What's new
Van's Air Force

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

Fuel Pump failure on takeoff

WingedFrog

Well Known Member
I mean mechanical fuel pump. Everything was normal regarding fuel pressure during taxiing and run-up. Full throttle and after about 10-20 seconds, just as I was reaching lift-off speed I got a series of audio alarms from the SkyView lady "fuel pressure" confirmed by the gauge going down to the red. Takeoff aborted just on time. I believe that without the audio warning I would have lifted off the runway and lost the engine a few seconds later. Thank you lady and a big thanks to Dynon.
Back to the ramp, I ran a test just in case it was a sensor failure. After taxiing back the fuel pressure was back in the green. I did a fake run-off and after a few seconds the pressure gauge headed down and the alarm came back. This time I kept the 4000 rpm on and about 5-10 seconds later the engine started coughing from fuel starvation.
This was after a one and a half hour flight, preparing to head back to my home airport. By the way my engine came with the new type of mechanical fuel pump and I have 160 hours on the Hobbs. My Rotax specialist told me that the new types also fail but at a lesser rate.
I also had recently got whiffs of gas that I attributed possibly to the sinking floats and I smelled gas when I shut down the engine after arrival.
So far I did not mention the electric fuel pump but all this happened with the electric pump off. I have a switch for the electric pump that I usually activate after run-up but this time I forgot to switch it on and I dare say that it might have been a blessing. If the electric pump had been on it might have hidden the failure of the mechanical one long enough for me to takeoff.
 
Why is the electric pump off?, is it to save runtime on the pump ?.

I mentioned this in a previous thread. I keep taxiing and run-up on mechanical fuel pump only to detect a possible failure of the pump. Just before takeoff I switch the electrical pump and check its impact on fuel pressure. In my opinion it's the best way to make sure that both pumps are working. This current experience will probably nudge me toward having a longer run-up while until now I was just checking the drop in rpms. I will now extend the 4000 rpm phase to exhaust the buffer of pressure in the fuel lines which should take about 5 to 10 seconds, all the while watching the fuel pressure indicator.
 
With the way the RV-12 system is designed, if either fuel pump fails during a time of high fuel usage (high or cruise engine power) the resultant drop in fuel pressure will usually take the fuel pressure into the yellow caution range while the engine continues to run just fine. Then the pilot can deal with the problem once a safe landing has been made.

All engineering decisions are a juggling act of compromises.

The decision of the RV-12 fuel system was to keep the pilot workload as low as possible by having a pump that runs all of the time, and to some degree, attempt to protect them from the inevitable.... making a mistake at an inopportune time.

There are other benefits also (that other owners who have installed a control switch have discovered).
 
With the way the RV-12 system is designed, if either fuel pump fails during a time of high fuel usage (high or cruise engine power) the resultant drop in fuel pressure will usually take the fuel pressure into the yellow caution range while the engine continues to run just fine. Then the pilot can deal with the problem once a safe landing has been made.

All engineering decisions are a juggling act of compromises.

The decision of the RV-12 fuel system was to keep the pilot workload as low as possible by having a pump that runs all of the time, and to some degree, attempt to protect them from the inevitable.... making a mistake at an inopportune time.

There are other benefits also (that other owners who have installed a control switch have discovered).

If the electric fuel pump was in parallel with the mechanical one, I might believe that it allows to keep fuel pressure to stay in the yellow in case of mechanical pump failure. As it is in line with the mechanical pump, depending on the failure mode of the mechanical pump, that might not be possible. In my particular case, when I heard the alarm I immediately switched the electric pump on but I did not have time to check if it did any good as I aborted the flight a few seconds later.
 
Vapor locking

Every time I post engine-related problems, I generally get negative points for "excess display of ignorance", but the symptoms sound much like my problems which I attributed to vapor lock problems:

Our plane has the stock fuel pumping scheme--electric pump always on.

After flights with 20-30 minute intervals between flights, (Young Eagle Flights) we experience power failure either on run up or takeoff.

The fuel pressure drops into the low yellow range (would have to confirm exact values).

Engine runs extremely rough--indicating one carb is starved.

Problem is solved by keeping the electric fuel pump running, and cutoff valve opened, oil check door open for air circulation.

Problem occurs during cool months, when high vapor pressure Mogas is used.
No problem during long, multi-hop cross country flight to Oshkosh using summer (low vapor pressure) Mogas.

I'm fixin to instrument a few thermocouples on the carbs/mechanical fuel pump, etc., to further investigate/characterize this.
 
I have also experienced vapor lock using 93 E10 auto fuel after hot restart. Procedure I now use is to leave master switch on to run electric fuel pump if I know I’m going to restart the engine within say 15 minute time span. Electric pump keeps fresh fuel charge at both carbs to replace fuel that vaporizes with heat. Works like a charm…
 
Last edited:
Rough running is not normal. The engine fuel pump could have a small hole in the diaphragm, not big enough to cause symptoms until the fuel is hot or has high vapor pressure. A hole in the diaphragm will allow air to enter or fuel to exit the fuel pump. Either could cause a drop in pressure which will allow the fuel to vaporize. Others have suspected vapor lock, but the problem turned out to be a bad fuel pump.
 
Rough running is not normal. The engine fuel pump could have a small hole in the diaphragm, not big enough to cause symptoms until the fuel is hot or has high vapor pressure. A hole in the diaphragm will allow air to enter or fuel to exit the fuel pump. Either could cause a drop in pressure which will allow the fuel to vaporize. Others have suspected vapor lock, but the problem turned out to be a bad fuel pump.

Actually both have been the cause rough engine operation.

This is the time of year that vapor lock is most prevalent because the OAT is starting to get warmer but people still have winter blend fuel in their tanks.

Even worse is if people stockpile fuel and happen to refill their storage drum right now and they catch the end of the season winter blend and then run it well into the summer.

Reducing the likelihood of vapor lock is another reason a fuel pump running all the time was incorporated as part of the fuel system design... if all of the portions of the system that gets heated (engine compartment, etc.) is under pressure, there is a lower likelihood of vapor lock (vs only the engine driven pump sucking fuel through most of the system and causing the system pressure to actually be below atmospheric).

Regarding parallel vs series pumps.... I guess it is possible there is a failure mode of the engine driven pump that could stop fuel flow from the aux pump, but operational history has shown that the check valves within the engine driven pump work just fine to allow normal flow. I am not aware of any Rotax 912 stoppage because of the engine driven pump stopping fuel flow. Both pumps installed in series is pretty much a universal standard for 912 installations.
 
In my case I don't believe in the vapor lock for several reasons:
- I know my plane very well and in 160 hours I have never had any sign of vapor lock even in the hottest months of North Carolina
- It happened on a day of mild temperatures on the Atlantic coast
- The engine had been at rest for 3 hours and the taxiing was short
- I did not do anything unusual before takeoff except forgetting to switch the electric pump
- I noticed a smell of gasoline on arrival, probably caused by the bleeding hole of the fuel pump
I have ordered a new fuel pump and will do the exchange early next week. We will do a postmortem on the failed pump.
Question for those having experienced vapor locks: do you see the fuel pressure gauge going down to the red when it happens? I have read mentions of engines running rough but I don't remember mentions of fuel pressure in the red.
 
second pump running

I will never second guess a pilot who aborts a takeoff and does it without harm or damage. I do suggest though, that you run the electric backup pump on takeoff.
I know you mentioned that it would have masked the problem had you had it running. That is why it is there. You should not have a distraction like an engine failure during takeoff. You got lucky....
So many other variables can come into play during that critical phase of flight.
If you begin to use the backup pump as designed.... you will note a loss of power when you turn off the electric at a safe altitude. (with a bad mechanical pump) Your hand is already on the switch. Simply turn it back on and return for landing.
Same for approaches.... I must admit I am tempted sometimes to leave it off when I have the runway "made". What could go wrong? Well, a number of things could create a go around at the last minute. They probably all fall in the category of "wow....I didn't expect that to happen".
None of them need to be complicated by failure of an engine driven pump.
While pumps are simple... they are not made of stainless steel and built like a bank vault. They can and will fail.
Hope this makes some sense and will be of help. Not trying to armchair quarterback... but I thought a post was prudent. Hope it all works out and you have many happy flights.
 
Running Electric Auxillary Fuel Pump

I see no harm in adding a switch to the electric fuel pump circuit in our RV-12s. I have one in mine as well. But, I thought most all GA aircraft with electric fuel pumps required the PIC to run this auxiliary pump on startup, and all takeoff, and landings. Sounds like a sensible procedure to me. At least that was part of my flight training, and it still works for me.

Tom
 
Check the fuel lines

I had the same issue with my plane. The only time there was a problem on take off was when I forgot to turn on the electric fuel pump.

Found the problem with the electric pump on while looking over the fuel lines. Turned out to be a small leak in the fuel line at one of the flares and without the electric pump on, the engine driven pump was sucking in enough air to cause the issue.

Fixed the leaking flare in the fuel tube and all is well.
 
Last edited:
Holy Moley!

I can't believe how many people change the design of the RV-12 fuel system by installing a switch and then don't use a checklist to be sure it is ON for takeoff. This would not look good on an NTSB report. :eek:

Is there any question why Van's did not include this switch in the design of the RV-12?
 
You should not have a distraction like an engine failure during takeoff. You got lucky....

I beg to disagree. First, doing the right thing when you face a problem is not luck it is knowing what you are doing.

I feel lucky not to have gone airborne with a disabled mechanical pump which would have probably occurred if my electric pump had been on. This does not mean I recommend doing it, it was just a mistake on my part.
Learning from this mishap and as I wrote before, I will now extend my run-up by a few seconds while watching the fuel pressure gauge, this run-up is performed with the electric pump off. I will then, with engine back at idle switch the electric pump on, watching again the fuel pressure gauge which should show a little bump. If the increase in pressure is absent, it means the electric fuel pump is not working.

And yes, compared to our S-LSA fellows I am feeling very lucky to be able to check both fuel pumps before takeoff.
 
Last edited:
Why not check electrical fuel pump pressure first when turning on master switch and engine not running? Should read ~ 2 psig. After engine start fuel pressure should increase another ~3 psig and then you know both pumps are working correctly.

I have retained stock wiring scheme with no additional switch to shutoff electric pump. If I want to run electrical system without flying for purpose of programming radios, troubleshooting, etc, I'll pull the fuel pump fuse and always tape it securely to the pilot's control stick so it can't be missed when reverting to flying mode.
 
Why not check electrical fuel pump pressure first when turning on master switch and engine not running? Should read ~ 2 psig. After engine start fuel pressure should increase another ~3 psig and then you know both pumps are working correctly.

I have retained stock wiring scheme with no additional switch to shutoff electric pump. If I want to run electrical system without flying for purpose of programming radios, troubleshooting, etc, I'll pull the fuel pump fuse and always tape it securely to the pilot's control stick so it can't be missed when reverting to flying mode.

This is a good point and I would expect these steps to be part of the official Checklist by VANs.

I still prefer my method, for the same reasons why we do the run-up just before takeoff.
 
from your commentary, it sounds to me that you don't use checklists. not a good idea.

Come on!
I am using a modified checklist adapted to the changes I made and suited to my own taste. For example I do not lock the canopy before I start the engine, just let it crack open. I also switch the electric fuel pump off after landing, etc... most of the original VANs check list is included.
 
I can't believe how many people change the design of the RV-12 fuel system by installing a switch and then don't use a checklist to be sure it is ON for takeoff. This would not look good on an NTSB report. :eek:

Is there any question why Van's did not include this switch in the design of the RV-12?

I agree that Van's should not install a fuel pump switch in their design. Their Job is to make the design as idiot proof as possible. That being said, I have a fuel pump switch. My reasons are that when working on the plane with the master switch on, the fuel pump is always on. If I pull the fuse, it also cuts off the avionics cooling fans, which I do not want. In the advent of an alternator failure, I like to be able to shed every thing except my comm radio. If the facet fuel pump fails, I want to be able to shut it off. Yes, and I do have a checkist
 
Is there any question why Van's did not include this switch in the design of the RV-12?

Well... they did, actually. In the design, but not in the finished product. If you look at the wiring diagram for the older D180 setup, there is a switch position for the fuel pump. At some point before production the switch was deleted, and the pump run directly to the fuse. I verified that there is an empty switch position on my PCB for the pump. It's set up for the same lighted switch used for the ignitions. All you'd need to do, I believe, is cut the hole in the face plate, solder in the switch and move the fuel pump wire from one connector pin to another. In fact that's what I had planned to do.

So why am I not doing it? I suspect for the same reason Van's deleted the switch in the first place. I can't think of a compelling reason to do have it there. I see no problem leaving the pump running in flight, since Van's decided to make it that way and hundreds of RV-12s are flying just fine that way. I do want to be able to shut the thing off while doing maintenance on the ground with the master on -- there's a fuse, and the builder also installed a breaker next to the fuel valve.

It's a cheap automotive fuel pump, not a $900 Weldon certified part. If it quits I'll replace it. It's one less checklist item to worry about. I don't really mind the extra breaker being there -- it would be pretty much impossible to accidentally turn it off -- but it's just not worth the hassle of installing the switch for no real advantage that I can see.

And from what I have seen and read, I trust that mechanical fuel pump about as far as I can throw the whole plane.
 
It is.

First step on the engine start page of the check list says
"Fuel Pressure - NORMAL"

In my case, the fuel pressure was NORMAL when starting first (after a 1:30 flight and a 3 hours stop) and during taxiing. It was normal during run-up and until about 10 sec. after I went full throttle (I was not looking at the fuel gauge during takeoff run until I got the audio alarm so it might have started going down a bit earlier). Taxiing back to the ramp it went back to normal. After stopping the engine and restarting it to make further troubleshooting fuel pressure was normal even with the electric pump off. In the fake run-up at 4000 rpm pressure looked normal for about 5 seconds then started going down, 2 to 3 seconds later the audio message came and after another 2 to 3 seconds the engine gave the first sign of fuel starvation. All these figures are from memory and might be off a bit but not much. Of course, with the electric fuel pump on fuel pressure would have looked even more normal...

One question, Scott, about the decision Vans made to not provide a switch: are there other LSA manufacturers who made the same choice?
 
In my case, the fuel pressure was NORMAL when starting first (after a 1:30 flight and a 3 hours stop) and during taxiing. It was normal during run-up and until about 10 sec. after I went full throttle (I was not looking at the fuel gauge during takeoff run until I got the audio alarm so it might have started going down a bit earlier). Taxiing back to the ramp it went back to normal. After stopping the engine and restarting it to make further troubleshooting fuel pressure was normal even with the electric pump off. In the fake run-up at 4000 rpm pressure looked normal for about 5 seconds then started going down, 2 to 3 seconds later the audio message came and after another 2 to 3 seconds the engine gave the first sign of fuel starvation. All these figures are from memory and might be off a bit but not much. Of course, with the electric fuel pump on fuel pressure would have looked even more normal...

One question, Scott, about the decision Vans made to not provide a switch: are there other LSA manufacturers who made the same choice?

The NORMAL that step in the check list is checking for is performance of the electric pump prior to cranking the engine for start. Thus evaluating the performance of the electrical pump. After engine start the performance of the engine driven pump is noted by seeing the pressure rise because of both pumps now operating.

I don't know what other low wing LSA's have done (high wing wouldn't typically use a second pump).
 
Last edited:
Every time I post engine-related problems, I generally get negative points for "excess display of ignorance", but the symptoms sound much like my problems which I attributed to vapor lock problems:

Our plane has the stock fuel pumping scheme--electric pump always on.

After flights with 20-30 minute intervals between flights, (Young Eagle Flights) we experience power failure either on run up or takeoff.

The fuel pressure drops into the low yellow range (would have to confirm exact values).

Engine runs extremely rough--indicating one carb is starved.

Problem is solved by keeping the electric fuel pump running, and cutoff valve opened, oil check door open for air circulation.

Problem occurs during cool months, when high vapor pressure Mogas is used.
No problem during long, multi-hop cross country flight to Oshkosh using summer (low vapor pressure) Mogas.

I'm fixin to instrument a few thermocouples on the carbs/mechanical fuel pump, etc., to further investigate/characterize this.

I think this is exactly right. The main purpose of that electric pump is to prevent vapor lock.

That RV-12 engine compartment gets pretty dang hot.

In fact I have gotten into the habit to use 100 LL as I approach the summer months solely to help mitigate this problem of crappy winter gas and high outside ambient temps.

If an engine driven fuel pump outright fails, the engine will fail.

I suppose there is some kind of engine driven pump failure that will still allow an electric pump to flow fuel to the engine at some reduced power setting possibly, but unlikely due to how the pump is designed.

And I love how every systems book says gravity feed of fuel is possible! Not.

Ask me how I know.
 
Last edited:
If an engine driven fuel pump outright fails, the engine will fail.

I suppose there is some kind of engine driven pump failure that will still allow an electric pump to flow fuel to the engine at some reduced power setting possibly, but unlikely due to how the pump is designed.

And I love how every systems book says gravity feed of fuel is possible! Not.

Ask me how I know.

There have already been failures of engine driven pumps on 912 powered airplanes where an aux. pump has aided in providing adequate fuel flow to keep the engine running normally.

All of the engine driven pump failures that I am aware of have not been catastrophic failures. They have caused a degradation in pumping performance that generally gives strong signs of something not being right (If the pilot pays attention).

Gravity works fine for fuel delivery in airplanes.... as long as the system is designed correctly.
 
There have already been failures of engine driven pumps on 912 powered airplanes where an aux. pump has aided in providing adequate fuel flow to keep the engine running normally.

All of the engine driven pump failures that I am aware of have not been catastrophic failures. They have caused a degradation in pumping performance that generally gives strong signs of something not being right (If the pilot pays attention).

Gravity works fine for fuel delivery in airplanes.... as long as the system is designed correctly.

If the engine driven pump's actual failure condition can keep allowing fuel to continue to flow to the engine, I would agree that is possible.

In my two experiences, the engine driven pumps failed on turbofan engines and all 6 electric pumps were providing fuel pressure, but the fuel wasn't able to get to the engine, this being a different design than the Rotax 912 system, comparisons are hard and one can watch the early 912 fuel pump mx videos that were put out on EAA by Rainbow for more info on that design and the problems that can arise.

http://eaavideo.org/video.aspx?v=4457527542001

In the event of a catastrophic engine driven pump failure on any engine that won't allow the fuel through to the engine, I think you are in big trouble.
 
Last edited:
I should have been more specific when I said you were lucky. I meant when the problem presented itself, you were in a good position to abort. I know planning is what we all do. But, failures won't go according to plan. I have crashed with an onboard fire and crashed with an engine failure. I always have a plan... but I still credited part of my survival to good fortune.
Anyway, I am sure you are doing your best to pick operating procedures to insure safe flight. Have a great weekend. nc
 
Tony,

The 12 is a great plane, but that doesn't mean it's perfect. I have had an electric fuel pump switch for four years without issue. I firmly believe in its value for revealing a mechanical pump failure.

Rich
 
Back in 2011/2012 era, there was an issue with some RV-12's where there would be high fuel flow indications, with accompanying startling alarms, when the airplane was flown at high density altitudes. At the time there were several conversations with both Van's and Dynon to try and figure out why this was happening. There were theories, but never a real answer. The "fix" was to shut off the electric fuel pump at higher altitudes. And so, many owners elected to install a fuel pump switch.

Now as to the safety aspects, thousands of pilots have flown Pipers and other low wings over many many years, using their checklists to remind them to turn their pumps on during takeoffs and landings. What is so different about these airplanes?

I would also make the point that with the electric pump switched off during preflight runup checks, you can be assured that the mechanical pump is working at full capacity. I dont believe you can be positive about that if the elctric pump is running all the time.

Bottom line, arguments can be made for switches or no switches. If your airplane is certified experimental you can make the choice that best fits your type of flying, and provides the most safety in your opinion.

Oh yes, SLSA owners, sorry about that, I guess you dont have a choice, unless you recertify as experimental.
 
I think this is exactly right. The main purpose of that electric pump is to prevent vapor lock.

That RV-12 engine compartment gets pretty dang hot./QUOTE]

I was curious just how hot the engine compartment gets. I was most interested in the temperature of the voltage regulator / rectifier and the ignition coils. In the case of my alternative engine installation, my engine compartment runs quite cool. I purchased the temperature strips off of Amazon.

IMG_0539.jpg
[/URL][/IMG]

IMG_0538.jpg
[/URL][/IMG]
 
Post Mortem on Fuel Pump

The A&P at the airport where my plane was stranded replaced the fuel pump. I opened the failed pump, here is the picture:

view


Comments:

- The membrane of the pump (right side) is intact, no leak visible
- On the left side there are two valves: one prevents the fuel to back down in the fuel line coming in and one prevents the fuel going out to back down inside the pump.
- A malfunction of either valve would cause a loss of fuel pressure
- I could not find an easy way to disassemble these valves from their housing but I don't think it would show anything abnormal because the fuel pump is operating with normal pressure at low and high fuel flow and fails only when warm (i.e. at run-up temp and at high fuel flow). If anyone can give me a method to access these valves I will inspect them.

Conclusions:
- I believe that one of these valves failed but I admit I could be wrong and that a possibility is that a vapor lock on the inlet valve could produce the same symptom as the valve failure. I have explained earlier in this thread why I believe that the odds that I faced a vapor lock are slim.
- The A&P ran a test before I decided to swap the pump. He could not reproduce the drop in fuel pressure with the electric pump off. According to my local airport A&P this is not surprising as he has had Rotax 912 customers with erratic fuel pressure drops that could not be reproduced on the ground with a cold engine.
- It would be interesting to keep this thread alive with additional cases of low fuel alerts and documenting what actions you took.
 
Last edited:
Skyview Blackbox Data from Savyyanalysis

I extracted the Skyview data for the period of takeoff that shows the fuel pressure drop. Please refer to the SavyyAnalisis chart below:

https://drive.google.com/open?id=0BzXL9s0_bVi9bzU5SlMwaEVoSVU

The top chart shows RPM in red and Fuel Pressure in blue
The bottom chart shows Ground speed in blue and OAT in Yellow
The scale for fuel pressure has been overwritten but is goes from 0 to 6 PSI
This charts shows:
- From 7:30 to 8:00 the runup
- From 8:30 to 9:30 the takeoff run followed by the abort
- From 10:30 to 11:30 the test back in the ramp area

Note that the fuel pressure dipped further down during the test because I let the engine run until sign of starvation.
Not shown: CHTR and CHTL were around 95 deg.C

From this, can we conclude that it is the manifestation of a pump failure or of a vapor lock?
My personal conclusion is that the combination of Skyview data and SavyyAnalysis charts is a fantastic troubleshooting tool. I have since my first flight downloaded every flight data and invite my fellow RV-12 pilots to do so; it just takes a few minutes after each flight and can bring great rewards.
 
Back
Top