how much confidence do you have in fuel gauges?

readytocopy995

Pre-takeoff checklist
Joined
Oct 29, 2019
Messages
168
Location
Camarillo, CA
Display Name

Display name:
buy_ridgewallet
Have a 172 with float style. How much "trust" would you put in these? Or basically how low would you actually feel comfortable running them in the real world?
 
A lot depends on what other equipment I have. A fuel flow meter is GIGO but very accurate. Nothing to really measure use accurately? I plan legs based on an hour reserve at the worst burn rate in the tables. But I always believe the gauges when they run low. And I want my crosscheck to only have reasonable discrepancies.
 
I’ve found the ones in the Archer to be quite accurate, but I definitely don’t rely on them.
 
Nada. Zip. Zilch.

My fuel gauge is on my wrist. I know what I took off with, I know my anticipated fuel burns, with a reasonable fudge factor, and I know when the tank should be empty.
 
I’ve flown quite a few airplanes and have only run across a handful that didn’t have fuel gauges that were trustworthy. The ones I didn’t consider trustworthy likely could have been fixed if the owner would have put some time and money into them.
 
This may come as a shock to some of you, but I don't keep track of how many hours I have been driving my car since the last fill up. I just stop to gas up my car when the gauge gets near empty. My airplane uses exactly the same technology to indicate the fuel level in the airplane.

Assuming you have flown and fueled a particular aircraft on more than one occasion, one should have a pretty good idea about how well the indication correlates to actual fuel levels. Now, that is not to say that fuel gauges can't fail - if you have been flying for 4 hours and the gauges still indicate full, probably best to land and investigate. On the other hand, if the gauge says empty after just 30 minutes, one can't rule out a fuel leak and one should land and investigate that as well.

 
I use elapsed time, supplemented with the gauges. I don’t intend to be “that guy.”
 
No way Jose. Always keep track using other (timed) methods while only using the gauges as a reference.
 
I don't trust the gauges when the show full. But I do trust the gauges when they show low.
I have the opposite problem. When mine read low, the tanks aren't. But I'd rather have that problem than ones that read like they have too much gas in them. On the other hand, it's kind of annoying to feel like I have to land to refuel when in fact I still have a couple hours' flying time left (including the reserve, though, so only 1.5 hours' flying time).
 
Iirc, fuel gauges are only required to be accurate at empty.

Fuel flow on the jpi has been proven to be very accurate. Gauges will maybe show if you are losing fuel.

I was out in the backcountry once where both my gauges were in the do not take off range. My jpi said I had 21 gallons, my home made fuel stick agreed, I took off and flew 45 minutes and still had 10 gallons.
 
Iirc, fuel gauges are only required to be accurate at empty.
You recall incorrectly. The requirements state it something like this: The airplane shall have a means for the flight crew, when seated at the flight controls to determine
  • the fuel quantity in each main fuel tank...
  • The fuel quantity indicators must read Empty when the tank contains only unusable fuel.

Now, a fuel gauge that's only accurate at Empty isn't any sort of a reliable means of determining the fuel quantity. The requirement to read Empty at the unusable level is a specification, but they don't give a specification for the rest of the range. Yet the requirement implies at least some sort of accuracy.

These gauges weren't too bad when they were new. But many are worn out, and don't give indications that are anywhere near useful. It's usually the senders in the tank that are shot, or their float arms have been bent and the sender is not properly indicating the floats level.

In addition, gauging tanks that are shallow and long and wide is difficult; any movement of the airplane sends the fuel sloshing all over the place. Tanks in wings also have dihedral to deal with. The sender has to be able to sense the unusable fuel level, meaning that it has to be at the inboard end of the tank, but it's the center of the tank where the sloshing affects gauging the least.

Edit: From the AC that replaced chunks of FAR 23:

upload_2022-7-30_10-1-54.png

https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_23-17C.pdf

Page 189.
 
Last edited:
You recall incorrectly. The requirements state it something like this: The airplane shall have a means for the flight crew, when seated at the flight controls to determine
  • the fuel quantity in each main fuel tank...
  • The fuel quantity indicators must read Empty when the tank contains only unusable fuel.

Now, a fuel gauge that's only accurate at Empty isn't any sort of a reliable means of determining the fuel quantity. The requirement to read Empty at the unusable level is a specification, but they don't give a specification for the rest of the range. Yet the requirement implies at least some sort of accuracy.

These gauges weren't too bad when they were new. But many are worn out, and don't give indications that are anywhere near useful. It's usually the senders in the tank that are shot, or their float arms have been bent and the sender is not properly indicating the floats level.

In addition, gauging tanks that are shallow and long and wide is difficult; any movement of the airplane sends the fuel sloshing all over the place. Tanks in wings also have dihedral to deal with. The sender has to be able to sense the unusable fuel level, meaning that it has to be at the inboard end of the tank, but it's the center of the tank where the sloshing affects gauging the least.


All true, but a “requirement” that “implies some sort of accuracy” isn’t really a requirement at all and it isn’t verifiable.

Also, “unusable fuel” should specify attitude. In cruise, a Beech Musketeer will drain a tank down to a quart or so, but in landing attitude with the nose pitched up a little it can’t get the last couple of gallons.
 
I flew all over Alaska in 206s and 207s with those type of gauges.

When the needles stop bouncing, it is time to worry.

But yes, a unapproved home made fuel stick and watch was used to determine fuel use.

Then again I have flown planes where the fuel gauge was the only way to determine how much fuel is in the tanks. Not a 172 though...
 
Also, “unusable fuel” should specify attitude. In cruise, a Beech Musketeer will drain a tank down to a quart or so, but in landing attitude with the nose pitched up a little it can’t get the last couple of gallons.
Somewhere else it specifies accuracy in level flight attitude. Can't put my finger on it, but I did read it not ten minutes ago.

Unusable fuel is defined as the fuel that will not flow out of the tank's outlet to the engine when the airplane is in the most critical attitude for flight. That would be a steep, flap-extended glide, or a Vx climb, maybe on an overshoot to clear an obstacle. So if the gauge reads Empty in level flight, better watch out.

We had a 150 run below usable fuel on final once. Engine quit. The instructor took control, dumped some flap and got the nose up and the power returned enough to safely complete the landing. He knew more than the average CFI. Whoever made the dipstick for that airplane had calibrated it from zero fuel, empty tanks, instead of having the unusable fuel as listed in the TCDS in the tanks.
 
I don’t even trust my fuel stick below about 12 gallons or so. Once the fuel is down below about 10 gallons, there’s almost no fuel directly below the filler cap so the stick can’t get a reading.
 
I have an Aerospace Logic digital fuel gauge with Cies senders. Pretty dang accurate for sure. Do I trust them implicitly? No, but it’s a far cry more piece of mind than stock.
 
Last edited:
All true, but a “requirement” that “implies some sort of accuracy” isn’t really a requirement at all and it isn’t verifiable.

Also, “unusable fuel” should specify attitude. In cruise, a Beech Musketeer will drain a tank down to a quart or so, but in landing attitude with the nose pitched up a little it can’t get the last couple of gallons.
It’s a regulatory requirement that doesn’t imply anything, so it is actually a requirement, and it does specify an attitude.
 
Have a 172 with float style. How much "trust" would you put in these? Or basically how low would you actually feel comfortable running them in the real world?
I was taught by an instructor in his 70s who had been teaching for 50 years. He gave me the best piece of advice I’ve gotten about fuel gauges: if they read more than half, you’re good. If they read less than half, it’s time to land for fuel.
 
All true, but a “requirement” that “implies some sort of accuracy” isn’t really a requirement at all and it isn’t verifiable.

Also, “unusable fuel” should specify attitude. In cruise, a Beech Musketeer will drain a tank down to a quart or so, but in landing attitude with the nose pitched up a little it can’t get the last couple of gallons.
§ 25.1337 (b)1. They have to read empty in level flight when the only fuel remaining is unusable.
 
Have a 172 with float style. How much "trust" would you put in these? Or basically how low would you actually feel comfortable running them in the real world?

Real world?

Ummmm, all my flying's "real world". You?

I have the panel mounted gauges, which are lousy, but I know what it "really means" at each notch.

I have wing-tank mounted gauges, that in the correct lighting I can see what's in the tanks rather well. I trust those 100%.
 
My fuel flow (engine monitor) is accurate to within 1 gallon (probably more accurate if I was more consistent when filling the tanks).
My wing mounted are only accurate on the ground (as designed).
My fuel level gauges are accurate to within -/+3 gallons per side, so my total can be off as much as 6 gallons.

I use the fuel level to insure I’m not leaking fuel during the flight.
I use wing mounted gauges to insure I haven’t lost fuel while parked.
FF I used to decide my actual consumption.

Since I’ve never landed with less than 15 gallons, it’s all probably doesn’t matter about lack of accuracy.
 
Accurate enough to know how much fuel is in the tank.


That would be +/- 0%? An exact amount? Then all fail.

If there’s an accuracy requirement it needs to be expressed. If an error band is not stated, there is no accuracy requirement.
 
Low tech sight tubes do well for the high-wing crowd. Seeing is believing.
 
Have a 172 with float style. How much "trust" would you put in these? Or basically how low would you actually feel comfortable running them in the real world?

An unknown gauge, not at all. One I fly with regularly and am familiar with, maybe.
 
That would be +/- 0%? An exact amount? Then all fail.

If there’s an accuracy requirement it needs to be expressed. If an error band is not stated, there is no accuracy requirement.
The stated accuracy requirement is that you need to be able to determine how much fuel is in the tank. If you can’t do that to the accuracy level required for your airplane, it doesn’t meet the standard.
 
Low tech sight tubes do well for the high-wing crowd. Seeing is believing.

My experimental has a translucent tank that can be seen during preflight. My gauge is accurate but can have some variance due to the capacitive fuel probe. 100LL & Mogas read different.

I know my fuel burn and keep up with a watch/timer. I'm pretty conservative and make it a practice to land with an hour of fuel on board.
 
I was taught by an instructor in his 70s who had been teaching for 50 years. He gave me the best piece of advice I’ve gotten about fuel gauges: if they read more than half, you’re good. If they read less than half, it’s time to land for fuel.

Is this advice for all aircraft or one particular airplane.??
 
Nada. Zip. Zilch.

My fuel gauge is on my wrist. I know what I took off with, I know my anticipated fuel burns, with a reasonable fudge factor, and I know when the tank should be empty.
Even then, if they are hugging empty, believe them. Like fuel flow meters, your watch doesn't account for fuel leaks.
 
The stated accuracy requirement is that you need to be able to determine how much fuel is in the tank. If you can’t do that to the accuracy level required for your airplane, it doesn’t meet the standard.


Okay, what is the accuracy level required for a given airplane? Pick a model and tell me, showing where that number is documented. TCDS, AFM,... I’ve never found a number for my Beech.

If there isn’t one, I can pick any error band I like and arbitrarily say it’s “good enough.” That’s not a requirement, it’s a wild-assed guess.
 
Okay, what is the accuracy level required for a given airplane? Pick a model and tell me, showing where that number is documented. TCDS, AFM,... I’ve never found a number for my Beech.

If there isn’t one, I can pick any error band I like and arbitrarily say it’s “good enough.” That’s not a requirement, it’s a wild-assed guess.
There is not a number published that I’m aware of, but a 3-gallon error isn’t a big deal in a Baron, where it would be a big deal in a Champ.

whether or not you feel the requirement is specific enough, it is a certification requirement.
 
There is not a number published that I’m aware of, but a 3-gallon error isn’t a big deal in a Baron, where it would be a big deal in a Champ.

whether or not you feel the requirement is specific enough, it is a certification requirement.


It’s a requirement that cannot be verified definitely, then, but verified only as a matter of opinion. I had to write, review, edit, and interpret technical requirements throughout my career in aerospace. This one doesn’t hold up.
 
Back
Top