Engine stumbled two back to back flights

[Hank]

36 minutes ago, hazek said:

If you run your tank dry and the engine stumbles and you switch tanks throttle is open though?

Yes. It's best to reduce the throttle to prevent overspend on restart, but not to Idle/Cut Off.

[Shadrach]

1 hour ago, hazek said:

If you run your tank dry and the engine stumbles and you switch tanks throttle is open though?

I try to avoid letting the engine quit as I think it can be easily avoided. The fuel pressure needle will give a fair amount of warning ahead of stumble. That being said, I have run tanks dry in flight many times before I became more considered in my approach. It never produced an afterfire nor an overspeed.  I've never reduced throttle when switching tanks after running a tank dry and never had an issue, however that does not mean it couldn't. I cruise at 2450-2500rpm. At that prop setting, any RPM fluctuations that occur when bringing the engine back online are well within normal operating range.  If one were cruising at 2700 and let the engine starve and then reintroduced fuel, it's possible that the engine might exceed redline for a short time. It's a good practice to reduce the throttle before reintroducing fuel, but I'd bet that many pilots are so eager to relight the fire that they neglect to do so.

[jetdriven]

The M20J will certainly overspeed.  There’s even a warning in the POH about reducing throttle rapidly at high indicated air speed that would cause it. It must be the design of the prop and the governor, even doing a go around or a stall recovery it will hit 2970 RPM and that’s actually in the POH for I think five seconds maximum, not because they think it’s a great idea, but probably because they couldn’t prevent it.

[Pinecone]

It you are expecting it (trying to run tank down), you should be ready to switch at the first sign of stumbling.  Which is NOT running the pump dry.

If carb, you can see the fuel pressure waver and not even have a single hiccup.

[Pinecone]

2 hours ago, cbarry said:

I do not see any legitimate benefit in deliberately running a tank dry.  Call me conservative, but I’m in the camp of the old WWII pilot sayings like “the bottom half of the tank is there to hold up the top half!”  

If you are pushing max range, you may want to run one tank dry so that you have all the fuel in the other tank.  Say you plan on landing with 10 gallons.  Which is one hour at cruise power.

If you have 5 in each tank, and say you have to go around and due to traffic, re-enter the pattern, there is a change you will forget to switch from the now very low tank to the other one.  

If you ran one dry, you have all the 10 gallons in the one tank and don't have to possibly remember to switch.

[hazek]

On 8/9/2025 at 9:57 AM, hazek said:

I was always wondering how does one preserve the most fuel on the fullest tank for that landing but I thought running a tank dry is extra wear if not actually harmful for the fuel pump? Plus a backfire might happen which harmful for the exhaust. Do people really do this routinely?

So, do we now agree that my original questions were valid?

[Shadrach]

8 hours ago, cbarry said:

I do not see any legitimate benefit in deliberately running a tank dry.  Call me conservative, but I’m in the camp of the old WWII pilot sayings like “the bottom half of the tank is there to hold up the top half!”  

Have you ever seen the inside of an M20 fuel tank up close?  I appreciate a good folksy expression as much as the next guy but there is real utility in concentrating remaining fuel in one tank. There is little chance of "sucking up" debris as some theorize.  Once one tank has been exhausted, one has completed one's fuel management duties for the flight. If one needs to divert or hold, there is no question about where the selector should be.  There are more than a few NTSB reports where the NTSB discovered many gallons of useable fuel in the tank not selected.

 

[EricJ]

18 minutes ago, Shadrach said:

Have you ever seen the inside of an M20 fuel tank up close?  I appreciate a good folksy expression as much as the next guy but there is real utility in concentrating remaining fuel in one tank. there is no chance of "sucking up" debris as some theorize.  Once one tank has been exhausted, one has completed one's fuel management duties for the flight. If one needs to divert or hold, there is no question about where the selector should be.  There are more than a few NTSB reports where the NTSB discovered many gallons of useable fuel in the tank not selected.

 

For a moment I was admiring your carpeted fuel tank. 

[Shadrach]

37 minutes ago, EricJ said:

For a moment I was admiring your carpeted fuel tank. 

Keeps things classy in the hangar and adds 5 minutes to the job for every fastener that's dropped...

[jetdriven]

Another thing is your totalizer may show 10 gallons remaining, which is an hour as Terry said, but the gauging systems we have arent accurate enough to tell you if it’s 2 gallons in one side and eight and the other or the vice versa. But if you run it all out of one side, then you know where it’s all at. And then you don’t have to worry about uncovering a fuel pick up because you have 10 gallons, or was it eight, or was it six.  
if you catch it fast enough, you can switch when the fuel pressure starts to flicker before you have a power interruption on a fuel injected engine, but it’s not always possible to get it, ATC calls, or you look away.  But this should be a non-event and recovering from it should be a non-event too.

[Hank]

4 hours ago, Pinecone said:

If carb, you can see the fuel pressure waver and not even have a single hiccup.

I've purposely run tanks dry twice in my C, both times planning to switch at the top of descent, making for easy landing with ~90 minutes fuel and 20-25 minutes flying time. 

The first time, the engine surged a couple of times rather like my lawn mower, and I switched tanks with no issues. The second time I was waiting and watching, and the engine smoothly shut down with no warning, surprising me a bit, and the RPM briefly went to redline on restart. 

[Skates97]

5 hours ago, Pinecone said:

It you are expecting it (trying to run tank down), you should be ready to switch at the first sign of stumbling.  Which is NOT running the pump dry.

If carb, you can see the fuel pressure waver and not even have a single hiccup.

Having a carb you get a fair amount of warning as the fuel pressure slowly drops. I've been running a tank dry on long cross country flights for almost nine years now. Never any issues. Three times I didn't catch the fuel pressure drop and the engine quit. Not a big deal. You will also see a surge in fuel flow as the bowl refills after switching tanks. Also as Mike mentioned early in the thread, the next time you switch to that tank there may be a brief interruption in flow as the lines that had air in them fill with fuel. Once I ran a tank dry, switched over, and finished the flight. Topped both tanks and started on the tank I landed on. One hour into the flight I switched tanks and saw my fuel pressure and flow start dropping so quickly switched back thinking "What was that?" Then I realized it was probably air in the lines from the previous flight. Switched back and after a brief moment the pressure and flow came back to normal and continued on.

5 hours ago, Pinecone said:

If you are pushing max range, you may want to run one tank dry so that you have all the fuel in the other tank.  Say you plan on landing with 10 gallons.  Which is one hour at cruise power.

If you have 5 in each tank, and say you have to go around and due to traffic, re-enter the pattern, there is a change you will forget to switch from the now very low tank to the other one.  

If you ran one dry, you have all the 10 gallons in the one tank and don't have to possibly remember to switch.

This is why I run it dry, my personal minimum is on the ground with 10 in the tank I'm landing on, not 10 gallons split between the tanks.

[redbaron1982]

I've never tried running a tank dry, and I'm kind of reluctant to do it.

One time, I was messing with the mixture while at 8kft WOT, and I pull it to far out, too quickly, and the engine kind of shutdown, and the RPMs surged, almost to the point that I would have to get the prop inspected.

I imagine that the effect would be similar between running out of fuel or pulling the mixture out too quickly. I don't know if my governor has issues or what. I asked here in the forum and they say it was normal.

Do you guys experience any RPM surge when the engine dies off after running a tank dry?

[jetdriven]

Yes, it’s quite normal. Some airplanes over speed pretty badly and some don’t seem to be bothered by it. But your first instinct ought  to be to bring the throttle back to about half, and then switch tanks and then after it stabilizes then bring the power back on.   The prop goes to low pitch to try to maintain on speed, and in a Mooney, for example, it won’t maintain on speed so it goes to full flat pitch, and then, as soon as the power comes on, the torque rises so fast the governor can’t react soon enough. Go to any airport and listen to airplanes takeoff, it seems to be exclusively serious guys, but some Bonanza guys will do this too, they the firewall the throttle so fast that it  overspeeds and surges on the takeoff roll. What do you think it does when you’re traveling 150 miles an hour. It could be a lot worse.

[Will.iam]

Interesting that the lycoming (j model) referenced can and does overspeed on resupplying fuel to the engine. My 252 (k model) continental engine (with Macaulay prop governor) does not overspeed on relighting of the engine. Also my governor is fully functional when the engine is windmilling in the glide as i can go to fully low pitch and this reduces the drag from the windmilling propeller drastically and when i push the prop lever back full forward for high pitch (2700rpm) as the propeller increases in speed the drag induced feels like someone deployed a parachute to slow down. In VVI terms I’m sinking at 800ft/min but pull the prop lever back and as the propeller slows to min rpms my VVI reduces to 350ft/min. This is huge if you are trying to glide to an emergency landing site or need more time to troubleshoot things before hitting the ground. The governor runs on oil pressure and a windmilling propeller turns the engine fast enough to create that oil pressure to operate the McCauley prop governor in my experience. 

[wombat]

28 minutes ago, cbarry said:

Again, the one time the fuel selector fails to feed from the tank containing useable fuel (after running the used tank “dry”), you’ll have fuel exhaustion and fuel starvation in the same incident report.

This is not a reasonable argument.

Sudden and unexpected fuel selector failure is not a significant risk and you are not significantly increasing safety by your methods.   By the time you've run the tank dry, you've probably already operated that system 5 or more times during that flight. 

There are way more fuel starvation accidents where the pilot neglected to switch fuel due to high workload prior to landing than there are fuel starvation accidents where the selector failed in flight.  In each and every one of the former, having run the other tank dry would have saved the situation.  If any of the latter have occurred at all, only a subset of those (where there was not an airport suitable for landing withing gliding range) would have caused a problem.

 

I have an airplane to be useful and the range is part of what makes it useful.  If I were to cut that range in half or less, that would make it much less useful.

Now there are some of us who only fly for pure pleasure to get up in the air.  In that case, yeah, sure, never go below 75% fuel.  But that's not what I do and the additional safety of never needing to switch fuel tanks in flight is probably lower than the increased accident rate due to increased number of landings anyway.

[Shadrach]

57 minutes ago, cbarry said:

Again, the one time the fuel selector fails to feed from the tank containing useable fuel (after running the used tank “dry”), you’ll have fuel exhaustion and fuel starvation in the same incident report.

Agree to disagree... while selectors do sometimes fail, there is a reason that the industry differentiates between fuel exhaustion and fuel starvation, and the reason isn't fuel selector failure.

[PT20J]

2 hours ago, Will.iam said:

Interesting that the lycoming (j model) referenced can and does overspeed on resupplying fuel to the engine. My 252 (k model) continental engine (with Macaulay prop governor) does not overspeed on relighting of the engine. Also my governor is fully functional when the engine is windmilling in the glide as i can go to fully low pitch and this reduces the drag from the windmilling propeller drastically and when i push the prop lever back full forward for high pitch (2700rpm) as the propeller increases in speed the drag induced feels like someone deployed a parachute to slow down. In VVI terms I’m sinking at 800ft/min but pull the prop lever back and as the propeller slows to min rpms my VVI reduces to 350ft/min. This is huge if you are trying to glide to an emergency landing site or need more time to troubleshoot things before hitting the ground. The governor runs on oil pressure and a windmilling propeller turns the engine fast enough to create that oil pressure to operate the McCauley prop governor in my experience. 

I know what you meant, but so as not to confuse others, prop control pulled back gives low rpm and high pitch. It’s the drag from the high pitch that causes the low rpm. 

[PT20J]

This whole thread seems to be getting a little silly. Some obviously like to run their tanks dry and some don’t. Each to their own.  Personally, I don’t see any great (actual, not theoretical) risk in doing so and I see no operational necessity to do so routinely. 

[Pinecone]

19 hours ago, cbarry said:

Again, the one time the fuel selector fails to feed from the tank containing useable fuel (after running the used tank “dry”), you’ll have fuel exhaustion and fuel starvation in the same incident report.

You can't fix everything.   But if you were running on the fuller tank, it is very unlikely that it will fail that way at that exact time.

 

[Pinecone]

16 hours ago, PT20J said:

This whole thread seems to be getting a little silly. Some obviously like to run their tanks dry and some don’t. Each to their own.  Personally, I don’t see any great (actual, not theoretical) risk in doing so and I see no operational necessity to do so routinely. 

I can see the value.   There is no "right" answer, what works best for you.

My personal way is to run the one tank to the low flue light (2.5 -3 gallons) and also plane on having at least 10 gallons in the other tank.  But if I really needed to push the range, I would not hesitate to run one dry.  The only want I could see my getting to this point is if something like a long XC and someone lands gear up  and closes the airport, and I have an unplanned divert.

[Hank]

23 minutes ago, Pinecone said:

I can see the value.   There is no "right" answer, what works best for you.

My personal way is to run the one tank to the low flue light (2.5 -3 gallons) and also plane on having at least 10 gallons in the other tank.  But if I really needed to push the range, I would not hesitate to run one dry.  The only want I could see my getting to this point is if something like a long XC and someone lands gear up  and closes the airport, and I have an unplanned divert.

Some of us do not have "low fuel lights" except in our ground vehicles. And I've had as many as five (5!) reroutes on what should have been a simple 505nm direct flight. (Three were due to MOAs going hot just before I entered them, and once when I was 5 miles inside the border.) That trip ended up at just over 4.5 hours. While I didn't run the tank dry, I did take it to 2.5 hours (26 gal per tank, no LOP with my carb). I felt better knowing I had at least 1.5 hours, probably 2 hours, remaining in the last tank while still a state away from landing.

Having run my tanks dry before, I had a good idea how long I could trust it before needing to watch the Fuel Pressure needle, and that made me feel a lot better.

[Hank]

10 minutes ago, cbarry said:

3 hours ago, Pinecone said:

You can't fix everything.   But if you were running on the fuller tank, it is very unlikely that it will fail that way at that exact time.

 

 
Very unlikely, yes but it’s important to remember the M20C accident in 2017 whereby fuel selector became obstructed (by remnants of a shop towel from who knows when) as simply an example of why deliberately, but unintentionally, paint oneself into a corner when there’s no legitimate reason to do so.  

That scenario will stop the engine when switching tanks any time, regardless of fuel level.

Same thing with an accident I remember where either a screw or pebble fell into the fuel selector area of the floor and jammed it when switching tanks.

Again, this is not affected by fuel level, the selector can jam on the first switch after filling.

If you are theorizing that this happens when the fuel line develops an internal obstruction, it could again happen at any time. When running a tank dry, it is typically to reach maximum range, and occurs after switching tanks several times, each an opportunity for an internal obstruction. Suppose this happens to you, @cbarry, when the tank you are switching away from has been run down to 5-6 gallons? That's below the minimum amount for departure (in case of missed appro or the need to go around) and may well unport if you try to descend quickly for that nearby airport . . . 

[Will.iam]

On 8/11/2025 at 9:08 PM, PT20J said:

I know what you meant, but so as not to confuse others, prop control pulled back gives low rpm and high pitch. It’s the drag from the high pitch that causes the low rpm. 

I don’t think so. It’s the blade pitched more toward the wind like a fethered prop just not as extreme but the drag is way reduced at the low rpm in fact if i could pitch the blade into a full feather i. E. Blade is parallel to the air the propellor and thus the engine would stop and i would have zero rpm’s and the least amount of drag. If i push the prop control full forward and command a min pitch / blades are almost perpendicular to the air then the air is pushing the hardest and i get maximum rpm and i really feel the drag come on more so than speed brakes and about the same a applying full flaps. 

[Hank]

Prop control fully forward puts the blades flatter to the relative wind, a much higher drag condition than prop control fully back. 

Prove it to yourself easily by pulling the throttle to idle, then pull the prop control fully out and feel the unpowered airplane accelerate. That's why we practice pulling the prop back in engine-oit scenarios, because the gliding distance is longer.