wrench... Interesting.

just how I was taught. 500 feet, flaps up, touch the handle and note CAPS alt. 1000 ft, fuel pump off, bring the mixture to the top of the green arc. That was what I was taught and it seems every bit of it is debatable.

Ron, you are a treasure in this community.

If you touch something on the left side of the panel, then the right, then the handle, it'd be kinda Catholic.

Never mind. Here ya go:

"Federal Aviation Administration studies indicate that piston engines in aircraft have a failure rate, on average, of one every 3,200 flight hours while turbine engines have a failure rate of one per 375,000 flight hours. Accordingly, for every turbine engine experiencing a failure, 117 piston engines will have failed."

That's an impressively big difference. Wow.

 

[FastEddieB]

I am very surprised to read that the engine-driven pump may need the electric pump to STAY RUNNING at times. That creeps me out. I guess the answer to everything is that "it has a chute."

To be fair, I've flown a fair amount of fuel injected planes where the electric boost pump may be required to avoid vapor issues. That's certainly not unique to the Cirrus.

 

[hindsight2020]

Several have asked why have an electric fuel pump if it isn't sufficient to keep the engine running. The obvious answer is to get fuel to the engine to start it. Once it is running, it can rely on the engine-driven fuel pump (or so you'd think). However, you'd sure think that if you're going to put in a separate fuel pump, you would design it to keep the engine running if the engine-driven pump failed. Hello? I ASSUME that the fuel pump in my Arrow will keep the engine running if the engine driven pump fails. I don't plan to test that theory, but I have always assumed that, and I think they state that in the literature. My Cirrus buddy assumed the same for his Cirrus - but he found out after the fact that it won't keep the engine running.

I am very surprised to read that the engine-driven pump may need the electric pump to STAY RUNNING at times. That creeps me out. I guess the answer to everything is that "it has a chute."

I'm an Arrow owner. It's not an assumption that the electric pump can keep the PA-28 line (from carbed O-320, to injected 360, to carbed 540) fed at max power. It's a fact, by design.
And it is stated in the POH. And yes, one of the many reasons I prefer Lycoming installations over Contis.

Everything is a compromise, but a lack of fuel pump pressure redundancy in a certified airplane, that's just a bridge too far for me. To be fair, I draw the same line with siamese mag'ed Lycos.

 

[denverpilot]

I'm kinda partial to gravity feeds. But if you're dying to put the tanks below the engine because you put the wing on wrong, ok... I'll flip on the electric pumps.

 

[PaulS]

I don't think the fuel injection systems work too well with gravity feed. There has to be a reason beside cost for not having an electric pump that can keep the engine running.

 

[Stingray Don]

I'm kinda partial to gravity feeds. But if you're dying to put the tanks below the engine because you put the wing on wrong, ok... I'll flip on the electric pumps.

Yep, I've yet to have my first gravity failure!

 

[Hank S]

I'm an Arrow owner. It's not an assumption that the electric pump can keep the PA-28 line (from carbed O-320, to injected 360, to carbed 540) fed at max power. It's a fact, by design.
And it is stated in the POH. And yes, one of the many reasons I prefer Lycoming installations over Contis.

Everything is a compromise, but a lack of fuel pump pressure redundancy in a certified airplane, that's just a bridge too far for me. To be fair, I draw the same line with siamese mag'ed Lycos.

My electric pump will also keep me in the air should the mechanical one quit. That's the whole reason to have the electric one on for takeoff and landing.

Having a $750,000 plane that is totaled because the $1000 fuel pump craps out is just mind boggling. Guess that's just another reason Cirrus put a chute in, in case something breaks and the backup system isn't capable of carrying the load. What's it got for mags, alternators and batteries? What other single thing can break and require popping the chute?

 

[Challenged]

If the engine driven pump dies, you just fly inverted.. simple fix.

 

[Dav8or]

To be fair, I've flown a fair amount of fuel injected planes where the electric boost pump may be required to avoid vapor issues. That's certainly not unique to the Cirrus.

It is very rare, but I have experienced this in my fuel injected Mooney. Climb to cruise altitude with electric pump off, then level off, reduce power and the engine stumbles and begins to die. Turning the electric pump on fixes it immediately. It hasn't happened in a couple of years now, but it has happened.

 

[MetalCloud]

So is it possible to put in a different / better fuel pump where this is not a perceived problem?

 

[RudyP]

My electric pump will also keep me in the air should the mechanical one quit. That's the whole reason to have the electric one on for takeoff and landing.

Having a $750,000 plane that is totaled because the $1000 fuel pump craps out is just mind boggling. Guess that's just another reason Cirrus put a chute in, in case something breaks and the backup system isn't capable of carrying the load. What's it got for mags, alternators and batteries? What other single thing can break and require popping the chute?

I wish it was $1000 - having bought a new one, it is actually closer to $8K. I too would like more redundancy in the fuel system but there are 3 reasons why Cirrus did not build it that way:
1) Full mechanical pump failure is exceedingly rare. In fact I was told it has never happened on a SR22 and led to an accident (at least a confirmed failure). Apparently someone in this thread has heard of one case so perhaps it isn't none any more but it is vanishingly rare when you consider that 6000+ of these planes are flying and most of them 200+ hours per year and you NEVER hear of a catastrophic mechanical pump failure
2) Since the electric and mechanical pumps are inline, a bypass would have to be designed in the event that the mech pump locks up and restricts flow, no matter how strong the electric pump is - not impossible, but a consideration
3) Cirrus thought the risk of accidentally turning on high boost in the wrong power regime and flooding the engine (e.g., on short final) outweighed the very few instances where the main pump fails and the electric would save the day. People HAVE died killing a perfectly running engine by flipping on the high boost setting.

Again, I'm in the camp that would like a real backup - with appropriate engineering and safety backups but I don't see it quite as black or white as others here. You may disagree with Cirrus' choice, but they didn't just not think of it, they made a conscious decision.

And, finally, since I mentioned that these pumps very rarely fail and yet also mentioned Ive bought a new mechanical pump for my plane, I should clarify that they do fail sometimes to operate to spec which leads to engine management and performance issues but not outright engine failure. Mine was not able to deliver the fuel pressure spec and fluctuated more than I liked. And, of course, no mechanical system is perfect - just because it tends not to catastrophically fail doesn't mean it absolutely can't.

 

[FastEddieB]

2) Since the electric and mechanical pumps are inline...

Is there a reason for that? Seems like in parallel would make more sense.

FWIW, I just looked and my Sky Arrow pumps are also inline.

As an aside, things like $8k fuel pumps helped convince me to sell my Cirrus, though to be fair that's hardly limited to Cirrus and likely not a Cirrus part. $172.99 for a ROTAX fuel pump, for comparison.

 

[ern]

I'm an Arrow owner. It's not an assumption that the electric pump can keep the PA-28 line (from carbed O-320, to injected 360, to carbed 540) fed at max power. It's a fact, by design.
And it is stated in the POH. And yes, one of the many reasons I prefer Lycoming installations over Contis.

Everything is a compromise, but a lack of fuel pump pressure redundancy in a certified airplane, that's just a bridge too far for me. To be fair, I draw the same line with siamese mag'ed Lycos.

 

[Dan Thomas]

2) Since the electric and mechanical pumps are inline, a bypass would have to be designed in the event that the mech pump locks up and restricts flow, no matter how strong the electric pump is - not impossible, but a consideration

There IS a bypass built into the mechanical pump.

 

[ern]

hello i have been reading the posts about mechanical fuel pumps on the cirrus

if you look at the record of the mechanical fuel pump you would be amazed at the failure rate it is like getting hit by .lightning

now columbia aircraft uses the same pump and they do have a backup electric fuel pump that will run the engine in the event the mechanical fuel pump fails

there is a solenoid in the columbia 400 that if the fuel pressure drops below a certain num,ber the solenoid flips and turns on a electric fuel pump they never had to use it once.

cirrus looked at the failure rate of the main engione driven pump and decided i was not worth the back up electric pump

 

[Racerx]

Always the new guys with single digit posts reviving the dead.

 

[Tantalum]

Always the new guys with single digit posts reviving the dead.

Would be nice if there was a checkbox, or some reminder, alert (if you will), for the would-be-poster prompting them that the thread is X years old and it should only be revived if they have salient and relevant details..

A good example would be an accident thread, and someone posting the final NTSB findings years later.. for instance

/S


..on a more serious note, maybe it would be a good idea to only allow established members with a post count minimum to pull a Lazarus, just like they do with links, etc.

 

[bflynn]

"Federal Aviation Administration studies indicate that piston engines in aircraft have a failure rate, on average, of one every 3,200 flight hours while turbine engines have a failure rate of one per 375,000 flight hours. Accordingly, for every turbine engine experiencing a failure, 117 piston engines will have failed."

weird. Our club flies between 6000-8000 hours per year. According to that stat, we should have around 2 engine failure every year. In 10 years, the only one I know of was a pilot running out of gas.

 

[RyanB]

Our club flies between 6000-8000 hours per year.

Dang, really? What kind of club are you in?

 

[overdrive148]

 

[Tantalum]

*I'm not sure that's how stats work.. a birth control pill is 99% effective. Doesn't mean there's one pregnancy for every 100.. umm.. "events"

 

[Racerx]

*I'm not sure that's how stats work.. a birth control pill is 99% effective. Doesn't mean there's one pregnancy for every 100.. umm.. "events"

There's lies. There's damn lies. Then there's statistics.

 

[wanttaja]

There's lies. There's damn lies. Then there's statistics.

“Statistics are like bikinis. What they reveal is interesting. What they conceal is vital!”
- Professor Aaron Levenstein

Ron Wanttaja