What's new
Van's Air Force

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

Chasing the issue - low fuel pressure on takeoff

John C

Well Known Member
The low fuel pressure on takeoff appears to be a design issue related to cavitation at the engine driven fuel pump. Controlling the pressure at the engine driven fuel pump appears to be the most direct fix.

The problem is not consistent. At times, the pressure at takeoff does not drop below 4-4.5 psi. At other times, the pressure can drop into the low 2s or even near 1 psi.

Often, I can lower the nose and pull back to 4000 rpm, see a pressure recovery, power up and continue the climb.

The two solutions offered prior were to use a blend of 100LL or just accept the low pressure if there is no engine falter. To me, neither is acceptable. I prefer, and the engine was designed for mo-gas.

Hopefully, the slightly higher rated Facet pump will boost the pressure at the engine driven pump to prevent the pump cavitation, if that is the issue.

The problem may only appear on the RV-12, at least according to two of the Rotax dealers and information from user?s groups or operators of other designs. Thus, it would appear to be a design issue related to the RV-12.

PRESSURE DROP IN SYSTEM:

There are several contributors to the lessening of the pressure at the engine driven pump.
Static pressure with the Facet pump: 3 psi
Pressure with engine and Facet pump running: 5.5 psi
Pull the c/b: 5.2 psi
At higher rpms such as runup: 3.5-4.5 psi, sometimes drops to 3 psi, rarely to 2 psi.
At takeoff, the pressure often drops to 1-2 psi.

Van's, Rotax and others suggested instrumentation or installation issues as the source of the indicated or actual low pressure problem.

1. Bad engine driven fuel pump: changed engine driven fuel pump, same problem. Result, both engine driven fuel pumps are ok.

2. Unusual blockages restricting flow: Using the Facet pump only, the static pressure was 3.0 psi. The pressure transducer is on the right forward side of the firewall, near the top. Flow was 33 gph at the engine driven fuel pump; 20 gph at the fuel pressure transducer; and 1.4 gph at the end of the fuel return line (you should also see 1.5 gph flow on the Skyview.). I did not check the flow to each carburetor as they only need 2.5 gph to each. Result, no unusual blockages.

3. Bad fuel pressure sensor: Static pressure about 3 psi and drops to 1-2 psi on takeoff. A new sensor was added and the results were the same. Dynon pressure was compared to a mechanical gauge on the ground, at takeoff and in flight (see later post for mechanical installation). The mechanical gauge was consistently about 0.1-0.2 psi less than the Dynon sensor. Result: Dynon sensor is accurate to demonstrate that the large pressure drop was real.

The system design also contributes to the pressure loss. It appears that the typical overall system loss is about 1.5 to 2.0 psi attributed to increasing fuel flow when and the rpm is increased from idle to 4000-5000 rpm (see above).

4. Contributing to the overall loss:
Long fuel line runs: 140 inches of 3/8 tube. The Facet pump in the middle confuses the issue for me.
Numerous components/fittings: 4 components and 10 fittings
Six 90-degree bends (plus two 45-degree bends)

5. Column pressure due to column height from tank to engine driven pump: (this is included in the overall system loss)
from full tank: 8 inches or 0.2 psi
from empty tank: 21 inches or 0.5 psi

Takeoff dynamics further contribute to the pressure loss.

6. Increasing column height due to pitching up is added to the overall system loss.
10 degrees pitch at 72 inches: 12.5-inch column increase or 0.3 psi

7. longitudinal acceleration adds to the overall system loss:

0.2 Gs longitudinal acceleration calculated from change in ground speed. With a 90-inch horizontal column, the 0.2 G longitudinal acceleration = 18-inch column increase or 0.45 psi pressure loss during acceleraton and/or initial climb.

The total loss is about 2.5-3.0 psi at takeoff. The portion of pressure loss due to the takeoff appears to be about 0.7-0.8 psi and may be the reason that the problem typically appears during the takeoff and initial climb phase of flight. The pressure drop at takeoff lessens the pressure at the engine driven fuel pump. Perhaps that is the primary trigger for cavitation at the engine driven fuel pump.

The higher rated Facet pump may fully offset the takeoff effects.
 
Testing the fuel pressure gauges

I followed advice from Van's, Lockwood, and Dynon. All were helpful. The following are not in the order of progress.

The fuel pump was replaced. The results were somewhat better for a few flights. The Dynon fuel pressure gauge was replaced and compared to a mechanical gauge. Both gauges were remarkable in agreement.

I inserted the mechanical gauge and new Dynon gauge into the fuel pressure gauge line as shown.

v2p1%20%28Copy%29-XL.png


Then the mechanical gauge was routed through the hinge pin access port. The duct tape fairing was added and flight checked.

v2p2%20%28Copy%29-XL.png
 
Some graphs show the pressure loss due to acceleration and pitch up

The following graphs were used to better define the fuel pressure drop at takeoff. Skyview captured the USER LOG DATA at 4 samples/second. A freeware program, DATPLOT, was used to create the graphs. The .csv files were converted to .xlsx, trimmed, manipulated, and equations added; and were converted back to .csv. DATPLOT requires that a value be in each cell of data to be graphed. That is, put a zero in each empty cell in the column of data to be plotted.

Fuel pressure has a 4-second lag, that has been corrected in four of these graphs.

The data titles may be hard to read.
Blue - fuel pressure; psi
Magenta ? fuel flow; gph
Green ? rpm; right scale

This flight had a pressure drop to 2 psi during the takeoff (4100). A second drop (4440) was introduced during a power up and acceleration while in flight.
v3p1%20%28Copy%29-XL.jpg



The next two graphs:
Blue ? fuel pressure; psi
Magenta ? fuel flow; gph
Green ? column pressure due to longitudinal acceleration; psi - right scale
Red ? column pressure due to pitch component; psi - right scale
Black ? sum of the two column pressures; psi - right scale

The column pressure due to longitudinal acceleration and pitch up is presented below. The longitudinal acceleration was calculated using the change in ground speed. The column pressure due to pitch up simply uses the increase in height of the fuel pump. It appears that both fuel flow demand and acceleration/pitch up are necessary for the large fuel pressure drop (see 4409 and 4660). For example, at 4409, the fuel pressure drop occurs when the fuel flow and acceleration/pitch up (black line) are present. At 4460, only the acceleration is present with no fuel flow demand or loss of fuel pressure.

v3p2%20%28Copy%29-XL.jpg

v3p2exp%20%28Copy%29-XL.jpg


Blue ? fuel pressure; psi
Magenta ? fuel flow; gph

Just an interesting graph. The fuel pressure can be traced from the 2000 rpm group along the higher lines to the lower fuel pressure group at 5000 rpm, then back along the lower lines as the rpm decreases.

v3p3%20%28Copy%29-XL.jpg


The first graph used fuel flow without the 4-second adjustment. The second graph has the 4-seconds correction applied to the fuel flow data. It seems to make more sense. the point is to be careful with the data, I am sure that we don?t know all of the anomalies in the system.

v3p4%20%28Copy%29-XL.jpg

v3p5%20%28Copy%29-XL.jpg
 
Low fuel pressure on takeoff and the Facet fix

Review of new Facet pump Notification 18-07-12

The previous posts may help explain why we are seeing the fuel pressure loss at takeoff and why the upgraded Facet pump could solve the problem.

Change to higher pressure aux fuel pump
--ES40105 vs ES40135
3.0-4.5 psi vs 4.0-7.0 psi
----30 gph vs 32 gph

I have been experiencing fuel pressure drops to low pressure on many takeoffs. Once I switched to the higher-pressure pump, I have not seen the pressure drops to the 1-2 psi range.

CONDITION --- Facet 40105 vs Facet 40135
engine off facet only ---- 3.0 psi vs 4.9 psi
engine idle facet on ----- 5.5 psi vs 5.7
engine idle pull facet cb - 5.2 psi vs 4.8
run up facet on ----------4.0 psi vs 4.9
takeoff facet on --------- 1.0-3.0 psi vs 3.6
cruise facet on ---------- 4.0 psi vs 4.9
Note; typically, the psi for each condition above varies a few tenths from flight to flight and day to day.

The higher-pressure pump appears to keep the pressure above the cavitation pressure at the engine driven pump. Thus, the fuel pressure does not dive into the 1s and 2s at takeoff. One concern was that if the pressure into the engine driven pump was too high, the output pressure would exceed the 8-psi limit. So far, the maximum pressure at engine idle with the new facet pump has not gone above 6.0 psi.

The pump is noisier. The noise does disappear at power up for takeoff. I must have gotten used to it as I no longer notice it when the engine is running during taxi.
 
Wow John you were busy on Christmas Day ciphering fuel delivery problems.

I hear what you're saying but I know of at least six RV-12's (D-180, Skyview, Dynon, and HDX EFIS) that don't exhibit loss of fuel pressure on take-off. I have personally experienced vapor lock on a summer day after restarting a heat-soaked engine. My procedure now is to power the master switch and let the electric pump recirculate cool fuel through the fuel distribution hoses above the warm engine. Before restarting a heat-soaked engine I'll run the electric pump for several minutes. Haven't had a problem since. Also, I run 93E10 exclusively...
 
The airplane first flew in 2016.

It is good to hear that some using mogas are not having a problem.
 
I'm in phase 1 of the first of our two 12's.

During the 2 hour flight which we need to do, I spent some time in the pattern - 15 circuits :(

During the climb on crosswind, the fuel pressure was dropping, but it recovered as I levelled off and throttled back.

I assumed it was just a characteristic of the model and engine :D
 
Safety Question

This is excellent information, thank you for sharing.

Reading all of this history has me asking myself a serious safety question.

In general the intent of an electric fuel pump in this design is to be a backup incase the engine driven mechanical fuel pump fails, noting it adds other side benefits such as helping prevent vapour lock. From what I interprete (possibly incorrectly) above, a failure of the electric fuel pump on takeoff could cause engine loss of power even with the mechanical pump operational due to too low fuel pressure? If this is in fact the case, our electric fuel pumps are no longer a backup but a critical device in which a single point of failure could cause a loss of thrust on takeoff...if this were true (please jump in and correct me), then should we not have 2 electric fuel pumps?

Thanks,
 
Two Facet Pumps?

Scott, I am comfortable with the one electric pump.

It appears that the cavitating pump is still pushing sufficient fuel, even though the fuel pressure is low. The Bing carbs can likely work with very low pressure as long as the fuel flow is 2.5 gph.

While playing with the pressure issue, I inadvertently made a series of 3-4 takeoffs with the c/b pulled. The fuel pressure did not dive. Go figure. I don't remember having a pressure drop with the c/b pulled. Also, others have added a switch for the Facet pump. I suspect others have either left the c/b pulled or left the switch off for takeoff. Wonder what they saw.

The 4th graph shows the fuel flow as very good even though the pressure was in the 2s. I never saw a lessening of fuel flow. If the fuel flow was taking a hit, I would be worried.

Also,the pressure recovers very quickly by simply leveling off a bit and pulling the throttle a bit. For example lower the nose 5 degrees from the climb angle and pull back to 4000 rpm for a second or two.

Jim, on your other post, you experienced vapor lock. Was that on initial start up and taxi or on takeoff. I can see where pumping 1.5 gph cool fuel through the system before start would cool the entire fuel circuit. But once up and running, the system is pumping 4-5 gph keeping the system cool. ((1.5 gph for the restrictor and whatever the engine is using at idle, 3 gph perhaps))
 
Last edited:
I have a switch. The lowest I?ve seen without cavitation on just the mechanical pump is just above 3 psi. With cavitation in hot Phoenix summers fuel pressure has taken a nosedive, but recovered to >3 psi when I turned on the pump.
 
I have a switch. The lowest I?ve seen without cavitation on just the mechanical pump is just above 3 psi. With cavitation in hot Phoenix summers fuel pressure has taken a nosedive, but recovered to >3 psi when I turned on the pump.
 
Jim, on your other post, you experienced vapor lock. Was that on initial start up and taxi or on takeoff. I can see where pumping 1.5 gph cool fuel through the system before start would cool the entire fuel circuit. But once up and running, the system is pumping 4-5 gph keeping the system cool. ((1.5 gph for the restrictor and whatever the engine is using at idle, 3 gph perhaps))

John

Vapor lock happened on the ground after restarting hot engine that sat for about 15 minutes. There are many threads in this forum discussing the merits of the electric fuel pump. Van?s recommends running the electric pump all the time - it?s very good advice?
 
My experience

I also had low pressure issues on take off but only with full fuel. I had just the tank vent. Added the vent in gas cap with the idea that with full fuel take off angle may have fuel blocking the vent. Never had a problem since I drilled the hole in the fuel cap in over 250 hours. Coincidence or solution?
 
John

Vapor lock happened on the ground after restarting hot engine that sat for about 15 minutes. There are many threads in this forum discussing the merits of the electric fuel pump. Van?s recommends running the electric pump all the time - it?s very good advice?

I agree with you.
 
I also had low pressure issues on take off but only with full fuel. I had just the tank vent. Added the vent in gas cap with the idea that with full fuel take off angle may have fuel blocking the vent. Never had a problem since I drilled the hole in the fuel cap in over 250 hours. Coincidence or solution?

I reciently drilled a slightly larger hole in my gas cap after replacing the mechanical pump and sensor (without solution) and it seems to have solved my problem
 
I have just the vent hole in the tank filler cap per Van’s original RV-12 design. My logic is that I run 93E10 exclusively and don’t want the filler neck vent because it would provide too much air exchange in the tank when the airplane is sitting for extended periods.

Unlike Avgas, the octane will degrade in 93E10 if subjected to the atmosphere. Also the alcohol in the fuel is hydroscopic and will suck moisture out of the air.
 
short surges of low fuel pressure

I installed the later-design fuel tank vent line, so that is not an issue. I also added an on/off switch for the electric fuel pump, at the recommendation of instructor in my Rotax iRMT class. For two years and 175 hours I have routinely used the electric pump only for takeoff and landing and have no recollection of fuel pressures low enough to be any concern. For no apparent reason on a flight a few days ago (81 deg F ambient) I had a couple brief instances of fuel pressure low enough to trigger the audio warning, 2 psi, when the electric pump was off. One instance was at cruise, one was taxiing at the end of a 1.5 hour flight. No apparent change in engine operation. Turning on the electric pump made the pressure immediately jump back to around 5 psi.


Today I carefully did three full stop and taxi back trips around the pattern with the electric pump on and the fuel pressure stayed around 5 psi at all times. This included takeoff climb at 5200-5300 RPM. I need to experiment with this some more, and will study JC's data carefully. But for now my guess is that there is a possibility of inconsistent engine-driven pump operation due to placement of the tank unusually far aft, which would explain why the kit does not provide a switch for the electric pump. I plan on leaving mine on all the time from now on, or at least until I get a clearer understanding of what is happening.
 
I also added an on/off switch for the electric fuel pump, at the recommendation of instructor in my Rotax iRMT class.

Did Rotax maintenance instructor say why he recommended shutting off electric fuel pump? Seems to me that he would have a good reason to tell someone to alter a safety feature designed by mfgr. I don't think I would want that liability...
 
Did Rotax maintenance instructor say why he recommended shutting off electric fuel pump? Seems to me that he would have a good reason to tell someone to alter a safety feature designed by mfgr. I don't think I would want that liability...

Jim, I realize you are a strong proponent of running the electric pump at all times, for valid safety reasons. However there are valid reasons for not running the pump at cruise in some airplanes. One of those is to prevent false high fuel flow indications and accompanying alarms when cruising at high altitudes. This was a known issue some years ago with certain airplanes. The cause was never definitively identified, and the solution employed by quite a few of us was simply to shut off the pump in cruise flight. Just adds one item to the before takeoff and landing checklists.
 
One of those is to prevent false high fuel flow indications and accompanying alarms when cruising at high altitudes. This was a known issue some years ago with certain airplanes. The cause was never definitively identified, and the solution employed by quite a few of us was simply to shut off the pump in cruise flight. Just adds one item to the before takeoff and landing checklists.

Why not just raise the upper alarm limit and let the pump run all the time? I'd rather have fuel pressure/fuel flow slightly above spec rather than no fuel flow because mechanical pump failed and electric pump was switched off...
 
But for now my guess is that there is a possibility of inconsistent engine-driven pump operation due to placement of the tank unusually far aft, which would explain why the kit does not provide a switch for the electric pump.

As has been explained here in the forum previously, the reason the electric pump on the RV-12 is operated full time with no switch available is to mitigate vapor lock.

This is the same reason the fuel pump is located at the tank in cars and trucks.

Vapor lock is much more likely when fuel is being pulled through the system. With a pump located near the tank (and operating), it is pushing fuel through the system and holding it at a pressure higher than atmospheric, making vapor lock much less likely.
 
switch recommendation

The instructor's recommendation for an on/off switch was a general observation on Rotax 912 operating requirements. He did not know why Van's chose not to include a switch on the RV-12 but said the decision puzzled him. I have always wondered about it and think I now understand why.
 
As has been explained here in the forum previously, the reason the electric pump on the RV-12 is operated full time with no switch available is to mitigate vapor lock.

This is the same reason the fuel pump is located at the tank in cars and trucks.

Vapor lock is much more likely when fuel is being pulled through the system. With a pump located near the tank (and operating), it is pushing fuel through the system and holding it at a pressure higher than atmospheric, making vapor lock much less likely.

Any ideas on if that additional fuel pressure is a consideration in Carb floats on the Bing Carbs turning into sinkers instead of floats? Does that fuel pressure cease to exist once the fuel enters the float bowl, since at that point, the carb jets are now creating lower pressure due to air stream through the throttle bore of the carb.
 
Last edited:
The carb bowls are continually vented to the atmosphere via the short clear plastic vent tubes. I suspect that the fuel pressure has nothing to do with sinking floats.
 
Any ideas on if that additional fuel pressure is a consideration in Carb floats on the Bing Carbs turning into sinkers instead of floats? Does that fuel pressure cease to exist once the fuel enters the float bowl, since at that point, the carb jets are now creating lower pressure due to air stream through the throttle bore of the carb.

The float bowls are vented to atmosphere with a small clear plastic vent tube. Fuel pump pressure is not "seen" by the floats...

DHeal and I typed at same time :)
 
Why not just raise the upper alarm limit and let the pump run all the time? I'd rather have fuel pressure/fuel flow slightly above spec rather than no fuel flow because mechanical pump failed and electric pump was switched off...

Well, Jim, that?s the beauty of this wonderful thing called experimental aviation. We all get to make the decisions that provide the best tradeoffs in our individual situations. No right or wrong, just choices. I chose the path that preserved the accuracy of the fuel totalizer function in the D-180. Even at my advanced age I can remember to run the pump at low altitudes. Checklists help!

In your case, if running the pump continuously causes no problems, by all means let her run!

Also - a word of caution - your last post could be interpreted as speculation and it would be a shame to get this thread shut down.
 
There are definitely two opposing strategies on whether or not to put a switch on the electric fuel pump. I doubt either camp will convince the other. We should all just agree to disagree and move on.
 
Effect of fuel pressure on float height

Any ideas on if that additional fuel pressure is a consideration in Carb floats on the Bing Carbs turning into sinkers instead of floats? Does that fuel pressure cease to exist once the fuel enters the float bowl, since at that point, the carb jets are now creating lower pressure due to air stream through the throttle bore of the carb.

Fuel pressure at the float needle pushes down on the float arm. The net effect is about 1/3 of the float weight. The depth of the float will increase as the fuel pressure increases.
 
Vapor Lock

I read Kathryn?s Report for continuing safety education. An accident in 2015 in California involved a Czech built LSA with Rotax 912ULS engine. The NTSB said the cause of the accident was attributed to fuel vapor lock which has been discussed many times in this forum.

See? http://www.kathrynsreport.com/2019/02/dova-dv-1-skylark-n919e-accident_2.html

The airplane manufacturer didn?t include a fuel return line with restrictor as Rotax installation requires. The airplane apparently flew successfully for 300TT and then developed vapor lock.

My takeaway on this accident is that this is analogous to shutting off the electric fuel pump in a RV-12. This topic has been discussed very thoroughly and can be searched for in this forum.

For what it?s worth, I?ll restate my opinion that the RV-12 electric fuel pump should run all the time as designed by Vans Aircraft in order to mitigate vapor lock.

See also? http://www.kathrynsreport.com/2018/1...-accident.html
 
Quote from NTSB report.
An examination of the fuel system revealed that when pressurized air was fed through the fuel filter, the air would not escape from the upper "y" fitting. Disassembly of the junction revealed that after the fuel filter there was a metered orifice installed that blocked air from continuing through the system
The NTSB report does not say that vapor lock was the cause of the accident.
 
The airplane first flew in 2016.

It is good to hear that some using mogas are not having a problem.

I own a 2015 RV-12 SLSA with 220 hours. I have been using high octane no ethanol fuel purchased at KAWO except when I couldn't and then used 100LL. I've never had a fuel pressure problem until I had to start using 100LL last month. Yesterday in a climb it went quite low into the yellow zone and I turned around and landed. My questions are: could the switch to 100LL be part of the problem? What is the risk of the engine quitting during a climb with very low fuel pressure?
 
The switch to 100LL is likely unrelated. Chances are that the fuel pressure is not really low. Instead the fuel pressure sensor is inaccurate. Verify the fuel pressure before blaming the fuel or fuel pump.
 
Back
Top