High fuel flow when RPM increases above 2000

[briantucker]

We're seeing our fuel flow sensor jump to ~21GPH on an IO-360 when the RPM goes past 2000 - normal flow is around 8-10gph depending on leaning. Any ideas what could cause it? There doesn't appear to be any fuel leaking and we can consistently reproduce this (bringing RPM back to 2000 results in normal flow readings). Engine runs fine otherwise. We have dual Lightspeed ignitions so my (naive) assumption is that it's not related to magnetic interference. Engine has 850 hours on it and this is the first we've seen this issue. Ignore the CHT/EGTs, we were fiddling with the mixture to try and fix the flow reading.

 

[BoydBirchler]

Fuel flow is accomplished, in the sensor, via the use of a little paddle wheel with a light on one side and a photocell on the other side that count passes of the paddles on the wheel to interpolate fuel flow. Two guesses: electrical interference or something is making the wheel spin faster (lose fitting sucking air but not lose enough to leak).

 

[Carl Froehlich]

Where did you mount the fuel flow sender? It is sensitive to location.

Carl

 

[briantucker]

Fuel flow is accomplished, in the sensor, via the use of a little paddle wheel with a light on one side and a photocell on the other side that count passes of the paddles on the wheel to interpolate fuel flow. Two guesses: electrical interference or something is making the wheel spin faster (lose fitting sucking air but not lose enough to leak).

Is there anything you can think of that would increase electrical interference only at high RPM? We have dual electronic ignition so voltages and magnetic fields should stay pretty static from my understanding. Fuel flow at higher RPM causing more air to get in would make sense, but fuel pressure is totally normal. Would you expect pressure to change if air was getting in?

Where did you mount the fuel flow sender? It is sensitive to location.

Carl

I'll have to check when I can get back to the plane but it has worked flawlessly since we've bought it about 200 hours ago and there's nothing in the log book about having replaced it so I'm guessing it's where the original builder installed it. I'm guessing location isn't an issue, but what should we be looking for with regards to location just in case?

 

[Carl Froehlich]

I'll have to check when I can get back to the plane but it has worked flawlessly since we've bought it about 200 hours ago and there's nothing in the log book about having replaced it so I'm guessing it's where the original builder installed it. I'm guessing location isn't an issue, but what should we be looking for with regards to location just in case?

Is it mounted following guidance? Such as:
- Mounted after the mechanical fuel pump
- No 90 degree fittings on the sender?
- The sender not mounted to the engine?
- The sender mounted so fuel flow is at least slightly up hill?

The attached photo is an example of how to meet all the above.
Carl

 

[lr172]

To me, It seems quite likely that the EI is throwing off interference above 2000 RPM. It is the only thing that makes any sense. It does not seem like the type of failure mode seen on the red cube. Something this unique with continuous repeatability makes it quite unlikely to be a red cube issue or anything related to fuel flow. I believe the LS is a capacitive discharge system, so the method of storage and release of energy is quite different than your typical simple coil based system that all the rest of us use. It stores energy in a large capacitor and then rapidly releases it to create a more intense spark. A std coil just stores energy as a magneic field and the collapse of this field creates a slower and more controlled energy release. This introduces an opportunity for strange behavior and possibly strong electrical interference not seen by other systems when something goes wrong. My first step would be to call LS for troubleshooting steps. CDI is not your grandfathers ignition system and should not be treated as such. THey have been known to create strange symptoms in the hot rod community where they are attracted to them like bears to honey.

definately check for grounding issues FWF. Proper grounding is far more critical for CDI systems than traditional systems.

Larry

 

[Carl Froehlich]

SNIP… It does not seem like the type of failure mode seen on the red cube. Something this unique with continuous repeatability makes it quite unlikely to be a red cube issue or anything related to fuel flow.

Larry

But, I had a repeatable (as in every flight) erroneous red cube fuel flow in a buddy’s RV-14A. The cube worked perfectly until the aircraft climbed above 5000’ or so. At that point fuel flow when to 14+ GPH mostly independent of throttle or mixture setting. Replacing the cube had not effect, the problem remained. The cube was mounted following the FWF package instructions - in exactly the worst spot (on the engine in a high vibration and heat area). Remounting the cube following the cube instructions (as I outlined in my previous post) and the problem was fixed.

We could only guess that the original mounting location, combined with fuel line routing, lengths and such exposed the cube to an altitude dependent flow oscillation, causing cube wheel to spin faster than normal for the fuel flow conditions.

So this specific problem may be related to RFI, but I would not assume that as the only possibility. Perhaps more data on conditions that create the issue would help.

Carl

 

[lr172]

But, I had a repeatable (as in every flight) erroneous red cube fuel flow in a buddy’s RV-14A. The cube worked perfectly until the aircraft climbed above 5000’ or so. At that point fuel flow when to 14+ GPH mostly independent of throttle or mixture setting. Replacing the cube had not effect, the problem remained. The cube was mounted following the FWF package instructions - in exactly the worst spot (on the engine in a high vibration and heat area). Remounting the cube following the cube instructions (as I outlined in my previous post) and the problem was fixed.

We could only guess that the original mounting location, combined with fuel line routing, lengths and such exposed the cube to an altitude dependent flow oscillation, causing cube wheel to spin faster than normal for the fuel flow conditions.

So this specific problem may be related to RFI, but I would not assume that as the only possibility. Perhaps more data on conditions that create the issue would help.

Carl

While I can't make an argument that your assessment here is incorrect, the fact that it worked perfectly for the OP for 850 hours before today, leads me to believe that he does not have a flawed fuel system design causing this and that interference is more likely. Your buddy likely had his problem from the day he setup the fuel system (the flawed design repeatedly gave erroneous readings when the variables lined up) and that doesn't seem to be the case here, as it popped up out of the blue after years or reliable indication. To me, that points more to a component failure than poor design.

For the record, I did not definitively state EI interference was the problem. Only that is seemed like, to me, the most likely cause. Further,I have seen CDI systems create weird things in the hot rod world , where the mallory systems are quite common.

Larry

 

[briantucker]

My partner was able to pop the cowl off yesterday and everything looks good. Also did a full power run up where he checked the mags, boost pump, and fuel selector and wasn't able to reproduce the issue on the ground. The only thing he noticed with our mechanic is that the stem on the fuel pressure sender appears to be loose. Is there anything with how those senders work that could somehow cause an erratic fuel flow reading when flying (but not on the ground)? Looks like this is our sender: https://store.vansaircraft.com/ie-411l-transducer-fp-50psi-ie-411l.html.

 

[vic syracuse]

Couple of things.... No, the pressure sender will not affect fuel flow readings.
It's possible with 850 hours that it has failed, especially if it is in a location without any heat protection covering.
If its close to the coils of the Lightspeed, check those coils and make sure one hasn't failed. Remove spark plug leads from the coil and measure resistance across the spark plug terminals. It should be around 8k-9k ohms.

Vic