Inop fuel gauge - what would you do?

[Everskyward]

FAA wrote it this way on purpose - but I am not sure of that purpose.

My guess is that if they did not write it this way, they would need to come up with a tolerance and a way to measure it for every certified airplane. It doesn't make sense that the only time the gauge needs to be accurate is when there is no usable fuel. Otherwise, you could have a dummy gauge that always reads zero even though it's not hooked up to anything. On the other hand, there's no way a gauge can be completely accurate. Accuracy needs to be defined.

I'm glad that the FAA is vague on many issues. But I realize many on POA are not.

 

[Tom-D]

I haven't worked with a metrology department for a long, long time. But back in the day, if there were no tolerances given, we assumed a couple things: 1) whoever wrote the specs forgot something, 2) since no error range was given, we had to assume that none was allowed (and we couldn't do that kind of measurement because all measurements will have some error). We would not assume that since no tolerance was given that ALL error was allowed.

FAA wrote it this way on purpose - but I am not sure of that purpose.

Simply because they know there are many ways to measure fuel quantity. in order to specify accuracy they would be required to specify method.

 

[Matthew]

Otherwise, you could have a dummy gauge that always reads zero even though it's not hooked up to anything.

Not necessarily - you could end up with a gauge that reads EMPTY when zero and anything else when not zero. Basically a light that is on when you reach zero usable and off when anything else.

All these comments have gotten away from what I was taught, "Never trust a fuel gauge."

 

[Everskyward]

Not necessarily - you could end up with a gauge that reads EMPTY when zero and anything else when not zero. Basically a light that is on when you reach zero usable and off when anything else.

But having a dummy fuel gauge would be a simpler and cheaper option!

 

[azure]

Sigh. This argument can end in only one way. Someone will write the Chief Counsel's office for an interpretation, and then we will have a legally binding accuracy standard for fuel gauges written by someone who has absolutely no understanding of instrumentation. The end result will likely be significant additional costs imposed on aircraft owners with no, or unmeasurably small safety benefit to owners, pilots, or the public. Because inaccurate fuel gauges are not the reason pilots run out of fuel, they run out of fuel mostly because of wishful thinking, refusal to do simple math, and passing up perfectly good fuel stops to save a few dollars. All that would still happen even if fuel gauges had +-1% accuracy because the problem is ADM, not lack of information.

 

[bradg33]

Sigh. This argument can end in only one way. Someone will write the Chief Counsel's office for an interpretation, and then we will have a legally binding accuracy standard for fuel gauges written by someone who has absolutely no understanding of instrumentation. The end result will likely be significant additional costs imposed on aircraft owners with no, or unmeasurably small safety benefit to owners, pilots, or the public. Because inaccurate fuel gauges are not the reason pilots run out of fuel, they run out of fuel mostly because of wishful thinking, refusal to do simple math, and passing up perfectly good fuel stops to save a few dollars. All that would still happen even if fuel gauges had +-1% accuracy because the problem is ADM, not lack of information.

I totally agree that fuel gauges generally have little to do with fuel starvation/exhaustion incidents. And I'm not saying I won't fly an airplane unless the fuel gauges are accurate. But, let's not kid ourselves into thinking that the myth regarding "only accurate at zero" isn't a myth. And most of the time the only reason gauges are wildly inaccurate is because owners are too cheap to fix them.

 

[Matthew]

But having a dummy fuel gauge would be a simpler and cheaper option!

It would be portable, too. Just peel off the double-stick tape and you are never grounded again!

 

[flyingron]

No "counsel" opinion is needed. You need to have "functioning" fuel gauges which is pretty much "working in accordance with the type certificate." It's just that the FAA doesn't specify accuracy requirements for them either as part of certification or for continued airworthiness. Such is the case with 99% of the stuff on your aircraft. I don't know why people are getting wrapped around the axle on this. There's no accuracy requirement on most instruments (altimeters and compasses are the exceptions). However, if my tach was only reading half the RPMs it should, the FAA would have no problem claiming that it's not working in accordance with the aircraft certification.

Similarly a fuel gauge that read only empty would not be construed as being airworthy, but there's still no "accuracy requirement."

 

[Checkout_my_Six]

Accuracy? Geez people the markings wouldn't allow for anything more accurate than +/- an eighth of a tank.

 

[Palmpilot]

...The regulations regarding fuel gauges do NOT have a published range of acceptable error, meaning that under a literal reading, they must be perfectly accurate as the FAA has not defined an acceptable range of error. The failure to define an acceptable range of error does not, using common sense (which I know many prefer to abandon here), mean that the acceptable range of error is 99%, so long as it reads zero when there's no usable fuel left.

I sure HOPE the people running our legal system are capable of common sense, since "perfectly accurate" would require violating the laws of physics! Personally, I would settle for "accurate enough to be reasonably useful."

 

[MIFlyer]

Fuel gauges can be deferred under an MEL but it's doubtful the 172 in question would have one. I looked on this site for one but the smallest Cessna I could come up with was a 310/320.

UND's 172's are on an MEL, according to a former student there.

on G1000 172s, I've flown one with intermittent red X on one of the tanks (can't remember which). It was in and out the day before the checkride, but held steady the day of. (Whew!) as 91.205 and the POH both require it for that airplane. Sounds like it's a not unusual problem in G1000 172s, hopefully they can dial this in, as it sucks to ground a plane over an inoperative guage that doesn't give you much info that your fuel planning and flow rates tell you anway (G1000 will show your GPH the whole time you're flying too, so the math is pretty easy and lets you back up your manual flight plan)

 

[MauleSkinner]

Ron is saying the FAA doesn't put any specific requirements for accuracy on fuel gauges. You can say all you want, but it doesn't change that.
If there was an accuracy requirement, it would be stated. Stating that you have to set the gauge so 0 is end of usable fuel is not an accuracy statement. Saying it is "calibrated" by itself is not an accuracy requirement. It's like saying the wheels on your aircraft must be round. Fine, but to what degree? Is a little flat OK.

You'll not find anything. Not in the FARs, not in an AC, not in an FAA order that says anything about fuel gauge accuracy. On the other hand, look at compasses. Part 23 and 25 tell you how accurate they need to be. Altimeters (at least for IFR) have very definite accuracy requirements.

The FAA puts a very specific accuracy requirement on fuel gauges...you have to be able to determine how much usable fuel you have from the gauge in flight.

 

[flyingron]

The FAA puts a very specific accuracy requirement on fuel gauges...you have to be able to determine how much usable fuel you have from the gauge in flight.

That doesn't imply an accuracy. Do you understand the difference between an indication and accuracy? 24 gallons is not an accuracy. 24 gallons +/- .5 gallons would be an accuracy statement. THere's such requirement. In fact, most aircraft fuel gauges don't have that great accuracy even as designed from the factory.

 

[TommyG]

I very rarely pay attention to fuel gauges. If I knew the take is topped when I go. For a 172 I estimate 11 GPH, which is an over estimate. I have never relied on fuel gauges, whether the G1000 digital or 152 inaccurate gauges.

 

[Stewartb]

I know of an accident where a 185 pilot crashed due to fuel exhaustion. He had been monitoring his Shadin fuel monitor when he ran out. Apparently he had a fuel leak upstream of the monitor impeller. Bummer. Personally I dip tanks before every first flight of the day. I cross reference that to my digital fuel gauges, which I don't care much for. Finally I check my FS-450. For years I just used a dipstick and a watch. Ultimately that's still my favorite tool but I have fancy electronics to validate what my Fred Flintstone method indicates.

I have an intermittent RH fuel sender. It ain't all the way broke so I haven't made any effort to fix it. No big deal.

 

[MauleSkinner]

That doesn't imply an accuracy. Do you understand the difference between an indication and accuracy? 24 gallons is not an accuracy. 24 gallons +/- .5 gallons would be an accuracy statement. THere's such requirement. In fact, most aircraft fuel gauges don't have that great accuracy even as designed from the factory.

ok, so if accuracy is not the correct terminology to define the usability of the gauge with respect to the certification criteria, what is?

Based on context, I'd say that's what everybody except you is discussing.

 

[TangoWhiskey]

Fast forward to today, five years later. How's the fuel gauge doing?

The two in the airplane I now own work GREAT! ;-) And I have a Fuel-Stik calibrated to MY plane from empty tanks, and a highly accurate fuel flow gauge tied into my JPI-830 and Garmin GNS430W.

Life is good. ;-)

 

[Tom-D]

The two in the airplane I now own work GREAT! ;-) And I have a Fuel-Stik calibrated to MY plane from empty tanks, and a highly accurate fuel flow gauge tied into my JPI-830 and Garmin GNS430W.

Life is good. ;-)

I don't really care how much you have spent or what kinda gauges you have. There is only two times when you know exactly how much fuel you have.

 

[TangoWhiskey]

I don't really care how much you have spent or what kinda gauges you have. There is only two times when you know exactly how much fuel you have.

I was answering Mari's question, not you. Unfortunately it didn't quote.

 

[Tom-D]

I was answering Mari's question, not you.

No matter who you were talking to, my statement is still true.

 

[TangoWhiskey]

No matter who you were talking to, my statement is still true.

So?

 

[Tom-D]

So?

So,, you can't figure it out.

 

[unsafervguy]

Show me why FAR 91 does not apply to EXP aircraft.

Since nobody answered your little test here it is:
91.205
(a) General. Except as provided in paragraphs (c)(3) and (e) of this section, no person may operate a powered civil aircraft with a standard category U.S. airworthiness certificate in any operation described in paragraphs (b) through (f) of this section unless that aircraft contains the instruments and equipment specified in those paragraphs (or FAA-approved equivalents) for that type of operation, and those instruments and items of equipment are in operable condition

It say standard catagory airworthiness certificate. Experimental is not a standard catagory Airwortiness cert.

Bob

 

[Scott Philiben]

Sigh. This argument can end in only one way. Someone will write the Chief Counsel's office for an interpretation, and then we will have a legally binding accuracy standard for fuel gauges written by someone who has absolutely no understanding of instrumentation.

So much of this is based on pilot here say - it needs to be addressed. There is a quality standard and it is referenced in the Advisory Circular 23.17. Equally if you review the maintenance manual for any aircraft - that would be the legally binding definition of airworthy. Piper has a pretty restrictive one on the Cherokee - see attached file

The end result will likely be significant additional costs imposed on aircraft owners with no, or unmeasurably small safety benefit to owners,pilots, or the public.

Interesting that this opinion isn't supported by fact. It is a commonly reported opinion and one you find in aviation print . - In Australia they found that 100% of all aircraft suffering a fuel related accident had non- functional fuel gauges. Fuel Gauges that did not show the remaining usable fuel in the tank. Australia requires pilots to re-calibrate the fuel indication every 4 years. This program is reducing fuel related accidents in Australia. Transport Canada is finding similar results more that 88% of fuel incident aircraft were flying with non functional fuel instrumentation, Transport Canada has petitioned the FAA for a action response. In the US in 80% of the cases the fuel gauges are not evaluated. You can't find what you are not looking for.

Because inaccurate fuel gauges are not the reason pilots run out of fuel, they run out of fuel mostly because of wishful thinking, refusal to do simple math, and passing up perfectly good fuel stops to save a few dollars. All that would still happen even if fuel gauges had +-1% accuracy because the problem is ADM, not lack of information.

"If you can't measure it - You can't manage it" This is another common pilot opinion again not supported by fact. So let's set up an argument. Fuel totalizers have become relatively universal in GA aircraft - yet the introduction of this safety equipment has had little if no effect on the fuel starvation issue. Aircraft with totalizers consistently run out of fuel. Modern aircraft with Garmin displays and fuel totalizers - run out of fuel. But here is a little known fact - 95% of new aircraft now come with a digital fuel level system - this has been in place since 2011. There are 2,500 aircraft flying with accurate digital fuel level and not one has succombed to a fuel related incident or accident. Next year the number of digitally equipped aircraft will rise to 4,000. This is an ongoing experiment - and it is proving Azure's conjecture to be without merit and that maybe we should re-think what the lack of accurate fuel indication in aircraft has cost us - in human lives, wrecked aircraft and the indelible mark in the general public of an aircraft on the news landing on a highway.

Common opinion - not based on factual information - may just be wrong .

 

[Scott Philiben]

I sure HOPE the people running our legal system are capable of common sense, since "perfectly accurate" would require violating the laws of physics! Personally, I would settle for "accurate enough to be reasonably useful."

Every large aircraft is equipped with a fuel level system capable of being accurate throughout the flight envelope of 3% of total fuel volume - many of these systems meet the 0.75% requirement for fuel volume. You can measure fuel in aircraft - it is done and as stated above. Most new GA aircraft are now delivered with a digital fuel quantity system. So while no law of physics was broken in this endeavor. One of these systems exceeds the 0.75% accuracy range or +/- 0.2 gallons on a 25 gallon tank.

What defies the laws of physics is that a pilot will believe that a fuel sender designed for automotive use and made of steel will last indefinitely without maintenance or calibration in a 30 to 50 yr old aircraft. An aircraft that is primarily made of aluminum and exposes this steel sender unit to induced moisture on descent is basically an advanced corrosion test. I boggles the imagination that pilots believe that this instrumentation will still operate effectively and provide operable information. Actually this slow corrosive decay robs pilots of what new instrumentation was supposed to do - and has lead to the very low expectation of this system in aircraft.

Fuel instrumentation prevents fuel related accidents - just like tachometers prevent overspeeding the engine. It just makes sense.

 

[Palmpilot]

Every large aircraft is equipped with a fuel level system capable of being accurate throughout the flight envelope of 3% of total fuel volume - many of these systems meet the 0.75% requirement for fuel volume. You can measure fuel in aircraft - it is done and as stated above. Most new GA aircraft are now delivered with a digital fuel quantity system. So while no law of physics was broken in this endeavor. One of these systems exceeds the 0.75% accuracy range or +/- 0.2 gallons on a 25 gallon tank.

I was just pointing out the meaninglessness of terms like "perfectly accurate." Accuracy requirements must be specified by tolerances if they are to be meaningful. The Piper example you posted seems quite reasonable to me. Tolerances that are common for large aircraft, on the other hand, may not be economically feasible or necessary for light aircraft.

Digital technology removes some sources of error, but not all. I'm not sure that it addresses the error sources listed in your paragraph below, for example.

What defies the laws of physics is that a pilot will believe that a fuel sender designed for automotive use and made of steel will last indefinitely without maintenance or calibration in a 30 to 50 yr old aircraft. An aircraft that is primarily made of aluminum and exposes this steel sender unit to induced moisture on descent is basically an advanced corrosion test. I boggles the imagination that pilots believe that this instrumentation will still operate effectively and provide operable information. Actually this slow corrosive decay robs pilots of what new instrumentation was supposed to do - and has lead to the very low expectation of this system in aircraft.

Fuel instrumentation prevents fuel related accidents - just like tachometers prevent overspeeding the engine. It just makes sense.

 

[Scott Philiben]

Digital technology removes some sources of error, but not all. I'm not sure that it addresses the error sources listed in your paragraph below, for example.

Actually you have to be precise here - A lot of Aircraft have Garmin instrumentation and that is digital - however most aircraft sensor outputs are analog - CHT, EGT, Etc. This was also true of fuel quantity as these sensors up until 2011 were primarily analog potentiometers.

With the change to digital recording and transmission of fuel level, there are two methods utilized, both manufacturers looked at the issues related specifically to aircraft fuel level. One system was designed as business jet system but failed in that marketplace and is now applied to GA aircraft and the other is a system designed for GA aircraft that is finding favor in business and helicopter platforms.

Both systems use aluminum as a primary material - (Recent Aircraft analog systems were aluminum as well) - one system is teflon hard-coat anodized to limit aluminum pit corrosion.

Both systems were designed to tolerate fuel contamination and by products.

Both systems were qualified to TSO C55a - so they met a plurality of aircraft related environmental issues - Vibration, Corrosion, Voltage Spikes - and meeting an accuracy requirement during those tests.

Both systems have some sort of microprocessor for fuel quantity processing and outputting some form of digital information - simple or complex.

The limits of these new systems is that GA aircraft use fewer systems in the tank - typically 1 or 2 where a business jet like the Eclipse may use 5. The quantity of senders used relates to addressing aircraft attitude effects on fuel and lightning strike redundancy.

The driving factor for incorporation in GA aircraft was not accuracy or a dis-satisfaction with fuel quantity indication

It was the fact that both of these systems were intrinsically safe - NO ELECTRONICS IN THE TANK - Resistive potentiometers can be made safe in the fuel environment it is done everyday in automotive however lightning issues specific to wing mounted fuel tanks allows an arc to form between the in tank wiring and structure.

Both of these systems though addressed the accuracy issue head on and meet a particular requirement.

Something to chew on - Aircraft fuel gauges are required - however the size of the gauge is not specified - Piper and Cessna used small gauges where accuracy was difficult to ascertain and this gauge size was maintained until the Garmin era - Beech moved to a larger gauge in the early to mid 60's as fuel quantity increased in the Bonanza.

 

[Palmpilot]

I'm glad to hear that they are addressing the residual analog links in the chain.

 

[Scott Philiben]

There is a lot of work in this regard @ the OEM level - digital sensors can provide lots of local information and assist in preventing catastrophic failure.

It is similar to the automotive world - I won't let my Dad buy a car without an automatic pre braking or a backup camera

Technology can improve safety - it just hasn't hit aviation as hard as it has the automotive world - yet.

I am witnessing some pretty fantastic things

 

[Palmpilot]

There is a lot of work in this regard @ the OEM level - digital sensors can provide lots of local information and assist in preventing catastrophic failure.

How can there be such a thing as a digital sensor? Doesn't there have to be some analog component in any transducer?

 

[MAKG1]

There is a lot of work in this regard @ the OEM level - digital sensors can provide lots of local information and assist in preventing catastrophic failure.

It is similar to the automotive world - I won't let my Dad buy a car without an automatic pre braking or a backup camera

Technology can improve safety - it just hasn't hit aviation as hard as it has the automotive world - yet.

I am witnessing some pretty fantastic things

Technology can also promise things it doesn't deliver. Be skeptical.

Backup cameras, for one, have an obvious hole. They look directly behind, but not for common moving hazards from the side. Those remain to be evaluated objectively. ABS systems have been thoroughly discredited -- they resulted in NO decrease in loss of control accidents, though they were sold as though they would. There are some situations where they make an avoidable loss of control incident unavoidable.

ALL digital systems suffer the risk of GIGO. They also suffer discreteness effects and sometimes dynamics that analog systems don't. And the level of fuel in the aircraft is fundamentally analog unless you count atoms. You will never make it truly digital.

 

[Scott Philiben]

In the basic element in some cases. Take the fuel quantity - One method is Time Domain Reflectometry - so a laser light is passed down a glass rod and reflections are measured
there is probably an analog component measuring reflected laser light - but it is quickly converted from Analog to Digital. The other measurement was Anisotropic Magneto Resistive and the basic characteristic is a change in resistance through a barber pole stack of silicone and silicone iron elements - however this analog needs to be boosted and converted - most times in the actual sensor itself as most of these sensors are attached to an ASIC of some sort.

Pressure transducers are typically piezo electric elements and analog - but the signal can be processed and a digital output provided

So like radios that are now digital - yes the antenna and receiver is likely analog but the signal processing prior to delivery is digital - culminating with an analog speaker delivery

All components are some mixture - even the keyboard I am typing this message on

Signal processing and Signal delivery are the key differentiators -

 

[Palmpilot]

In the basic element in some cases.

Since we're talking about measuring an analog world (at least above the quantum level), there will always be some point where the information must be converted from analog to digital.

Take the fuel quantity - One method is Time Domain Reflectometry - so a laser light is passed down a glass rod and reflections are measured
there is probably an analog component measuring reflected laser light...

I don't think there's any "probably" about it.

... - but it is quickly converted from Analog to Digital. The other measurement was Anisotropic Magneto Resistive and the basic characteristic is a change in resistance through a barber pole stack of silicone and silicone iron elements - however this analog needs to be boosted and converted - most times in the actual sensor itself as most of these sensors are attached to an ASIC of some sort....

That would seem to work counter to the goal you mentioned earlier, of having no electronics in the fuel tank.

 

[Scott Philiben]

Actually on the no electronics - the Anisotropic Magneto resistive sender can take the measurement outside the tank through the tank wall

And yes it probably does - the Time Domain Reflectometry - but I am not privy to the actual components