[prkaye]

Particularly at lower power settings, I am seeing significant fluctuations in my fuel flow readings, even in a steady cruise configuration. I have a carb and a red cube sensor just upstream of my carb. At around 50-55% power, I will see my fuel flow readings move up and down by up to 1GPH. At higher power settings the magnitude of these fluctuations decreases.
Other readings (MAP, RPM, engine sound) remain steady.
Does this suggest a problem, or is it typical for a carb' ed engine?

 

[Ironflight]

That is pretty normal for the configuration that you describe Phil - especially (and more-so) if th boost pump is on. I generally see a fluctuation about the "steady state" of maybe +/- 0.4 gph, and if I flip the boost pump on, I gain another 1.0 or so. None of it is a real (I believe) flow change, just indication. The good news is that it averages out to the actual flow - unless you leave the boost pump on all flight, in which case, you'll definitely have more fuel left onboard than your totalizer will indicate.

Paul

 

[Bob Martin]

The good news is that it averages out to the actual flow - unless you leave the boost pump on all flight, in which case, you'll definitely have more fuel left onboard than your totalizer will indicate.

Paul

Paul, Thanks for this info. I don't know why this didn't occur to me before.
I installed the red cube and set the K factor accordingly on xcountry flights.
but when flying locally the accuracy would change from .1 to 1gallon off in the safe direction. Now I understand the accuracy is changing with the amount of boost pump use. Which is more on multiple local flights. Thanks for pointing this out in a way that clicked for me!

 

[[email protected]]

double check the connections to the transducer...

The red cube is fed with 5 volts and generates a square wave from the impeller turning with flow... the count of the breaks in the signal is very accurate but the signal is susceptible to some interference if the connections are not very solid and they are subjected to vibration, and can show up as drifting or erratic flow. Double check your connections to and from the transducer.

I chased this for a bit with my setup (fuel injected with the red cube between the fuel meter and flow divider - downstream of both fuel pumps) and only eliminated it (drifting high) after changing the connections at the red cube to inline d-sub pins with heat shrink.

I have no fluctuations with this configuration regardless of the state of the fuel pump.

 

[Ron Lee]

I get the same fluctuations on my carbed engine. Not sure of the flow sensor but I have a Rocky Mountain engine monitor. To get instantaneous fuel flow I have to monitor the fuel flow for a short time (30 seconds or more) and mentally average the values. Good thing is that fuel remaining is very close.

 

[Ironflight]

I get the same fluctuations on my carbed engine. Not sure of the flow sensor but I have a Rocky Mountain engine monitor. To get instantaneous fuel flow I have to monitor the fuel flow for a short time (30 seconds or more) and mentally average the values. Good thing is that fuel remaining is very close.

Yup - I'd say it's about a 30 second period.

 

[Ron Lee]

I don't know what the period is. I see it go high, low, mid range. Not even sure if there is a pattern. I just have to wait for enough different fuel flow values so that my mental averaging is semi-close. It may take a minute.

I only do that if I am recording performance data since the actual value is not that important most of the time. Fuel remaining is the more important value

 

[Ron B.]

With the electric fuel pump on, the fuel flow indicates higher than with the pump off. Is the actual flow the same in both conditions and the reading incorrect or will fuel consumption be higher if the pump was forgotten on for the entire flight?
Ron

 

[Finley Atherton]

I have tried unsuccessfully for over three years to solve the big FF fluctuations I get on my VM1000C (see http://www.vansairforce.com/community/showthread.php?t=32144.) Post #4 gives a graph of the sort of fluctuations I get. I think the basic problem for me is that the VM1000C does not have any sensitivity settings to electronically damp out the fluctuations you get with a carb. Check to see if you can adjust the sensitivity on your FF gauge.
Some time ago Bob Martin installed a FC-!0 Fuel Flow gauge which has sensitivity adjustment for his carb engine and reported reasonably steady readings, so, recently I ordered the FC-10 from ACS and will be installing it as a stand alone gauge and maybe the problem will be solved?

Fin
9A 0-320 (carb)

 

[videobobk]

Like Paul noted, I get a higher FF reading with the aux pump on. I initially set the gauge (AFS 2500) to give me accurate average readings, but that was during the initial 40 hours. These were mostly short flight with higher boost pump usage. I had to reset once I started doing more X-C, or I would have had less fuel remaining than indicated, not a good thing. Now I am pretty much dead-on doing X-C or I have more than indicated. I don't know how to correct this, but I don't see it as a problem.

Bob Kelly

 

[Flybuddy2]

I had the identical issue using the red cube (Skyview). I swapped to a floscan, mainly because I needed a smaller transducer where I had it mounted. After the swap over to the floscan I had no more fluctuations.

 

[Ron Lee]

Bob, it is a function of the K-factor or number of pulses that equal some volume of fuel. It may be temperature related as well.

 

[videobobk]

Bob, it is a function of the K-factor or number of pulses that equal some volume of fuel. It may be temperature related as well.

It is number of pulses, or that is how I understand it. I haven't noticed any differences with temp changes, although there may be some small change. All I know is that I usually have about a gallon more fuel left than the instrument says. I like it that way!

Bob

 

[Norman CYYJ]

With a carbed engine and low power settings you will get fluctuations. As the fuel gets used up in the carb the float goes down and opens the fuel valve (higher fuel flow). When the bowl is full the valve shuts off( lower fuel flow) thus the fluctuations. When running at higher power settings the fuel valve on the float remains open and you will get a more steady state fuel flow.

 

[DanH]

So why does the flow indication fluctuate when measured at the carb inlet?

For that matter, why does flow indication increase with electric fuel pump operation? There's no operating return line to the tank. EGT's don't change so there's no additional fuel flow into the engine. There does not appear to be an actual increase in flow.

 

[Ironflight]

So why does the flow indication fluctuate when measured at the carb inlet?

For that matter, why does flow indication increase with electric fuel pump operation? There's no operating return line to the tank. EGT's don't change so there's no additional fuel flow into the engine. There does not appear to be an actual increase in flow.

My Guess? (I won't even give it the dignity of calling it a theory...) Maybe the pressure pulses from the boost pump are causing the little wheel to spin a little faster with each pulse, effectively causing it to "slip" in the fluid, and giving more counts per minute. As I said - it's a guess.

I am intrigued by the post prior to yours that explains the "wandering" as an effect of the carb float/needle motion - it sounds like an interesting working hypothesis. I must admit though, I still see the fluctuation at cruise power, so I am not sure that the needle is just staying open. But the cyclical nature sure would correspond to float/needle motion.

Paul

 

[Finley Atherton]

I am intrigued by the post prior to yours that explains the "wandering" as an effect of the carb float/needle motion - it sounds like an interesting working hypothesis. I must admit though, I still see the fluctuation at cruise power, so I am not sure that the needle is just staying open. But the cyclical nature sure would correspond to float/needle motion.

Paul

On my 0-320 (carb) the fluctuations appear to be mainly caused by changing G forces. With accurate flying in smooth air my FF readings are reasonably stable. Any turbulence will increase the fluctuations with the readings fluctuating wildly in strong turbulence. (my system does not have any electronic dampening filter/setting)

I can simulate this in smooth air with abrupt stick movements which immediately cause the FF to fluctuate.

At first I assumed that the G forces were changing FF by causing the float to move however I don't think it is this simple as I have observed that positive G reduces the FF and negative G increases the FF which is the opposite to what you would expect if the float was forced lower by positive G and visa-versa.

Fin
9A

 

[Kevin Horton]

I initially had way too much fuel flow indication movement on my RV-8 with fuel injected engine. The fuel flow indication would constantly move up and down over a range of about 1 gph. I didn't like the original location between the electric fuel pump and the engine-driven fuel pump, as it put the sensor on the suction side of the engine-driven pump. I feared it might make the fuel system more prone to vapour-lock issues.

I moved the fuel flow sensor to a new location after the engine-driven fuel pump, where it would always be under positive pressure. Now the fuel flow indication is much move stable, only moving up and down 0.1 gph.

I have no solid theory as to why this new location is so much better than the old one.

 

[DanH]

Kevin, that's interesting. My EI FT-60 (red cube) is between the electric and engine-driven pumps and does not fluctuate more than 0.1, if at all. It does rise in GPH with the electric pump operating. The indication remains steady, just higher than the true fuel flow.

Same airframe, and both are angle valve Lycs. The difference would have to be in details of plumbing. Have any pictures taken before moving the cube? And did you also get an increase in indicated flow when you turned on the boost pump? Said boost pump is the common AFP package?

 

[Kevin Horton]

Dan,



The only photo I could find was from part way through the engine installation process, hence there is not yet fire sleeve on the hose, and the wires are dangling loose. You can see the FloScan sender bolted to the firewall. The fuel comes in from the right side in the photo, and goes out the left side. You will note that there is a 90 degree bend in the hose on the way in, and another one on the way out. It is possible that it didn't like that bend. Or, maybe it was getting vibration from being mounted on the firewall, and the vibrations interfered with its operation. Or, maybe the Red cube is more tolerant of this stuff than the FloScan sender is.

The new location puts the sender on a long straight run. It is now wrapped in fire sleeve, as it isn't too far from one of the exhaust pipes.

 

[DanH]

Kevin,
Well, rats.....you have a Flowscan and I have an EI, so no direct comparison.

FWIW, my red cube is much lower on the firewall. There's an elevation rise between the AFP pump and the point where the line penetrates the firewall with a 90 degree bulkhead fitting, then a level run through the red cube, then another elevation rise to the engine pump. I didn't want anything to trap an air bubble.

I called EI today and asked about an increased GPH indication with electric pump usage. The tech with whom I spoke said they want the red cube installed downstream of all pumps. With an FI installation they prefer it between the servo and the flow divider.

In the case of an installation like mine (FI, with the cube between the pumps)his explanation for increased GPH indication with boost pump operating was something along the lines of "pressure pulses cause the impeller to move backwards and count additional breaks in the light beam". However, he didn't have a clear explanation for the source of these pulses. I would point out that pressure pulses would do nothing. The fluid (fuel) is nearly incompressible, so reverse movement of the impeller would require actual reverse flow.

He had no clue as to why the carb guys would see a fluctuation with the cube mounted between pump and carb. It is their prescribed method.

 

[[email protected]]

impeller is not in the main flow...

Kevin,
I would point out that pressure pulses would do nothing. The fluid (fuel) is nearly incompressible, so reverse movement of the impeller would require actual reverse flow.

As I understand it the impeller is not in the main flow... there is an offset channel or passage on the side of the wiring exiting the red cube that catches a proportional flow with the impeller and signal.