What's new
Van's Air Force

Don't miss anything! Register now for full access to the definitive RV support community.

Looking for cause of engine quit

Blain

Blain

Well Known Member
Building a -8. Bought an RV -7A through insurance auction for the engine and firewall forward components. Contacted the builder-pilot for details of the incident. Very forthcoming and honest.

Certified IO-360-L2A, slick mags, Precision Airmotive injection. 500 TSN.

Pilot report: crossing Sierra Nevada mnt range west to east. Begins decent from approx. 11,500 into Carson City. Sometime during decent realizes the engine is out. Went thru mag check, has fuel pressure, ran out of options and made a successful forced landing except for pole-vaulting over the nose gear and parking it on the lid. Occupants extracted and walk away.

So FAA enters cause as "Icing", which seams unusual with injection and no visible moisture. Inspector doesn't even remove cowling.

Fast forward to this week.

All I need to do is find the "smoking gun". What was the cause?

Removed cowling.

First check the "P" leads for grounding. Switch functions correctly.

Connect fuel source at wing root bypassing tanks. Fuel pump switch on results in good flow and 30 lbs pressure. Small leak at line entering servo.

Secure ship for test start. Anchored to my farm tractor.

Cranks for just a few seconds then fires. Settles into good idle. Oil pressure good, fuel pressure good. Shut down to check everything out.

Fire up again, this time going up to 1500 for mag check. RPM stable.

So I did this for a total of about 15 minutes of run time allowing for cooling since cowl was off.

Now I can't find the smoking gun. Although I bought it well enough to replace mags and injection if I have to but I'd rather determine the cause then just throw parts at it.

Looking for experienced ideas that might lead me to the cause.

Oh, by the way, Have an RV7a "project" airframe looking for a new home:D
 
The "project" sounds tempting. Just finished a -4 project that took almost as long as building one from a kit. Post some pictures of the plane.
 
Just a thought here......

Blain said:
Begins decent from approx. 11,500 into Carson City. Sometime during decent realizes the engine is out.
Click to expand...

I wonder if he was low enough on fuel that the nose down attitude in the decent (which would be pretty steep to get into Carson) unported the fuel pickup in the tank???
 
Been over that range many times, if....he was leaned out for 11,500 Ft, and descended 5,000 Ft and not richen back up, he could have just starved the engine, too lean.
 
What was the temperature at altitude? Could it have been ice crystals forming in the fuel and clogging the filter, then melting and redissolving after the landing? This would depend on the position of the restrict ion relative to the fuel pressure sensor, as you said the fuel pressure indicated normal.
 
What Dave said...

John and Martha, King Schools...Cessna said:

The very kind and competent person on the other end of the line said, ?It?s not in your model year?s handbook, but in subsequent years there?s an explanation of how to prevent this. When an airplane?even one using avgas?flies at altitudes where it?s very cold, the water that is regularly dissolved in the fuel precipitates out in the form of ice crystals, which then can block your fuel system. Then as your fuel system warms up, the ice melts and the water dissolves back into the fuel. When you drain warm fuel from the sumps, you won?t find water, because it is dissolved in the fuel. The solution is to use a fuel system icing inhibitor like the product called Prist, or isopropyl alcohol.?
 
I think John might be right. You probably will never see it happen again... to you anyway. Fly it and have FUN
 
Icing?

So I think of icing as in the throttle body. But ice in fuel now starts to make sense. Subject flight was in March so the possibility of condensation in the tanks would be higher then summer, (at least for this area that has more distinct seasons)

At the same time it's very likely that temps at that FL were well below freezing.

As I think about it however I have to question fuel pressure he said was present. Unless the clog was after the mech fuel pump. This seems unlikely with the temps under the cowl?
 
...

So if it was icing in the fuel, what is the best way to combat that. Can just go buy prist at the gas station. I regularly fly 91 octane and fly high.
 
Blain said:
So I think of icing as in the throttle body. But ice in fuel now starts to make sense. Subject flight was in March so the possibility of condensation in the tanks would be higher then summer, (at least for this area that has more distinct seasons)

At the same time it's very likely that temps at that FL were well below freezing.

As I think about it however I have to question fuel pressure he said was present. Unless the clog was after the mech fuel pump. This seems unlikely with the temps under the cowl?
Click to expand...

Keep in mind there doesn't even need to be much condensation for this to happen. There is almost *always* some water dissolved in avgas. From the moment it leaves the refinery it is beginning to equilibrate with the moisture in the air.
 
So how do you fly internal combustion engines at A flight level that is at or below freezing? Why isn't it a more common occurance?
 
Blain said:
So how do you fly internal combustion engines at A flight level that is at or below freezing? Why isn't it a more common occurrence?
Click to expand...

Great question Blain. If one was looking to cross the Sierra now from the Sacramento area to the Reno area, the temperature aloft forecast indicates:

-16 @ 12,000 feet[/B]

Certainly cold enough to make water freeze.
 
Last edited:
.
It sounds like he was at idle power if it took him awhile to "realize" the engine quit. Not much heat under the cowl if that happened?
Click to expand...

So how do you fly internal combustion engines at A flight level that is at or below freezing? Why isn't it a more common occurrence?
Click to expand...

I wouldn't descend at idle power settings in that temperature range.
Not many pilots do and probably why such occurrences are rare if in fact
the icing theory holds "water"
 
Blain said:
So how do you fly internal combustion engines at A flight level that is at or below freezing? Why isn't it a more common occurance?
Click to expand...

The freezing point of water is variable, changing with the concentration of impurities in the water. Generally, (I don't have access to the appropriate information for the water/avgas system right now), the freezing point of water drops as the concentration of water is decreased. As I said, I don't have the pertinent information right now, but when water is limited to very low concentrations, one would expect that its freezing point would be quite low. Therefore, you have to get the fuel/water solution quite cold to freeze out the water.
 
When ice accretes out of solution it does so in small individual crystals. If the crystals have some point in which they can accumulate, like say at a fine filter element, they can accumulate to form a clog. Certified aircraft with fine fuel filters, particularly with jet fuel which is well known to hold lots of water in suspension, are required to have bypasses. If ice (or other contanimants) accumulate to create a filter block, the bypass opens and keeps the engine running. I can recall many times as an tech, checking these filters after a reported bypass and finding nothing unless that aircraft was still cold, when one might find ice.

So for our avgas airplanes this is much less likely to happen, but it can, particularly if there is a filter that is in an unheated area.
 
Fuel icing is an interesting theory. Let's look at practical details.

First, the guy was in cruise, at a much higher fuel flow, just before the descent. Blockage would be more evident at the high fuel flow, so insufficient fuel supply at idle seems unlikely.

Ok, so let's suppose nosing over sloshed fuel crystals from the low spot in the tank to the fuel tank pickup. A simple tank switch should have restored power, unless you make the further assumption that ice crystals somehow concurrently blocked the other tank outlet, despite there being no prior flow to pack crystals against it.

In an RV, the fuel line enters the warmer cockpit very near the tank, and blockage isn't likely in a length of line anyway. From there I assume the component arrangement specified by Precision; flow first to the fuel valve, then to the boost pump, then to a very fine mesh (32 micron) inline filter.

Precision's 32 micron filter has the potential for ice blockage, but again, blockage should have been more evident at cruise fuel flow than at idle fuel flow. In addition, low flow means more dwell time in the warmer environment of the center console area, and turning on the boost pump would supply 25~30 psi against the screen.

Anything non-typical, like fuel filters in the wing roots? Mogas in the tank? 32 micron filter prior to the boost pump?

You mentioned a small fuel leak at the servo. Are you sure of the source? Look close at the four ram air tubes for any sign of fuel staining.
 
In addition to Dan's analysis on icing what most folks don't kow is that this was a short flight over the Sierra from Auburn - not hours in cruise. I would doubt that the fuel temp dropped more than a degree or two in the tanks during the flight. 11,500 sounds high to many folks - that's just a quick climb for those of us near Tahoe.

The original builder/pilot is our neighbor, so we know a little more than most about it. We were disappointed that their was no further investigation of the cause of the power loass - it remains a head scratcher. BTW - not a case of unporting - he had lots of gas - I stored over twenty gallons that came out of the plane after it was put back on its wheels for awhile.
 
DanH said:
Anything non-typical, like fuel filters in the wing roots? Mogas in the tank? 32 micron filter prior to the boost pump?
Click to expand...

My thoughts exactly.....

Speculating on a cause, without knowing details of the fuel system design is likely pointless (at least in the context of considering the likelihood of this accident in our personal RV).

As the accident data proves, and is repeatedly preached here on VAF, fuel related accidents in RV's are often related to system design (sometimes even the smallest design change can have a very negative and unexpected influence). I don't mean to imply that is the case here, just pointing out that we do not know.
The OP mentioned running the engine with an alternate fuel tank (bypassed the main tank connection to the system. That right there could have eliminated a number of things that could have caused fuel starvation (vent system, fuel pickup problem, etc.)
 
Ironflight said:
BTW - not a case of unporting - he had lots of gas - I stored over twenty gallons that came out of the plane after it was put back on its wheels for awhile.
Click to expand...

Mike S said:
Got it, thanks.
Click to expand...

Was the interior of the tanks inspected?

There has been an accident resulting from a loose fuel pick-up connection at the cover plate bulkhead fitting. Once the fuel level goes below the fitting level, the system starts sucking air. This could easily occur in a transition from cruise to nose down.
 
rvbuilder2002 said:
My thoughts exactly.....

Speculating on a cause, without knowing details of the fuel system design is likely pointless (at least in the context of considering the likelihood of this accident in our personal RV).
Click to expand...
Standard Vans design.

DanH said:
Fuel icing is an interesting theory. Let's look at practical details.

First, the guy was in cruise, at a much higher fuel flow, just before the descent. Blockage would be more evident at the high fuel flow, so insufficient fuel supply at idle seems unlikely.

Ok, so let's suppose nosing over sloshed fuel crystals from the low spot in the tank to the fuel tank pickup. A simple tank switch should have restored power, unless you make the further assumption that ice crystals somehow concurrently blocked the other tank outlet, despite there being no prior flow to pack crystals against it.

In an RV, the fuel line enters the warmer cockpit very near the tank, and blockage isn't likely in a length of line anyway. From there I assume the component arrangement specified by Precision; flow first to the fuel valve, then to the boost pump, then to a very fine mesh (32 micron) inline filter.

Precision's 32 micron filter has the potential for ice blockage, but again, blockage should have been more evident at cruise fuel flow than at idle fuel flow. In addition, low flow means more dwell time in the warmer environment of the center console area, and turning on the boost pump would supply 25~30 psi against the screen.

Anything non-typical, like fuel filters in the wing roots? Mogas in the tank? 32 micron filter prior to the boost pump?

You mentioned a small fuel leak at the servo. Are you sure of the source? Look close at the four ram air tubes for any sign of fuel staining.
Click to expand...

I will get a better look at them. The b nut was slightly loose but fuel pressure was normal so I would think the pump was compensating?

Ironflight said:
In addition to Dan's analysis on icing what most folks don't kow is that this was a short flight over the Sierra from Auburn - not hours in cruise. I would doubt that the fuel temp dropped more than a degree or two in the tanks during the flight. 11,500 sounds high to many folks - that's just a quick climb for those of us near Tahoe.

The original builder/pilot is our neighbor, so we know a little more than most about it. We were disappointed that their was no further investigation of the cause of the power loass - it remains a head scratcher. BTW - not a case of unporting - he had lots of gas - I stored over twenty gallons that came out of the plane after it was put back on its wheels for awhile.
Click to expand...

Thanks for chiming in Paul.
 
Update

As mentioned earlier in this thread I've been looking for the smoking gun. Well on Saturday I may have found a contributor to the loss of power.

Final assembly taking place at airport. Going to start it for the first time in over a year so I removed 1 set of plugs and started pulling prop through to get oil into places it hasn't been in a while.

Then, there it was, Avgas. A puddle forming. Coming from mechanical fuel pump drain.

The diaphragm is leaking.

So here is the question; long decent from about 12,500 down to 4,500. Electric pump off. Injected 360. Could the pump be introducing air into to fuel system?

I would think that a small volume of air would take a while to purge even after the electric pump is switched on.

Could this be the smoking gun?
 
bret said:
Here is an excellent thread on the subject. http://www.vansairforce.com/community/showthread.php?t=73721
Click to expand...

Great pictorial dissection.
Doesn't address how the engine reacts to a leak on that side. Does it just pump fuel out or will it draw air in? My thinking is air is easier to pull in then liquid, therefore at least part of the cycle would involve introducing air into the system.

Then to take it a step further, it seems if there were a pocket of air in an injector line each pressure impulse would be absorbed by that pocket and not transferred to the nozzle.

I'm looking at what could have happened in a short period. Maybe 10 minutes. Low RPM, fuel demand low.

We know the electric pump overcomes the mechanical. So as long as the electric was running, engine ran fine and fuel was probably being pumped overboard. But once that positive feed is removed will the mechanical draw air?
 
Blain said:
Doesn't address how the engine reacts to a leak on that side. Does it just pump fuel out or will it draw air in?
Click to expand...

In theory it can do both. The oil and fuel diaphragms move together, so the volume of the vent space between them is fixed. A hole in the fuel diaphragm thus makes the vent space an extension of the pump chamber. Given a hole in the fuel diaphragm, boost pump pressure at 25 psi would result in fuel flowing out the vent all the time. Boost pump off would make vent chamber pressure (which is atmospheric) higher than fuel chamber pressure during the fuel chamber's expansion stroke, meaning air would enter the fuel chamber. How much would depend on the size of the hole in the diaphragm.

Then to take it a step further, it seems if there were a pocket of air in an injector line each pressure impulse would be absorbed by that pocket and not transferred to the nozzle.
Click to expand...

There is no pressure impulse in the lines to the fuel nozzles. It is literally a constant flow system.
 
Blain said:
Thanks for the explanation, Dan. So do I consider this the cause?
Click to expand...

What Kurt said. Prudence requires an attempt to duplicate the problem. Sometimes it's not possible to do so, but when you can...

Your initial post says you had 30 psi with the boost pump on, and ran it at idle and 1500 RPM. Did you run it with the boost pump off? If it ran fine on the engine driven pump alone, with normal fuel pressure, I wouldn't bet on the leaky diaphragm as the cause of engine stoppage.

Here's a classic, a little piece of **** loose in a fuel system. This airplane had been assembled by a very good team, test flown repeatedly, then flown Texas to Alabama without a problem. The very next time it was flown it scared the snot out of James, followed by a LOT of hunting for a smoking gun.

http://www.vansairforce.com/community/showpost.php?p=641029&postcount=73
 
DanH said:
What Kurt said. Prudence requires an attempt to duplicate the problem. Sometimes it's not possible to do so, but when you can...

Your initial post says you had 30 psi with the boost pump on, and ran it at idle and 1500 RPM. Did you run it with the boost pump off? If it ran fine on the engine driven pump alone, with normal fuel pressure, I wouldn't bet on the leaky diaphragm as the cause of engine stoppage.

Here's a classic, a little piece of **** loose in a fuel system. This airplane had been assembled by a very good team, test flown repeatedly, then flown Texas to Alabama without a problem. The very next time it was flown it scared the snot out of James, followed by a LOT of hunting for a smoking gun.

http://www.vansairforce.com/community/showpost.php?p=641029&postcount=73
Click to expand...

Did not run it with boost off. It was a ground start on damaged airframe. I limited it to just a few minutes. Never got out of cabin during run up or I may have saw fuel dribble. Really didn't even think of trying that at the time. I was just excited to have a running engine!
 
There are two diaphragms, creating a chamber in between. This chamber has the overflow outlet. That outlet has a restrictor in it. It is designed so that this chamber fills with fuel when the main diaphragm leaks, allowing the second diaphragm to provide the backup pumping function (chamber fills with fuel, as Dan mentioned). The restrictor is sized such that enough pressure can still be exerted by the second diaphragm upon the gas in the main chamber (ruptured main diaphram and filled chamber allows upper diaphram to act as the main diaphragm). Some fuel is expelled through the restrictor on each pump cycle, so there may be some evidence of this if it happened.
 
Video

That was educational. It leads me to believe that if the chamber is full of fuel and draining overboard there may not be an open route for air to enter the pump.
 
Ironflight said:
In addition to Dan's analysis on icing what most folks don't kow is that this was a short flight over the Sierra from Auburn - not hours in cruise. I would doubt that the fuel temp dropped more than a degree or two in the tanks during the flight. 11,500 sounds high to many folks - that's just a quick climb for those of us near Tahoe.

The original builder/pilot is our neighbor, so we know a little more than most about it. We were disappointed that their was no further investigation of the cause of the power loass - it remains a head scratcher. BTW - not a case of unporting - he had lots of gas - I stored over twenty gallons that came out of the plane after it was put back on its wheels for awhile.
Click to expand...

Paul, was it AVGAS or auto gas?
 
sailvi767 said:
Paul, was it AVGAS or auto gas?
Click to expand...

Avgas, and very clean. After talking with the original owner, I believe that the leaking pump is a very high contender for the cause of this. Can we prove it? No - but I have heard no other more likely and reasonable solution. We will probably never be sure, but changing the pump like Blain is doing is the right thing regardless.

Paul
 
Steep decent and fuel problems

I know for sure that in my aircraft that if I use a steep decent with the throttle pulled all the way to an idle, and when the fuel tanks are less than 1/2 full, the fuel will move forward in the tanks uncovering the fuel pick-up. I have a sensor that goes off when air is coming into the fuel system.
 
RV3bpilot said:
I know for sure that in my aircraft that if I use a steep decent with the throttle pulled all the way to an idle, and when the fuel tanks are less than 1/2 full, the fuel will move forward in the tanks uncovering the fuel pick-up. I have a sensor that goes off when air is coming into the fuel system.
Click to expand...

His descent wasn't steep enough to create a significant nose down attitude - 6,000 feet in about fifteen miles. And there was 2/3 fuel, so it would be pretty tough to unport the tanks.
 
This thread is looking like a game of battleship. So, Paul, is the"a" theory that that the mechanical pump failed, released debris then it plugged something downstream of that? The op said it had fuel pressure. It seems it would flow fuel with fuel pressure ( measured downstream of the pump) unless there is an obstruction. Even if the tank fuel was not low temp, the wing root connection could cool the fuel rapidly , but not through a 3/8" tube with 6-7-8 gph. That is pretty fast moving unless it was near the critical temp. Even if that were so, it would seem to plug the boost pump filter first, resulting in no fuel pressure.

So, is there any chance that the air filter was plugged with ice? No mention of the NTSB report or temps are posted here.


I would think taking the servo and distributor apart and inspecting and cleaning, or sending it to Don for same would be in order. Also, removing every single hose, tube and (or) ensuring it has no obstruction or debris, then doing a first flight style series of pressure (leak) and flow testing would be prudent to ensure that all (non icing) bases are covered. Just my 1 cent.
 
Ironflight said:
His descent wasn't steep enough to create a significant nose down attitude - 6,000 feet in about fifteen miles. And there was 2/3 fuel, so it would be pretty tough to unport the tanks.
Click to expand...

Ice in the fuel system could have certainly been the cause if the actual temp of the fuel was below freezing. Since we don't have a fuel temp gauge it's hard to know. Based on the flight profile the fuel could have been below freezing. If we knew the temp at the departure station and how long the aircraft sat their before the flight we could make a pretty good guess. Fuel icing brought down a British Airways 777 so it could be a factor.
 
You can also pull the spark plugs and run it on the starter with the ignitions off. It will either create a fuel pressure indication or fuel will pour out the pump overflow tube.

Vic
 
vic syracuse said:
You can also pull the spark plugs and run it on the starter with the ignitions off. It will either create a fuel pressure indication or fuel will pour out the pump overflow tube.
Click to expand...

I dunno Vic. That would be further away from the conditions present when the engine quit running....throttle plate closed, elevated RPM, pump case warm. Throttle plate closed means very low fuel flow; the diaphragm(s) would be operating with little displacement, near the top of the stroke. Elevated RPM has a subtle effect here; it determines the velocity at which the diaphragm is raised. Temperature relates to vapor formation pressure.
 
Last edited:
I checked it on my RV10 before I posted, as I am in the process of chasing down some fuel flow indication problems.

With plugs removed, and cranking with starter, engine pump showed 7-10 PSI.

Vic
 
Defective pump removed

New Tempest on workbench ready for install. I got fuel out of the overflow by hand propping with plugs out. Thats how I discovered it.

I think the only question at this point is will the following conditions allow air to be introduced into the servo.

Long decent, electric not on.

Hand propping seemed to produce only an outflow of fuel. I have silicone line so it was visible.

Question is what will the pump do at say 1200 rpm? Will it suck air back in or push it out?
 
BillL said:
This thread is looking like a game of battleship. So, Paul, is the"a" theory that that the mechanical pump failed, released debris then it plugged something downstream of that? The op said it had fuel pressure. It seems it would flow fuel with fuel pressure ( measured downstream of the pump) unless there is an obstruction. Even if the tank fuel was not low temp, the wing root connection could cool the fuel rapidly , but not through a 3/8" tube with 6-7-8 gph. That is pretty fast moving unless it was near the critical temp. Even if that were so, it would seem to plug the boost pump filter first, resulting in no fuel pressure.

So, is there any chance that the air filter was plugged with ice? No mention of the NTSB report or temps are posted here.


I would think taking the servo and distributor apart and inspecting and cleaning, or sending it to Don for same would be in order. Also, removing every single hose, tube and (or) ensuring it has no obstruction or debris, then doing a first flight style series of pressure (leak) and flow testing would be prudent to ensure that all (non icing) bases are covered. Just my 1 cent.
Click to expand...

My ground run was only with the boost pump on. In retrospect I realize I should have shut it off to validate the function.

FAA final did not include an icing theory.
 
BillL said:
This thread is looking like a game of battleship. So, Paul, is the"a" theory that that the mechanical pump failed, released debris then it plugged something downstream of that? The op said it had fuel pressure. It seems it would flow fuel with fuel pressure ( measured downstream of the pump) unless there is an obstruction. Even if the tank fuel was not low temp, the wing root connection could cool the fuel rapidly , but not through a 3/8" tube with 6-7-8 gph. That is pretty fast moving unless it was near the critical temp. Even if that were so, it would seem to plug the boost pump filter first, resulting in no fuel pressure.

So, is there any chance that the air filter was plugged with ice? No mention of the NTSB report or temps are posted here.


I would think taking the servo and distributor apart and inspecting and cleaning, or sending it to Don for same would be in order. Also, removing every single hose, tube and (or) ensuring it has no obstruction or debris, then doing a first flight style series of pressure (leak) and flow testing would be prudent to ensure that all (non icing) bases are covered. Just my 1 cent.
Click to expand...

First thing I did was send the servo to Don. No defects found. All new lines from TS Flightline.

Guess I should have addressed these with multiple quotes. I may get a scolding from DR:eek:
 
You must log in or register to reply here.
Back
Top