fluctuating fuel pressure

[SixPapaCharlie]

My instructor pointed it out to me a couple flights back and I have researched online and the responses are all over the place. So I decided to create a thread to add some more all over the place responses next time someone searches this. First let's get some search terms out of the way.
"Comanche 250", "Lycoming 540", "Lycoming fuel pressure", "I almost died".

So what I am noticing is that I see a significant fuel pressure drop at full throttle / climb out. I have read that I only need .5 PSI in the 250 for full power and I am told "This is normal" but I have been told a lot of things are normal or "Happen to all the guys" and none of it is ever true.

Why would my fuel pressure go down at full throttle and normalize at lower throttle and in cruise.
Fortunately I film a lot and I went back and it appears that this has been the case since I bought it.
"Then why are you only just now asking about it?" Engine is butter smooth and those gauges are 61 years old and bounce all over the place and I am not super in tuned to those things and thus an unsafe pilot.

At idle, the fuel pressure is in the green. I go full throttle to takeoff and it drops very close to the red on the lower end. Engine performs fine. Fuel pressure remains low on climb out and starts to slowly come back to the green and remains in the green through out the flight. It is 100% consistent and repeatable. I can do a run up with the elec. pump on or off and I see the drop at full throttle.

There is no behavior that leads me to believe there are any issues other than that needle.
I don't have any engine monitoring equipment other than the old needles so nothing with any level of precision.

Is this cause for concern? What say Comanche / Lycoming 540 owners?

 

[mark.staggs]

Not what you asked, but I see similar results in a Mooney M20C with a Lycoming O-360. At 1000 RPMs on the ground the fuel pressure is at or near the top of the gauge. On full power climb out it drops to 1-2 PSI until I've gained some altitude and power has been reduced.

I think this is probably normal considering the more fuel the engine is consuming the less pressure the pump can make in the lines--kinda like a waterhose with no sprayer on it has a floppy hose (heheh) vs one with with a spray nozzle that's not letting as much flow through and has a firmer hose.

 

[Jim K]

Why would my fuel pressure go down at full throttle and normalize at lower throttle and in cruise.

You're burning less gas. The fuel pump is capable of a certain maximum flow. The engine at full power is burning almost as much gas as the pump can supply.

When you reduce power for cruise, the engine burns less fuel, the needle valve in the carburetor that allows fuel into the bowl is less open, and that results in higher pressure.

I'd be more comfortable with 3-4psi than .5, but maybe that's a Comanche thing. Is it that low even with your boost pump on?

 

[jsstevens]

 

[Gary Ward]

It is common for the fuel pressure to drop under heavy load consumption. Gas is one of the more difficult fluids to suck and push through lines because it can break apart easy under vacuum conditions(get vapor bubbles in it) and cause cavitation at the pump inlet. The warm weather also adds to this.

At this point I wouldn't worry about it if everything is in order. Is there a blast tube to cool the fuel pump?

 

[eman1200]

that's all I got

 

[SixPapaCharlie]

. Is it that low even with your boost pump on?

Yes.

I don't see anything official but several people on forums stating .5 PSI at full throttle is acceptable range.
I am looking for something more official stating that. Either way. my gauge won't show .5, just red or green and I am guessing .5 is in the red zone.

 

[Jim K]

Yes.

I don't see anything official but several people on forums stating .5 PSI at full throttle is acceptable range.
I am looking for something more official stating that. Either way. my gauge won't show .5, just red or green and I am guessing .5 is in the red zone.

I assume you're running the boost pump during climbout....does shutting it off cause the gauge to drop further? Are you sure your boost pump is working? I think that's a bigger concern than the absolute pressure. If you lose your engine driven pump and your boost pump isn't working, you're a lousy glider.

 

[Chip Sylverne]

First, you have to understand how a carburetor works. There is a bowl from which the jets, the little openings that let fuel into the engine, draw fuel. How much fuel depends on the manifold vacuum, controlled by the butterfly valve connected to the throttle control. As the throttle valve is opened, vacuum increases through the carb and sucks more fuel through the little holes.

Ok, now you u understand that, what controls the amount of fuel in the bowl, you ask? That's controlled by a needle valve in the carb inlet connected to a float. When the fuel level drops, the float drops and opens the valve. When the valve opens, the fuel pressure in the line drops because the fuel is getting past the valve. At wide open throttle, the engine is sucking as much fuel as it can through the little holes, continuously draining the bowl, so the float remains down, the valve remains open, and the pressure in the line stays low, and fuel is flowing like crazy, in your case like 22gph.
As you power back, fuel demand lessens, flow reduces to 14gph, the float rises and the needle closes, reducing the fuel flow in the line but increasing pressure, like putting your thumb over the end of a hose. Think of your thumb as the needle valve. With it off, you have flow, but no pressure. With it on, you have pressure, but less flow.

I understand your concern. Seeing fuel pressure at the bottom of the green at take-off flipped me out a bit too, but it's normal. .5lb is all you need at full flow. A fuel totalizer is money well spent.

 

[hindsight2020]

it's a carbed setup, it's normal.

Still, given the condition of the firewall fuel line fittings from the other thread, might be time to do a lookie loo at the age of the mech fuel pump and see if you're taking "fly on condition" of this well-worn bird to its deadbrained limit.

 

[Chip Sylverne]

Attached is the Lycoming fuel flow spec page. You really ought download the manual.

 

[Dan Thomas]

Are all the fuel screens in the system clean? The fine-mesh filter in the gascolator, and the even finer-mesh screen in the carb's fuel inlet can present resistance to fuel flow if they're contaminated, causing low pressure readings at high fuel flows.

Too many airplanes don't get much attention to those screens, especially the carb inlet screen.

 

[Magman]

All LOW WING Light GA aircraft have a fuel system that SUCKS the fuel from the tank.

A few aircraft with the symptoms you described were later found to have tiny

leaks in the fuel line in the cabin. Not easy to pinpoint as it’s air leaking IN

rather than fuel leaking OUT. There were a couple that would run fine on the

ground but falter at rotation. When the pump is raised above the source fuel

flow and power dropped off . Determining whether this happens on one particular

tank aids diagnoses. Suction Leaks are not easy to identify.

 

[Dan Thomas]

All Light GA aircraft have a fuel system that SUCKS the fuel from the tank.

Well, all except for the high-wingers that use gravity flow, like the Cessna 120/140, 150/152, 170, 172, 175, 180, maybe some old 182s. Like all the ragwing Pipers. Like an awful lot of older designs like Champs and Citabrias and Cubs and Luscombes and Maules and more. All carbed, all gravity feed, no pumps, meaning that there is positive pressure in the lines all the time.

Some higher-powered carbed high-wing airplanes have fuel pumps, like the R182. The fuel flow is high enough that the gravity feed is a bit too weak, and the carb on the R182 sits higher than the bottom-mounted carbs on most others, reducing its head pressure. All fuel-injected airplanes have pumps.

 

[Magman]

Dan, is absolutely right. My intent was to include “low wing” but my finger

had a mind of it’s own. I did edit that post.

Cardinals have an engine driven pump and an electric back -up.

Just in case the “ Law of Gravity “ gets repealed.

 

[Gary Ward]

Are all the fuel screens in the system clean? The fine-mesh filter in the gascolator, and the even finer-mesh screen in the carb's fuel inlet can present resistance to fuel flow if they're contaminated, causing low pressure readings at high fuel flows.

Too many airplanes don't get much attention to those screens, especially the carb inlet screen.

First time I took it out, the fine screen was clean but the outside wasn't indicating to me that it has not been out in a while.

That is white grease, not pipe dope.

 

[bflynn]

I've read about low fuel pressure in the past from people who had JPI fuel flow monitors. Not sure of which type they were talking about, but they confirmed full fuel flow at .5 psi, somewhere in the 26 gph range for an IO-540. You're seeing .5 psi because you're using all the fuel the pump can supply. The fuel is never pushing on the pressure sensor, it's moved on before it can build up any force.

If it bothers you, I recommend the same thing - a JPI so you can verify the flow directly other than by pressure.

 

[tyndall]

I've read about low fuel pressure in the past from people who had JPI fuel flow monitors. Not sure of which type they were talking about, but they confirmed full fuel flow at .5 psi, somewhere in the 26 gph range for an IO-540. You're seeing .5 psi because you're using all the fuel the pump can supply. The fuel is never pushing on the pressure sensor, it's moved on before it can build up any force.

Ask yourself, does that sound right?

We have redundant systems and fail-safes for safety, but the mfg is ok with running the fuel supply at the ragged edge of starvation on climb-out, while using pumps that wear out and screens that get dirty.

 

[tyndall]

First, you have to understand how a carburetor works. There is a bowl from which the jets, the little openings that let fuel into the engine, draw fuel. How much fuel depends on the manifold vacuum, controlled by the butterfly valve connected to the throttle control. As the throttle valve is opened, vacuum increases through the carb and sucks more fuel through the little holes.

Ok, now you u understand that, what controls the amount of fuel in the bowl, you ask? That's controlled by a needle valve in the carb inlet connected to a float. When the fuel level drops, the float drops and opens the valve. When the valve opens, the fuel pressure in the line drops because the fuel is getting past the valve. At wide open throttle, the engine is sucking as much fuel as it can through the little holes, continuously draining the bowl, so the float remains down, the valve remains open, and the pressure in the line stays low, and fuel is flowing like crazy, in your case like 22gph.
As you power back, fuel demand lessens, flow reduces to 14gph, the float rises and the needle closes, reducing the fuel flow in the line but increasing pressure, like putting your thumb over the end of a hose. Think of your thumb as the needle valve. With it off, you have flow, but no pressure. With it on, you have pressure, but less flow.

Your garden hose analogy assumes the flow stays the same. The engine pump flows more in relation to engine rpm. So it's like opening up the tap more as you take your thumb off the hose. If the supply increases enough, there will be pressure. More flow than needed = higher pressure. If you can't build pressure, your supply is undersized. Considering the multitude of bigger engines with needle valves the size of your finger not losing pressure at maximum power, trying to justify minimum allowable pressure because the flow squeaks by, sounds like confirmation bias.

 

[Chip Sylverne]

Your garden hose analogy assumes the flow stays the same. The engine pump flows more in relation to engine rpm. So it's like opening up the tap more as you take your thumb off the hose. If the supply increases enough, there will be pressure. More flow than needed = higher pressure. If you can't build pressure, your supply is undersized. Considering the multitude of bigger engines with needle valves the size of your finger not losing pressure at maximum power, trying to justify minimum allowable pressure because the flow squeaks by, sounds like confirmation bias.

The engine is already at max rpm so the output from the mechanical pump is at it's peak. The electric boost pump is on at take off power/wide open throttle. The tap is wide open, flow is at it maximum. Under those conditions, tell me from where the additional output comes to maintain constant pressure in the supply line.

 

[Racerx]

I think Bernoulli is in play

 

[Dan Thomas]

First time I took it out, the fine screen was clean but the outside wasn't indicating to me that it has not been out in a while.

It's the inside of that screen that catches the crud, and it's hard to see with that angled fitting in it. Holding it up to the light and looking through it sideways can reveal the junk. Getting it clear means cleaning it, inside and out, and blowing some air through it from the outside. When it's really clean and dry outside, do what us old guys go: Stick the screen end in your mouth and blow. The air should flow freely through it. Then dry it again with compressed air; your breath could leave moisture in it.

 

[Racerx]

View attachment 118117

It's the inside of that screen that catches the crud, and it's hard to see with that angled fitting in it. Holding it up to the light and looking through it sideways can reveal the junk. Getting it clear means cleaning it, inside and out, and blowing some air through it from the outside. When it's really clean and dry outside, do what us old guys go: Stick the screen end in your mouth and blow. The air should flow freely through it. Then dry it again with compressed air; your breath could leave moisture in it.

It's not just you old guys. Just yesterday there was a fluid leaking from our older service truck. Taste test confirmed a coolant leak.

 

[bluerooster]

It's a pressure vs volume thing. If there was a gauge to measure fuel flow in gpm, it would show an increase commiserate with the decrease in pressure.

 

[Jim K]

It's not just you old guys. Just yesterday there was a fluid leaking from our older service truck. Taste test confirmed a coolant leak.

Not just the old guys that leak?

You might want to see a doctor about that.

 

[Llewtrah381]

The engine pump flows more in relation to engine rpm

While I agree, it seems like this isn’t linear: the mechanical pump is a diaphragm type, which would seem to mean its efficiency would drop off as RPMs or pulses increase, since there’s some lag in the movement of the diaphragm during back-and-forth movement, which would seem to get worse with increasing frequency. By contrast, demand would seem to be more linear with RPM. At higher RPMs, it would seem that pump pressure increases slower than fuel flow demand. This is all an uneducated guess but it seems like a plausible reason to see decreasing pressure with increasing demand.

I agree with other responses: the real issue is if the flow can meet the demand regardless of the pressure: that’s really the parameter that matters.

I’m gonna guess it was cheaper to measure pressure rather than flow and pressure was used as a proxy for flow, with a very low pressure still indicating the flow rate was sufficient to meet the demand.

 

[Dan Thomas]

It's not just you old guys. Just yesterday there was a fluid leaking from our older service truck. Taste test confirmed a coolant leak.

I'd be careful about tasting any sample from a puddle on the ramp. It might have been left there by a desperate pilot who just landed after a long flight...

 

[Daleandee]

I'd be careful about tasting any sample from a puddle on the ramp. It might have been left there by a desperate pilot who just landed after a long flight...

Or the dog traveling with them ...

 

[Dan Thomas]

While I agree, it seems like this isn’t linear: the mechanical pump is a diaphragm type, which would seem to mean its efficiency would drop off as RPMs or pulses increase, since there’s some lag in the movement of the diaphragm during back-and-forth movement, which would seem to get worse with increasing frequency. By contrast, demand would seem to be more linear with RPM. At higher RPMs, it would seem that pump pressure increases slower than fuel flow demand. This is all an uneducated guess but it seems like a plausible reason to see decreasing pressure with increasing demand.

The O-540's pump runs off a cam that turns at a bit less than camshaft speed. It's slow, and these pumps were adapted from automobile/truck pumps that had to run much faster than they do in the airplane.

I’m gonna guess it was cheaper to measure pressure rather than flow and pressure was used as a proxy for flow, with a very low pressure still indicating the flow rate was sufficient to meet the demand.

Yup. And especially back then when these airplanes were built. The small and light precision flowmeters we use now have sensitive little pickoffs that count the revolutions of a tiny waterwheel affair in the flow sensor. That sort of stuff was rare up until maybe 30 years ago.

 

[Capt. Geoffrey Thorpe]

90W diff oil tastes the worst.

 

[Doug Reid]

When the aircraft was certified, the fuel supply to the engine had to be 150 percent of what was needed at full power. On my Cessna 195, I have an engine driven fuel pump even though it is a high wing airplane. I have been told that if the fuel pump fails in flight, it is supposed to bypass the gravity feed fuel from the tanks and keep the engine running. Cessna 190 has a slightly smaller engine and does not require a fuel pump.

Now, if the screens at the carburetor were clogged, that would restrict the flow and cause the pressure in the line to increase...not decrease. Same as the analogy of putting your thumb over the end of a garden hose. Sounds to me like the fuel pump is doing all it can to supply adequate fuel at fuel power...maybe this is normal ?

 

[Racerx]

I'd be careful about tasting any sample from a puddle on the ramp. It might have been left there by a desperate pilot who just landed after a long flight...

I think that goes back to the old guy thing.

 

[kyleb]

Do you see a corresponding gain in the pressure reading as you descend? That might indicate a pressure sender with a restricted vent. Pressure senders have a reference pressure (basically an open vent to the outside). If that vent is clogged or undersized, you see erroneously low pressure readings on climb and high readings on descent.

 

[kyleb]

The O-540's pump runs off a cam that turns at a bit less than camshaft speed.

Isn't there a pushrod that rides on a specific lobe on the cam and drives the pump? Wouldn't that mean the pump operates at a speed that is 1/1 with the camshaft?

 

[Dan Thomas]

Isn't there a pushrod that rides on a specific lobe on the cam and drives the pump? Wouldn't that mean the pump operates at a speed that is 1/1 with the camshaft?

The pushrod runs on a cam on an idler gear between the crank can cam gears. The Lycs are all that way except for the 720, and that one has a rotary gear-driven pump.

The six-cylinder arrangement:



In cars the pump ran off a dedicated cam on the camshaft.

 

[RyanB]

I'd be careful about tasting any sample from a puddle on the ramp. It might have been left there by a desperate pilot who just landed after a long flight...

Just yesterday there was a fluid leaking from our older service truck. Taste test confirmed a coolant leak.

Wait, seriously?

 

[Dan Thomas]

Now, if the screens at the carburetor were clogged, that would restrict the flow and cause the pressure in the line to increase...not decrease. Same as the analogy of putting your thumb over the end of a garden hose. Sounds to me like the fuel pump is doing all it can to supply adequate fuel at fuel power...maybe this is normal ?

You are right, a clogged carb screen would cause a higher pressure reading. The pressure gauge is connected downstream of the pump, which is downstream of the gascolator, so a clogged gascolator screen could cause a pressure drop.

I must have been caffeine-deficient when I wrote that.

 

[Dan Thomas]

Do you see a corresponding gain in the pressure reading as you descend? That might indicate a pressure sender with a restricted vent. Pressure senders have a reference pressure (basically an open vent to the outside). If that vent is clogged or undersized, you see erroneously low pressure readings on climb and high readings on descent.

This system takes fuel from downstream of the pump and runs it through a tube directly to the gauge. No sender. The designers of old airplanes either didn't have suitable senders available, or they weren't too bothered about ducting pressurized fuel into the cockpit.

Old cars also did that, with the oil pressure. A line to the gauge itself. Even the coolant temperatures were measured using a closed system that contained an oil that expanded when heated and worked a pressure gauge calibrated in degrees. Cessna did that, too, with oil temp, up until around 1980 or so.

 

[kyleb]

This system takes fuel from downstream of the pump and runs it through a tube directly to the gauge.

Gauge or sender, the sensor has a reference pressure (an atmospheric vent) that its comparing against the fuel pressure, looking for the difference, right? So the reference pressure port on the gauge (or sender) *could* be the problem.

 

[kyleb]

The pushrod runs on a cam on an idler gear between the crank can cam gears. The Lycs are all that way except for the 720, and that one has a rotary gear-driven pump.

The six-cylinder arrangement:

Thanks for the diagram. It has been >20 years since I had the accessory case open on a Lycoming. I remembered it differently.