Fuel flow problem

[JR]

Hi,
Just installed D180 with fuel flow sensor
Problem is it ranges from 50lph in climb out back to 25lph in cruise then sits around 33lph even up to 40 lph for long periods - 60-80 min. This variation is all at the same rpm and similar load.
After 4 hour flight actual fuel used is dead on 25 lph, whats going on?
I put in K factor, Unit is installed in engine compartment in straight length of pipe, vibration maybe but should be OK.
Variation is my concern, we can always adjust factor to bring it into line otherwise.
JR

 

[dynonsupport]

This is tough to troubleshoot - the easiest thing to do might be to try a different installation location for the sensor and see if that changes the behavior at all.

 

[JR]

Um, in a Jabiru theres not many other places to put it. The entire fuel line is just a few feet long and after taking into account fuel pump and firewall fitting theres not another straight length.

The bit that fools me is that the erroneous readings are very stable, ie it will sit on 35 lph with maybe 0.5lph variation for hours, then after a change in operation and back to the same rpm settings will sit on 40lph with little variation, next cycle will sit on 28lph

Whatever data the screen is getting is being averaged well as when you alter rpm it changes fast.
What are the chances of a bad transducer? What output comes out of these transducers? How about induced noise in wiring?
JR

 

[dynonsupport]

When you're getting the different fuel flow readings - are you at different mixture settings?

Also, are you using the 12V output from the engine monitor as the fuel flow sensor's power supply?

So the sensor outputs signal pulses in direct proportion to the speed of the impeller that spins inside of it. When they fail, it's usually in an "all or nothing" way. If the impeller were getting stuck or was being slowed down, you wouldn't expect it to be able to have stability over even short periods of time.

We're not able to warrant sensors that haven't clearly failed - especially when there's such a big installation-dependent component in getting an accurate fuel flow reading. In cases like this, we're willing to meet you half way though. If you replace the sensor at your cost, and it fixes the reading, we'll refund the replacement cost.

 

[PilotKris]

What seams weird to me is that you said the total fuel used is dead on.

That tells me that the K-factor is set correctly and system is functioning perfectly.

PilotKris

 

[JR]

Total fuel used isnt correct either. So K factor could be an issue.
12v is coming from D180
I will have a play yet to see if I can improve things. I just want to fix it, Ill worry about whos paying later on.
In regards o the mixture settings, Jabirus carbs are automatic, and the flow variation I mentioned is all at same rpm and altitude.
Additional to this the real refill fuel matches historical usage exactly - 25lph
I deal with many kinds water meters and understand how they work (and fail) This seems like a pick up problem, getting double pulses or may be noise in the cables bit its a bit too stable for noise
Im still investigating
JR

 

[JR]

Just checked install and have ruled out some simple stuff
Unit is installed upright, in fuel line with rise after unit, wiring doesnt run with any suspect others, just CHT/EGT wiring. It is close (4") to Mag #2 though. Hose is zip tied to engine support brace, but no where else to supprt it.
Near perfect up and downstream pipe lengths
Next Ill play with K factor but the one Im using is from the cal tag it shipped with so shouldnt be 20% out!!! and this doesnt account for variation in readings.

Any more hints?
JR

 

[PilotKris]

My understanding is that the Trueflow sensor MUST be installed horizontally (or as close to horizontal as possible).

4" from the mag shouldn't cause any problems as long as you're using shielded wire (grounded only at the D180)

PilotKris

 

[JR]

The wires arent shielded so this could be an issue.
I read the Floscan site and it just says to ensure wires are upwards to free vapour efects. And it actually is very close to horizontal, near the top (but not quite) of a large radius bend.
JR

 

[PilotKris]

That's your problem.

The transducer needs to be installed with the fuel line running horizontally and oriented so that the wires exit at the top of the transducer.

It sounds like you've got the fuel line running vertically.


PilotKris

 

[JR]

Kris hi,
No its horizontal, I had another look and the hose had twisted the unit maybe 20 deg to the left, so wires werent exiting vertically, I reset position and now seems to read OK.
Summary is these things are very sensitive to vapour locks and MUST have wires leaving perfectly vertically
Could have been an air bubble trapped from commissioning which has now been released
Thx for the help

 

[jaba-who]

JR.

Your description of the problem is EXACTLY the same as mine in my Jabiru 430.  So whatever is the cause is common to jabirus and perhaps our setup.  
The only thing I have noticed to add is that the odd high flow rates are a bit worse with the electric fuel pump on.  eg at same RPM I might have say 33 litres per minute (when I know it should be about 22 litres a minute) and then if I turn off the electric fuel pump  it drops down to some number closer to the expected 22 lpm after a few minutes with the pump off.

I have kept record of fuel use ( full tank to full tank) and it definitely is far less than what the Dynon is making out. So it is definitely not using what the Dynon says it is.

I have disconnected the fuel line from the mechanical fuel pump and run the fuel into a calibrated cylinder both under gravity alone and also with the fuel pump on and have adjusted my K factor such that on repeat tests I have accuracy to within the 500 mls/hr that the dynon seems capable of so I don't think its a dodgy sensor as such. There is something about the addition of having the engine running as well, that stuffs it up.

Kris.
I understand your comments about the sensor being horizontal but I think the manual says (or maybe it was the distributer here in Oz) it should be in a section of fuel line that is rising (not horizontal) I'll check up when I get home. But it also says it should have a section of straight fuel line both before and after,  which is pretty hard to get in a Jabiru. I had put it down to that and have thought I will get around to extending the fuel line at some stage.

The thing that has made this not a "must fix now" for me is that the Jab has an endurance of about six hours and I never fly much more than a couple of hours at a time so I have plenty of spare.

I will try an horizontalize my transducer and double check the vertical state of the wires and see if I can fix mine too.

 

[jaba-who]

I repositioned my sensor yesterday. It was not truly horizontal - had a bit of uphill angle (in direction of fuel flow) although it was positioned so the wires came out in a vertical plane but with a slight rearward lean.

I installed longer fuel hose before it so I could position the sensor absolutely horizontal with the wires vertical.
Trouble is there is not enough room to fulfill all the other requirements of - six inches of straight line in front and after AND to have rising fuel line after the sensor.

At this stage I have about 2 - 3 inches straight in front and maybe 2 inches straight after with downward lefthand bend after.

Took it for a run for two hours. Dynon said I used 59 liters. Actual use 54 litres. So it is overreading by 2.5 litres an hour.

So the question is.
Is this a time to just re-K factor it until it gets closer or is it a time to spend some serious (and major) time and effort to reroute the fuel line to fulfill all the suggested fuel line factors.

If I have to do a major reroute of the fuel lines but can still only fulfill some of the requirements do you have a known order in which to choose.

ie ?straight sections more important than rising after-section
?rising section more important than straight sections
? is it more important to have the before section straighter than the leaving section etc etc.

 

[dynonsupport]

Well you should adjust the K-factor for sure, and see if it works well across different flight situations and regimes. Also remember that the K-factor is there to be adjusted, and we'd expect that changing the physical installation would require you to adjust it.

As for your questions about which items are most important - we honestly don't know. We have customers that have installations that go against every recommendation and get good results, and then customers like you that pay attention to the details and still get some oddities in the reading. Predicting what factors are causing turbulence in a particular fuel line is just inherently hard.