Engine Failure yesterday

68 posts in this topic

Posted (edited)

Few comments on a fuel pump.  I do have a Facet fuel pump. Funny enough I just reached 1000 feet and switched it off, it was  then that the engine died..... within seconds.

Thr Lake Amphibian has an engine driven pump with an elect boost backup.  

Procedure there was to keep your finger on the elect pump switch when you turned it off... keep an eye on fuel pressure guage.... and be ready to flip it back on if you noticed any pressure drop or motor stumble.

Engine way up high on that plane, tank down low.  Got a guy into good fuel pressure monitoring habits.  The elect pump was the first thing you went for if the motor hiccuped.

Edited by Yamma-Fox

Share this post


Link to post
Share on other sites

Posted

Yep fuel pump was the first thing I went for.....

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

Yeah,  it must have been a total coincidence in this case...  no matter if the elect pump was on or off.

Not sure if it helps figuring things out on your case,  but my KF 2 would have trouble transferring fuel down from the wing tank to the 6 gall main (behind the firewall) in flight.

I would land and then the flow would start down.  Seems like the fuel cap vent on the firewall tank would overpower the fuel cap vent on the wing tank.  More pressure in the firewall tank (which was basically a huge header tank) seemed like it was keeping the fuel from flowing down during flight.

With that experience, I agree with nlappos and his theory:  wing tank (without cap) under neg pressure which made flow down to the header stop (like my kf2 would).  Elect pump could pull it but the vac pump couldn't.

But in this theory I still have trouble understanding how you would ended up with a header tank full of AIR....  I can see how it may have been under neg pressure, but I can't see where the air would be introduced....  unless it could be pulled back backwards through the vent line by the fuel pump.  But your vent line has a SOV on it.

The air in the header is what I can't explain unless there is an open vent.

Edited by Yamma-Fox

Share this post


Link to post
Share on other sites

Posted

More strange details "So I stuck a jerry can under the header tank and opened the bleed/vent valve.......it to quiet a while gurgling and bubbling and filling up before it started to bleed out the bottom".

As long as fuel from the main tank (or air from a vent) can get into the header tank, any fuel in the header tank should immediately bleed out the bottom. Why would there be a need for "filling up" before starting to bleed from a bottom valve? It sounds like you are the bleeding from the air vent valve and not from a bottom purge valve... Are you sure the "from header tank to engine" is taken from a low position on the header tank?

Share this post


Link to post
Share on other sites

Posted

Here are some photos

 

IMG_20171220_193912.thumb.jpg.5fef41a023IMG_20171220_193847.thumb.jpg.81e112cf39IMG_20171220_193927.thumb.jpg.22fb00a065

Share this post


Link to post
Share on other sites

Posted (edited)

FWIW,  that vent line that you have (with the valve that I assume goes down to a belly drain) goes, in my KF5, back up to top of the wing tank with no valve.

Seems like that is a more moden and possibly better design with a vent that is always open (any fuel slosh through it returning to tank).  With the open vent there can not be air in the header tank at any time (unless wing tank is totally empty).  

Being openly vented back to the tank also helps prevent an air lock in the fuel line:

With your design it seems possible that, if air was in the header tank, that there would be no way for it to get out (with your vent valve closed).  In that case you would have the Airlock:  air in header trying to rise the only way it can go... back UP your fuel feed line!

Without enough positive pressure (fuel cap) to push fuel through the airlock... and not enough suction (elect fuel pump off) it is possible that the airlock could win the battle and stop flow.

Edited by Yamma-Fox

Share this post


Link to post
Share on other sites

Posted

This looks right.
On the side of the header tank, 3.5in from the bottom, is where you should have a line to the engine (and you mentionned a valve close to the firewall so you can turn off the fuel flow if so needed...) .
On the bottom of the header tank there should be a purge valve to empty the whole system...
On top there is one line arriving from the tank (with what appears to be a closed valve but I assume it was open during the flight..)
Also on the top, a second line that appears to go to the bottom of the plane so that when this valve (with the arrow and "open") is opened to let the air out when filling it up you don't get fuel inside the plane. Unless this last line is going to the engine (and not to let air and fule out under the plane) it looks good. 

Share this post


Link to post
Share on other sites

Posted

FredStork that is exactly right.

Share this post


Link to post
Share on other sites

Posted

Yamma-Fox, the problem with having an always open vent on the header tank is when you have a reduced fuel flow due to dirt in the bottom of your main tank...
With a partially clogged fule line from the main tank to the header tank there is a risk that the consumed fuel is replaced with air from the open vent rather than fuel. With a pump in the system, as most of us have, the pump will not make the difference until the header tank is empty and the pump start chewing air... With a cosed vent the pump might be able to suck fuel, rather than air, through the clogged line... 
 Note that air in the header tank is not a problem unless there is only air in the tank, with a closed vent the volume of air will remain constant. Use fuel will be replaced with new from the header tank and the air will remain. In my plane there is a an alarm that goes off when the level in the header tank start to go down, i.e. when the first wing tank is empty, I switch to the other wing tank and open the header tank vent valve until the alarm stop blinking... Next time it start blinking I better land...

Share this post


Link to post
Share on other sites

Posted (edited)

FredStork that is exactly right.

So the mystery remains... starting with a far from empty main tank and assuming you were not banking heavily to the right during the entier flight, there is no way your header tank could contain more air after the flight than it did before... There is no way in for the air... 
Maybe you had closed the valve between the main tank and the header tank after your exciting landing, "just in case"... If so, and if it remained closed when you started to purge the fuel before folding the wings, there could be a lot of gurgling and bubbling when air goes up through the purge and fuel goes down, and so with very limited flow in the beginning... This could give the impression of the header tank filling up... 

 

Edited by FredStork
1 person likes this

Share this post


Link to post
Share on other sites

Posted

FredStork that is exactly right.


Maybe you had closed the valve between the main tank and the header tank after your exciting landing, "just in case"... If so, and if it remained closed when you started to purge the fuel before folding the wings, there could be a lot of gurgling and bubbling when air goes up through the purge and fuel goes down, and so with very limited flow in the beginning... This could give the impression of the header tank filling up... 

 

I think that's it.  I shut everything off before exiting he aircraft.

Share this post


Link to post
Share on other sites

Posted

Fred, 

I think the header tank was drained by the engine, which quit when the header was empty. The header was not refilling, because the pressure difference (suction) from the main side was too great.

I drew up a simple sketch of the single tank Avid system

Fuel System.jpg

1 person likes this

Share this post


Link to post
Share on other sites

Posted (edited)

Fred, 

I think the header tank was drained by the engine, which quit when the header was empty. The header was not refilling, because the pressure difference (suction) from the main side was too great.

I drew up a simple sketch of the single tank Avid system

Fuel System.jpg

I'm sorry to say but I think you are wrong and the reason is that there is no way the header tank could have been drained.

On your drawing (correct according the Gfry and his photos) we should assume that the bleeder valve, going out under the belly, was closed during the flight. Another given is an unknown amount of air in the header tank as shown in your drawing. 

We now need to find a way to remove the fuel from the header tank and replace it with air... Where do we get that air from?
Not from the bleeder valve because it is closed.
Not from the main tank as it is still far from empty.
Not from the engine side as there is where we are pulling out the fuel. 
With lower pressure the same amount of air take more volume but not enough to empty (fill?) the tank unless it was almost empty of fuel from the beginning. It is more likely that it was almost full of fuel. The lable with the arrow and "open" (on the photo) tend to indicate that purging is a part of the pre flight.

In Gfry's last coment he "admits" that he started to drain the header tank with the valve between main and header closed. This is likely to gurgle and bubble and giving the impression the header tank is filing up while it in reality is air comming up through the purge valve.
The header tank was never empty - it could not have been...

I think the explanation is that as long as the pump was running it countered the negative pressure in the main tank and successfully sucked/pulled fuel through the system. The moment the pump was turned off, the flow was inversed and air came, from the carburetor, to the pump and further. Air in the pump prevented restart of the flow.
(This flow of air through the system would not have altered the volume of fuel in the headertank, the air would have been sucked through the fuel still in the headertank before it continued thrught the line up to the main tank and freedom...)

Edited by FredStork

Share this post


Link to post
Share on other sites

Posted

Fred, 

I think the header tank was drained by the engine, which quit when the header was empty. The header was not refilling, because the pressure difference (suction) from the main side was too great.

I drew up a simple sketch of the single tank Avid system

Fuel System.jpg

I'm sorry to say but I think you are wrong and the reason is that there is no way the header tank could have been drained.

On your drawing (correct according the Gfry and his photos) we should assume that the bleeder valve, going out under the belly, was closed during the flight. Another given is an unknown amount of air in the header tank as shown in your drawing. 

We now need to find a way to remove the fuel from the header tank and replace it with air... Where do we get that air from?
Not from the bleeder valve because it is closed.
Not from the main tank as it is still far from empty.
Not from the engine side as there is where we are pulling out the fuel. 
With lower pressure the same amount of air take more volume but not enough to empty (fill?) the tank unless it was almost empty of fuel from the beginning. It is more likely that it was almost full of fuel. The lable with the arrow and "open" (on the photo) tend to indicate that purging is a part of the pre flight.

In Gfry's last coment he "admits" that he started to drain the header tank with the valve between main and header closed. This is likely to gurgle and bubble and giving the impression the header tank is filing up while it in reality is air comming up through the purge valve.
The header tank was never empty - it could not have been...

I think the explanation is that as long as the pump was running it countered the negative pressure in the main tank and successfully sucked/pulled fuel through the system. The moment the pump was turned off, the flow was inversed and air came, from the carburetor, to the pump and further. Air in the pump prevented restart of the flow.
(This flow of air through the system would not have altered the volume of fuel in the headertank, the air would have been sucked through the fuel still in the headertank before it continued thrught the line up to the main tank and freedom...)

Makes sense to me

 

 

Share this post


Link to post
Share on other sites

Posted

I had a forced landing a year ago. Fuel starvation. I have 2 poly insert tanks in the right wing. After flying an hour I looked up at the fuel level super low on fuel. Turned back to the airport watching the air bubbles going into my header tank. I learned I only have about 4 minutes of cruise running off the header tank. My engine never sputtered just quit.  Lucky I had a harvested bean field within gliding range. Landed without any trouble. After rolling to a stop and taking a few deep breaths I heard a tricking noise. Looked up at my inboard tank and it was filling up with fuel and trickling down into my header tank. Went outside and found the vent tube on the out board tank was facing backwards with fuel all over my wing. It must have created a vacuum in the out board tank, so the inboard and header tank ran dry. I just let the outboard tank drain into the inboard tank then turned the vent back into the wind, started it up and flew back to the airport with no problems since. Make sure your vent tubes are facing forward. Wasn't rocket science to figure that one out.

Share this post


Link to post
Share on other sites

Posted

I did  switch it on as soon as it started to splutter.  However it seemed useless as when I checked the header tank back on the ground it was bone dry.  My guess is the fuel venting backwards then created an airlock in the header tank.

I would say your engine burned the fuel in your header tank.  Negative pressure wouldn't let fuel flow to your header tank.  Been there done that.

Share this post


Link to post
Share on other sites

Posted (edited)

Ok Fred, I'll respectfully spar with you here a little!  :)

I hear your argument against an open vent, and conceed that yes if BOTH finger strainers were plugged up then the open vent system could pull air.

But in that extremely rare case I don't think the closed vent header design would have very much more of a chance pulling through the clog anyway... marginal advantage.

The wing to header is designed to be a gravity fed system, and IMO nothing should impede that..  definitely not a column of air in the header... thats an airlock!   You might be right that a strong pump can pull enough neg pressure in the header to force fuel down past that airlock, but that doesn't sound ideal to me at all.  Without a strong enough pump you are SOL!

They have switched to the vented header on all newwer designs... right??   There probably is good reason for that.

Edited by Yamma-Fox
1 person likes this

Share this post


Link to post
Share on other sites

Posted

Interesting discussion. Since you moved the plane back to your hanger or  shop, have you checked to see what kind of fuel flow you get from the wing tank to the header? besides the vacumn being created by the missing cap, there's also a question of how well the fuel is actually flowing through the finger screen and any in-line filter that may be installed. Just wondering.

Jackak

1 person likes this

Share this post


Link to post
Share on other sites

Posted

Ok Fred, I'll respectfully spar with you here a little!  :)

I hear your argument against an open vent, and conceed that yes if BOTH finger strainers were plugged up then the open vent system could pull air.

But in that extremely rare case I don't think the closed vent header design would have very much more of a chance pulling through the clog anyway... marginal advantage.

The wing to header is designed to be a gravity fed system, and IMO nothing should impede that..  definitely not a column of air in the header... thats an airlock!   You might be right that a strong pump can pull enough neg pressure in the header to force fuel down past that airlock, but that doesn't sound ideal to me at all.  Without a strong enough pump you are SOL!

They have switched to the vented header on all newwer designs... right??   There probably is good reason for that.

Hi Yamma-Fox!

It is good to be challenged – either I’m wrong or didn’t express myself clearly.

It is often easier to resolve a problem by using extremes – like “empty”, “completely clogged” etc.  but from that you then have to go back a step or two. Maybe I only did that in my mind.

If you are using an old fiberglass tank you should expect there to be “things” floating around that you prefer not to have there… One day those “things” might end up in the wrong place and while not completely blocking the flow it will reduce the flow.

If the remaining (reduced) flow is slower than the your actual fuel consumption you will, with the header tank vent closed, get help from the pump to increase the flow. Hopefully enough to prevent starvation.

But if the header tank vent is open and the fuel flow is slower than your actual fuel consumption the level in the header tank will slowly decrease and the header tank will fill up with air from the open vent with the speed of actual fuel consumption minus the reduced flow from the main tank (the reduced flow will depend on the volume of fuel remaining in the main tank and therefore steadily decrease). Same thing with or without pump. 

As the only thing we want to have going to the engine is fuel we should only have fuel inlets open, if you want air open the air inlets... Leave the back door and one day someone not invited will enter...

And I’m sorry, I have to contradict you again… A column of air in the header tank does not create an airlock - not as long as the fuel level in the header tank is higher than the outlet from the header tank (unless you fly inverted…).

I have a see-through header tank so I know that I cannot get all the air out of the tank by venting it. There is always some left at the top. The tank is in conception almost identical to the old "4 minute tube" but with a 7 liter (1.85 gal) capacity and a level alarm built in. It is also placed in the same "behind the seat" position.

Once I have emptied one wing tank the level in the header tank is lowering and the upper part of the tank filling up with air (yes, I know, not yet the same scenario as the header tank is now my “temporary” main tank vented through the primary wing tank).

This goes on until the alarm wakes my up and I switch to the other wing tank. By now I have exactly your scenario with a header tank with a closed vent and a major air column on top expected to be fed by the main tank…

Ta-Da ! Suspense! Will the fuel start to flow from the wing tank to the header tank or will I crash and burn?

 drip_dripJPG.thumb.JPG.4a854f4a0fb90b269

An airlock is when air (or vapor) is preventing the passage of the liquid – but this will not happen, in this configuration, until the fuel level drop to the level of the outlet (“Fuel to engine” on the drawing). If there is air at the top of the header tank the fuel will just drip through the air and down to its fuel friends further down in the header tank.

 

Question for you... you mention "BOTH finger strainers"... one in the main tank - where is the other? Or are you like me having double outlets on the wing tank?  

 

Share this post


Link to post
Share on other sites

Posted (edited)

Haha!  I was pretty sure I wasn't going to convince you that easily!  ;)

And I am pretty sure that after writing this reply I probably still will not.  But each person reading this will have two sides to this discussion  (which I am sure has been wrestled with before here in a different thread :)).  Each can decide for themselves which theory they prefer to agree with.

Ok, so yep in my example I was referring to a two tank system with a finger strainer in each wing tank.

But my main premise is on pure gravity flow, which is spelled out to be tested (at the firewall) on my kitfox.  It needs a certain rate to be demonstrated all the way to empty wing tanks.

And that is my challenge to your idea.  I think that (with a column of air in the header) you would experience airlock during a gravity flow check:

I am guessing that with a full wing tank you might have reasonable flow, but at some point the force of the air trying to rise will overpower the lower head pressure when the wing tank gets lower.  I say at some point your flow rate will be interrupted by air gurgling up the wing fuel line with fuel remaining in your wing.  At that point you are counting on either a fuel pump to create neg pressure in the header (which is not what it is necessarily designed for),  or adequate positive pressure through the fuel cap (which may or may not always be available).

Here's my point:   Does the closed vent header system work?   Absoulutely!  But IMO it does count on a larger number of conditions being met than the vented system.  I say that things (other than the rare "mostly" clogged finger strainer situation) can go wrong easier with a closed system.

I'll and close up here and let you have the last word since I am beginning to repeat my points and we are in jeaporady of a hijack citation, but I will leave anyone else reading this friendly sparring match with the main question that I believe supports my argument:

If the closed system were better, then do later evolutions of fuel systems use it?  I haven't seen all later versions, but it the ones I have seen are vented.

Edited by Yamma-Fox

Share this post


Link to post
Share on other sites

Posted

I don't recall reading about a Kitfox that has a vented header having these kinds of problems but it is perhaps apples and oranges situation due to the dual wing tanks.  

It is very hard for me to imagine the the little electric pump pulling enough of a vacuum to actually empty the header and not having enough of a vacuum to pull fuel from the tank but of course a smalll vacuum in the header would at some point stop the flow.  Two things that are not known is the amount of vacuum that little pump will pull and the actual vacuum that might have been created by the open fuel tank.  I can see that shutting off the pump would upset a delicate balance and stopping the flow.   Two wing tanks are bound to be a much safer system with either a vented or non vented header with the lost cap or the reversed vent tube. 

I guess I will stick with my vented header until some engineer can show me the science to make me change my mind.

Share this post


Link to post
Share on other sites

Posted

It has been a good discussion.

I just have a last comment in regards to "air lock" as I think it is important to understand this - it can be useful in other situations.

In a system with a main tank feeding into a header tank through a line that is as narrow as those we normally use (i.e. so narrow that the surface tension will prevent air from trickle up through the fuel line) you would get an air lock if the headertank had enough air in it. See, I agree with you!  But this is true ONLY as long as we are not consuming any fuel from the header tank. ...but as we are more concerned about what happens when we need the fule than when we don't.... 

As soon as you let fuel out of the header tank it will be replaced with the same amount of fuel from the main tank as the fuel gone out has to be replaced with "something" and as the only thing that have access to the header tank is fuel from the main tank that is what it will be (sometimes refered to as "nature hates emptiness"). The amount of air will stay the same but is not preventing the passage.

It will be more like an "automatic air valve" that opens when you consume fuel from the header tank and automatically closes when you turn off the engine. It is ttransparent to the flow. You will not know the air is there unless you can see it. 

With a high wing and wing tanks setup, and therefore a gravity feed configuration, you can use both configurations open or closed header vent. In the ideal world (see earlier posts) both will work just fine.
 

But it has been fun - we need another "Avid Flyer Murder Mystery" to resolve!

Share this post


Link to post
Share on other sites

Posted (edited)

I don't recall reading about a Kitfox that has a vented header having these kinds of problems but it is perhaps apples and oranges situation due to the dual wing tanks.  

It is very hard for me to imagine the the little electric pump pulling enough of a vacuum to actually empty the header and not having enough of a vacuum to pull fuel from the tank but of course a smalll vacuum in the header would at some point stop the flow.  Two things that are not known is the amount of vacuum that little pump will pull and the actual vacuum that might have been created by the open fuel tank.  I can see that shutting off the pump would upset a delicate balance and stopping the flow.   Two wing tanks are bound to be a much safer system with either a vented or non vented header with the lost cap or the reversed vent tube. 

I guess I will stick with my vented header until some engineer can show me the science to make me change my mind.

WyPaul,

don't worry, I will not try to convince you...
The Mikuni pumps are indicated for 3-5 psi, (5 psi being 1/3 atmosphere). I don't know the spec for the actual pump used but 5 psi is probably a good guess. Emptying the header tank without pulling fuel from the main tank would mean creating a perfect vaccum if we assume the header tank was close to full from the beginning, with 1/3 atmosphere (and that is on the pushing side..) this is obvioulsy not possible. The header tank would have had to be 2/3 full of air to start with to appear empty by 1/3 atmosphere. 

Gfry kindly tested for us and we know that his fuel pump was stronger than the venturi generated negative pressure in the tank... but we don't know by how much. It doesn't have to be very strong to pull/suck the fuel back through a narrow line with a difference less than 1 meter between lowest an highest point once the pump is turned off.

Would having both wing tanks communicate help? Maybe but not sure, there would be more fule to suck back to the non caped tank but without knowing the actual negative pressure it is impossible to say. It sounds like the flow was reversed very rapidly, this indicating that the negative pressure was more than symbolic... 

 

Edited by FredStork

Share this post


Link to post
Share on other sites

Posted

Good job, and welcome to the club!

Share this post


Link to post
Share on other sites

Posted

So finally got round to some ground runs yesterday (after finding a new fuel cap).  I think the Mikuni pulse pump contributed to the failure.  Turns out at idle the fuel pressure was right on 0.2 bar and at full power was flickering just below 0.2 bar.  With the Facet pump on pressure was up at 0.4 nearly 0.5 bar.  So as soon as I switched off the facet pump and at full power the Mikuni was not enough to deal with fuel venting back out over the wing.  Doesn't explain why switching the facet back on did not restore flow though.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now