GA Accidents-Fuel

[Yetti]

On 6/23/2016 at 3:14 PM, fuellevel said:

You need an intersection in your venn diagram.

 

The guy had to put it down because he misjudged one of the factors in the Time in air = gallons in the tank / fuel burn

[fuellevel]

Skill and Procedure - Planning 

Just got off the phone with the pilot 

The FAA inspector ran the numbers and said he should have 13 gallons in the tank - the pilot's own plan showed twelve 

No joy there 

-----------------------------------

What is really infuriating - is that if the pilot wanted to find out what caused the fuel loss - he would have to pay the fee for the teardown.  

A determination of cause had been made by the inspector when he arrived  even after he ran the numbers. 

You can't find what you aren't looking for 

[Yetti]

Does anyone know the actual fuel burn?  Leaking injector, fuel servo gone south

 

 

[fuellevel]

See above - It would have cost the owner to find the determination.

No totalizer onboard.   Rock solid gauges - little movement in turbulence

but accurate reporting on prior flights.  

[carusoam]

" we cling to the opinion most fuel accidents are a pilot issue.  "

Fuel accidents are an issue. the pilot has to come to grips with it before it can occur.

1) How does a better fuel gauge  solve the problem the accident pilot had?  The guy was running out of fuel no matter how it got measured.

2) He emptied one tank.  This leaves relying on the last single gauge.  If it is wrong, things may likely go quiet....

3) As 'good' pilots, we build in flexibility in our flight plan.  Example: We pad the t/o distance by 50% over calculated numbers.  With experience we can whittle this down some.  But the padding remains.

4) PMc is not alone in his thinking.  Each critical instrument in a Mooney has a back-up device.  If an instrument goes south, the PIC decides how he is going to handle it.  When he flies on to the final destination not knowing if he will get there, is he surprised when he comes up short?

5) Even my autos with crummy fuel gauges have a cross reference with miles driven on the tank.  (Yes, There were days when I couldn't afford to fill the tank).  My autos are 'fix as fail' devices.  Different than my plane.  One bad FL sender/gauge has run me out of fuel three times over the past two decades.  The tank is intentionally only keeping a few gallons in it for fuel stability reasons.  The first time the car was new to me. The second, on my way home from church. Third, on my way to the gas station after years sitting in the garage...  Fuel senders that stick may make you angry.  The kids walking home from church will be angrier.  and will remember longer...

5.5) in the 90s, There was also a half gallon of 'emergency' fuel in the trunk. Some high octane stuff that didn't evaporate well... Kind of like having Josés VW valve for emergency fuel...

6) With all of the extensive writing here, I have missed or forgotten part of the premise.  

Even on the last leg, without a FF meter the pilot doesn't know what rate he has been burning fuel at.  Having an accurate FL gauge can't help with that if he is planning on running out exactly on touch down.  How much padding did he use VFR day or IFR night or less?

7) always have two gauges working.  FF and FL, or two FLs.  When one is no longer working, the size of the padding needs to be adjusted.  

8) when the size of the required padding is larger than the fuel left in the tank, land immediately...

9) full throttle climb at full mixture uses a lot of fuel.  Low throttle descent doesn't save enough fuel...

10) relying on a single gauge for accuracy will leave you empty handed when it fails.

Scott P.,

Am I missing something?

I am thinking it may be a low cost alternative to add two CIES FL gauges in place of adding a single FF gauge to an existing plane that doesn't have one.

a really good idea if the FL is non-linear closer to the bottom of the tank...

I am interested in more accurate FL gauges to go with my FF gauge and totalizer.

Blaming the pilot for running out of fuel is like blaming the sales guy for not getting the sale.  Somebody is going to be directly responsible.  It's in their job description.  

Best regards,

-a-

[Yetti]

Nope - The intersection in the Venn diagram is Successful or unsuccessful flight or something like that

[fuellevel]

I have to redraft re-lable that as it really is pilot skill and plan difficulty  - Hamburger run - full tanks - no or low risk  - indication is a non factor 

Cross country - partial tanks - higher risk  - no indication higher risk 

 - and indication of any type (total or fuel tank amount) provides a warning or risk mitigation

The fault Tree supports that component  - a totalizer can intercept a few lines -

a gauge can intercept any line less it has failed and then it contributes by mis information to a potential harmful conclusion 

The gauge on the bottom indicates  it's mis information impact on the chain moving up   

 

carusoam  -  one tank emptied itself in excess of planned value, he nursed it to dry carefully watching fuel pressure, and then switched to the other tank  and watched that tank fall precipitously to zero.  (supports an issue downstream of the tank).

 He was lined up on the runway and saw the security fence and feared catching the fence and flipping over so he turned and landed in a fuel near the airport.  This is from my interview notes.    Remember -  military experience.     

So yes multiple methods of ascertaining critical information are important - trust but verify - Time in tanks is excellent backed with starting fuel value and a watching time and remaining fuel alonf the way.  One system is not a panacea or replacement for any of the others. 

[carusoam]

Spell check?  Fuel = field in above post...?

-a-

[fuellevel]

I am having a dialogue over accuracy and stability on COPA  and this discussion on Mooneyspace 

On one hand  - the accuracy deviations in flight of a gallon might be an issue 

Other hand   -  and this assessment of pilot responsibility  

Two sides to the Oreo - in the center is the good stuff 

[fuellevel]

carusoam

Doubling up is not necessary 

the Aerospace Logic gauge gives a trend function for fuel tank level over time 

(Fuel Flow)  You can see the accelerated flow at the beginning (Climb) followed by cruise 

The Bar graph will actually warn you to switch tanks based on fuel volume 

so if you get a fuel imbalance warning - and it is happening faster than you planned 

problem  

[fuellevel]

This was Cirrus forum discussion, so I animated it.

 3 years ago on COPA  - what is old is new  

[Bennett]

In my opinion, valid only to me, everything is the panel is little more than a tool. I want to know how to use the tools, and their limitations. I want lots of redundancy, and I think about lots of "what ifs" on any given flight; plan Bs, Cs, and Ds. I do the same in offshore sailing. Sure I would like accurate fuel gauges, and I might well replace the factory units, but in all candor, I pay more attention to fuel flow, fuel required, and fuel used as reported by JPI, but most importantly through very conservative trip planning for at least 20 gal in the tanks when I land. Hell. I've landed 50 miles short of my planned destination to take on more fuel rather than cutting into my personal reserve requirement. Sweating out low fuel is not something I want to do. I just don't see the point of risking life trying to stretch range. I do run LOP when I want to avoid an intermediate fuel stop, and this works well in some situations, but if the winds are not favorable to allow this, I will make the stop.

[fuellevel]

Bennett

The pilot in the example said so as well.  He used 13 gallons as his reserve in a Cherokee for this flight.  Having that experience he would now stop in Brooks County no matter what.   He said - now he would only fly 2.5 hrs effectively cutting range by a third.   His passenger was his son in law, the implication was impactful.  

AOPA Safety - recommends in some aircraft examples not getting fuel closer than an hour to your destination -  That person is now a Mooneyspace poster.  I saw the statistics

given, but can't remember the reasoning why. 

----------------------------

This was an example of a very poor accident investigation, that ends up as a statistic supporting a conclusion.  

I know enough now to suspect every fuel accident report I read and look for inconsistencies. 

A self perpetuating bias

[Bennett]

Why would a pilot not stop for fuel if he felt he was going into his reserve, and a suitable airport was reasonably close. My J has 64 gallons, and since I almost always fly alone, every departure is with full fuel - I personally check it. This leaves 44 gallons for my flight with my self imposed reserve of 20 gallons. My decisions as to fly fast ROP or fly longer LOP are a mixture of experience and conservative flight planning. LOP at under 10 gallons/hour vields at least 4 hours flying time, still leaving 20 gallons of reserve. Obviously I pay attention to time, fuel flow, (JPI 830), panel fuel gauges for what they are worth, at least one sight gauge can be seen from the cockpit, and being sensitive to way the aircraft is flying (heavy/light wing), TAS, winds aloft from XM, etc. After flying Mooneys for over three decades I believe that I can sense some aspects of the "rightness" or "wrongness" of some situations in flight. As I mentioned in an earlier post I have had fuel siphon in flight from an improperly seated fuel cap, which became apparent from both the panel gauges, and a seeing a visible sheen on the wing. Yes, I landed at the next suitable airport. I have also had a fuel drain leak in flight, although there was no leak seen after I sumped the tank before the flight. Maintaining more than the minimum prescribed fuel reserves simply makes sense to me. Need longer flight legs - the Monroe conversion, such as I had in my 261 allowed me to spend over 7 hours at a time in the cockpit (much younger then).

[fuellevel]

Bennett  -  He crossed that decision airport with sufficient reserve and to his plan.  

He had 2 1/4 more hours time in the tanks for less than 50 minutes of flight.

To your point - Plan C And Plan D were not evaluated.

 After I descend - all my fuel will be gone what next.   I wouldn't fault him.  

I remember a very hairy sea kayak crossing on Lake Michigan  - Fall day - fairly brisk wind and whitecaps we set out for a 20 mile crossing 

10 miles off shore,  We turned over and were swamped with a wave. 

I righted the boat and oriented the stern into the waves and directed the other person to start bailing after a few waves and after some heart lifting progress

 they lost the bailing bucket.   I directed us both to get back in and paddle out of danger, that went OK and I was making some progress with the pump

 until we hit the reef.   Standing waist deep in water and waves I made repairs, emptied the boat and busted my butt home shivering from the onset of hypothermia.

  As an offshore guy, I thought you'd appreciate that one, 

Some days - it just doesn't go your way.  I had plans,  gear, wearing a wetsuit, and I can still tell the story -

I have not made an open water crossing since.    

[carusoam]

Phillip,

You have really shown us something here today.  The graph of fuel used over time is pretty interesting.

It gives the impression that the fuel senders have been working steadily throughout the flight.  If they stopped working or became intermittent the graph would would have anomalies.

One type of leak would show up as a continuous fuel flow when the tank is not being used.

Another type of leak may show up as a higher than expected FF.  The FF gauge is showing one number, but the FL can see a higher number.  Fuel is leaving the tank but the leak is occurring before the FF sensor.

What types of leaks does this system actually notify the pilot of?

Bennet gave a couple of examples he has had in Mooney flight.  I have experienced some similar leaks. Dirt stuck in the drain.  Fuel evacuating out the top.  Flow problems caused by a stopped vent.

Fuel pumps have a tendency to leak when a bladder develops a tear.

Have you researched these types of failures and how your system (or others) can alert a pilot to the problem?

do you have a copy of various fuel system drawings for Moonies.  They come from various POHs that may be posted here.  They may give Fuel Pump models or types and other details...

Another day is going by...

Best regards,

-a-

 

[fuellevel]

Anthony - This is Scott  

Yes the first step,  is getting a quality fuel sender in the wing tank.  

• This is a monumental struggle because - no current pilot expected or expects performance from this aircraft system sector.  
• That wouldn't be fair - If a Capacitance system were offered - Because of a large aircraft familiarity,  that concept might gain traction 
  • Pennycap was introduced in 1969 by Cessna - It was not a well received solution - Cessna only.  It wasn't an answer.
  • We are more reliable and more accurate over a capacitive system  -  float guy has plane credibility.

The concept of pushing intelligence to the sender level,  gives a lot of flexibility for warnings and capabilities not considered today.

It provides capabilities that make flying easier and yes with an increase in warning capability and monitoring  - safer 

-  Charles Beck - International Avionics  - the manufacturer of the Mooney Master Caution Panel really advanced fuel safety with an advanced low fuel level warning for Mooney

All of what you are asking has already occurred, in order to to make automated tank switching on a single engine a viable concept. 

We can monitor trends and direct appropriate warnings.  

We reviewed the egg timer solution of the TBM 900 or the much more sophisticated system on the Pilatus PC-12 

It culminated in our own OEM single tank solution.   

Like I said before- We aren't here to keep bad pilots out of trouble   - we make good pilots better. 

 

[carusoam]

Scott P. / Phil S.,

Sorry buddy. I'm still working on my memory skills. 

You can only imagine how challenging it is for some people to follow your post.

Best regards,

-a-

[fuellevel]

I started over in the commercial aspect.  

but yes in addressing each concern - it is a bit of a mishmash 

I deleted a lot of responses to clean it up if I was quoted. 

 

[Yetti]

In flight Analytics, / IoT / Edge computing.  Using a Raspberry PI ($70) and a 9 degrees of freedom board and some flight profiles it would be easy to have a bitching Betty to .warn of impending stall spin...

For one of my work projects we are doing a Fly no Fly based on inputted personal minimums and NWS reports and TAF.

 

 

Edited July 4, 2016 by Yetti

[Bob - S50]

15 hours ago, Hyett6420 said:

Ok I'm going to wade in.   As far as I'm concerned this boils down to cost.  Can I with the most powerful computer on the planet work out:- 

1. the fuel i have put in the tank - yes
2. my endurance based on historical performance and fuel burn - yes
3. monitor over time the existing guages i have and what they show after each hour of flight - yes
4. Montor guages using 4 work out if i have a fuel flow issue (ie more than i expect) - yes. 

Nothing and I mean nothing that you put in my cockpit is more powerful than the most powerful computer on the planet....yes your brain!  So why should I pay for something that I firmly believe adds no value to me as a pilot.   

I honestly feel that this whole post has been a strong sales pitch.  Apologies if it is not but that is exactly what it feels like.  

Now if someone invents a system that looks at the angle of attack, speed and bank angle and automatically reduces the angle of attack, increases power etc so we eliminate stall spin accidents on turns to Final now THAT IS something that should be installed on every aircraft and people would pay for.  

Andrew 

Andrew,

I don't think it's just a sales pitch.

1.  Agree, you can and should know how much fuel you have on board before you start.

2.  Agree IF you don't develop a leak and IF you don't get distracted and forget to lean properly.

3.  Maybe.  What if those existing gauges are lying to you?  That seems to have been the case for the pilot of the Piper that was mentioned in this thread.  We have a digital fuel gauge hooked up to standard floats.  Over about a 3 year period, the output from our left tank has changed enough that we know the indications are as much as 5 gallons higher than actual.  Standard resistance floats degrade with time and can hang up due to friction.  So I'd say it depends on whether you are looking at the stock needles or a digital gauge.  With the stock needles, you might not notice that they have not moved the correct amount.  We are getting ready to re-calibrate our gauge either with Cies floats or stock floats if we can't get the Cies floats.

4.  Again, you can only do that if your gauges don't lie to you.

I don't know if you have a fuel flow/used or not.  We do.  Even if you do and you develop a fuel leak upstream of transducer, it won't show up as fuel used.  Could be a loose clamp, leaking gascolator, venting fuel, broken line, etc.  If your gauges are lying to you, you'll never see it.

In other words, garbage in, garbage out.  Put incorrect info into that most powerful computer and you'll get incorrect results.

Bob

[bonal]

Bob, very good points but the thing is I know my gauges are lying and any device can malfunction if you have a fuel loss and your true fully accurate gauge also fails your still in the same boat. Unfortunately there is no way to guarantee a perfect solution. I would love to have an accurate system but in my case fuel gauge is like number 100 on the list and I'm on number ?

[fuellevel]

Well - the topic started out with the standard why do pilots run out of fuel.   Typically this falls into the following dialogue only

• Lack of awareness 
• Lack of training 
• Lack of care 

This is well supported by AOPA, and the aviation editors of the various flying oriented magazines 

Reinforcement of checking starting fuel value (use of a stick)  and planning are the two tools of preventing this incident from occurring 

It is widely reported that fuel totalizers help but they do not address all potential failure paths,

confusingly the advent of fuel totalizers, while well liked and trusted have not moved the accident statistic in any great manner

Several, Cirrus aircraft prior to 2012 have run out of fuel with a fuel totalizer and a traditional gauge, for example.   

Totalizers have not prevented fuel related accidents,  even in technologically advanced aircraft. 

Given the unchanging statistic - you could state

Neither education or totalizers have presented an answer to fuel related accidents.  

Yet we continue to reinforce these methods 

-------------------------------

It is widely reported that fuel indication lacks clarity and accuracy in aviation 

Aircraft with an operational low fuel warning and a high wing (Cessna) suffer almost zero (1 or 2)  fuel starvation / exhaustion events 

Aircraft with a digital fuel quantity systems have to date - suffered zero fuel starvation / exhaustion events.   

It may be, but not proven at this time,  that quality fuel indication may be the method that allows good pilots better tools and

will provide a timely warning assist those careless pilots an opportunity to correct the preflight mistakes they may have made.

Most new aircraft delivered have digital fuel quantity systems, and soon nearly all new aircraft will have a digital fuel quantity systems. 

If the current trend of no fuel starvation with digital fuel indication continues - fuel related accidents will go down. 

---------

If this is in any way compelling,

• Make the fuel quantity system in your aircraft functional 
  • If you can't make the fuel quantity system in your aircraft work
  • Avail yourself of any number of other options,  to make it work with the degree of accuracy intended by the manufacturer.  

 

• To bonal's point, you can only determine if a system is not functioning when it deviates from normal operation.   Trying to determine if a randomly operational system is operating in another random manner that signifies failure - is useless.   That being said most digital systems are set up as life of aircraft 50,000 hr systems.    Resistive senders are 2,000 to 3,000 hr systems in aircraft. 

[bonal]

I certainly don't feel like re reading this so not even sure your product is available for our Mooney's and one with bladders to boot. If your looking for a volunteer for beta testing mine would be available of course me being the CB I am and having to spend all of my available budget on things need fixing it would have to be on your dime. 

Im out.

[carusoam]

Summary of a PP's understanding of what he has read around here....

1) The combination of decent FL and FF gauges affords the pilot the ability to run out of fuel with accuracy.  

2) Adding digital bits and pieces allows the same pilot to run out of fuel with precision.  

3) Using a high wing fuel system allows the pilot to somehow run out of fuel without anyone noticing.

 

The challenges seem to be...

1) Plan A: It takes being aware of the situation prior to departure.  Enough fuel + proper reserve.

Required: calibrated and operational FL indicators, or a calibrated stick proving to the pilot he has the fuel he thinks he is starting with.

Supported by: Fuel delivered quantity by the pump.

 

2) Plan B: It takes being aware of the situation on the fly.  Fuel used or fuel remaining by the totalizer or FL gauges.

Required: FF and totalizer with accuracy in the one gallon range.

Supported by: FL in the one gallon range.  

 

3) Plan C: It takes being notified of the situation before running out of physical options.

Required: 

- operational annunciator panel that has had it's test button pressed to arm the low fuel level circuit.

- know how much fuel is left when the light goes on.  Some LBs had been set for 3 gallons usable per side. 15 minutes to enact the landing procedure is not very much unless you are over an airport.

- in VFR conditions the lights on the anunciator panel may not be very effective without actively including them in your scan.

- FL or fuel remaining (FF)connected to the GPS with an active flight plan is a great way to be given a heads up if the fuel is not enough to get to the final destination.

 

4) Plan D: An accurate fuel use plan compared to actual fuel used is the heads-up that is needed to keep from accidentally running out of fuel.  It all starts with knowing how much fuel you are starting with.

- know the fuel used in start-up, taxi, T/O, Climb, cruise, descent, landing.  Climb and cruise are the most significant.  Descent can be included in cruise with minimal accuracy losses on the safe side.

- switch tanks according to a plan.  Do you switch at the TOC, followed by every hour after that?  Some people change each 10 gallons or so...  Since climbing in a NA plane uses a non-consistent FF when leaning using the Taget EGT / blue box method.  I burn the lesser known amount of fuel from one tank.  I usually start from the left tank for consistency.

- some people like to intentionally run a tank dry.  This leaves the last usable fuel on one side. Without a real good FL gauge, the feeling of unease becomes stronger after one tank is empty.  The ability to adjust rate of descent into a short field is impossible after the second tank runs dry.

 

5) Plan E: Knowing what the legal reserve looks like on the FL indicators. 

- know what 5, 10, 15 gallons looks like on the FL indicators.  Does it even show up?  Does it become unreadable in flight?

 

6) Plan F: Verify there are no fuel leaks or fuel use surprises.  Fuel use surprise may be caused by simply not remembering to set the mixture properly.

- leak after the FF meter.  Check to see if FF matches the various flight segments T/O and climb FF, and cruise FF (ROP or LOP)

- leak before the FF meter. Check to see if FL matches expectations each time the valve is switched.

-Checking fuel used from two independent sources would be cool!  CIES vs FT101...  Hmmmmmmmm 

 

7) Plan G: Commercial wide scale adoption Of CIES digital floats would be aided by word of mouth of three groups...  Bonal has offered his plane as a test bed and has shown himself as an excellent member of MS.

- Short body Mooneys A, B, C, D, E...  (Bonal fits in this group)

- Mid length body Mooneys F, G, J, K... (A group of 10 is requested for float arm development / study related to tank design)

- Long Body Mooneys L, M, R, S...

 

let me know What I missed.  There has to be plenty more...

Best regards,

-a-