VAF Forums

VAF Forums (https://vansairforce.net/community/index.php)
-   RV General Discussion/News (https://vansairforce.net/community/forumdisplay.php?f=2)
-   -   The ever changing fuel flow (https://vansairforce.net/community/showthread.php?t=182569)

bjdecker 05-17-2020 02:31 PM

The ever changing fuel flow
 
Some background -- My RV-7 has a YIO-360-A1B6 with Dual Plasma III ignitions, Garmin G3X Touch system, with the EI FT-60 "Red Cube" fuel flow transducer located in the fuel line between the engine driven pump and the fuel servo inlet, almost the same location as per RV-14 plans (page 49-05).

Over the past 115+ hours, I have never been able to get a stable or accurate fuel burned total from flight to flight. I'm always adjusting the K factor (mostly up) by a few % after each fill up.

For example, I started with 68000 as the magic #, flew for a couple of hours with various power settings, filled up the tanks - Pump at airport said I burned 15 gallons, G3X said I burned 13.5, adjust the magic number. Wash-Rinse-Repeat, Pump says I burned 21 gallons, G3X said I burned 22.6, and so on/so forth.

Here's a picture of my installation. Any suggestions would be appreciated.

http://www.distractedpilot.com/pics/IMG_0445.jpeg

KayS 05-17-2020 02:45 PM

The manual says that the device should not be mounted directly to the Engine to avoid vibration. Maybe that's the root of your problem. Any reasonable way to attach it to the firewall?

Carl Froehlich 05-17-2020 02:48 PM

Yep - fixed the exact same problem on an RV-14A. The fuel flow would be about right, then above 7K? or so it would go to ~14gph regardless of mixture setting.

As you point out, the RV-14A had the red cube mounted similar to yours (and the way Van?s says to mount it).

We moved the red cube off the engine and all is well. Remember the install instruction with the red cube specifically says not not mount it on the engine. Doing some digging, I found this specific RV was not unique, others had the same problem. Note - very few had the issue and many RV are flying just fine the way you have yours.

For the RV-8 build I mounted the red cube on the engine mount with Adel clamps between the mechanical fuel pump and the throttle body. No issues.

Carl

bjdecker 05-17-2020 03:28 PM

Thanks y'all,

I can move the cube off the engine and put it on the firewall or on the engine mount frame.

What's odd, is that this is the same location as my last RV-7 (same engine, etc.) and I never encountered this issue in 500 hours of flying. Started at 68K and left it there for the duration. Fuel totalized to within .1 gallon per fill up.

Must be a vibration thing - just the right frequency/amplitude, etc.


B

Mark Dickens 05-17-2020 04:01 PM

Of course, this is a religious issue so take my input however you want. I have my red cube mounted vertically on top of the engine immediately preceding the fuel distribution spider (ie after the servo). The reason this is important (and again, this has been beat to death over and over again) is that the servo determines the fuel actually provided to the engine. I have mine mounted with all steel fittings and after I calibrated it, it's been very, very accurate in predicting fuel quantity remaining. I check it at every fill up and it's always within a few tenths one way or the other of the actual amount pumped.

Gash 05-17-2020 06:26 PM

Same problem here
 
I had the same problem with very erratic fuel flow indications in my RV-14A, especially at higher altitudes where I like to fly in the mountain west. I finally got fed up with it and moved the red cube off the Vans RV-14 standard location on the engine. It's now mounted by adel clamp on the engine mount. Luckily I had enough fuel line slack to be able to move it there without needing to buy different length fuel hoses. The results were immediately noticeable. My fuel flow indications stabilized, and so far (about 20 hours since I made the change) everything appears to be working normally.

PilotjohnS 05-18-2020 03:18 PM

sensor
 
The red cube uses a impeller that passes a sensor. if it is subject to vibration or pulsing fuel loads, the sensor seems to sometime double count the passing of the impeller, from what i understand.
I suspect you would see an error difference whether or not the engine is leaned to the point of roughness.

BillL 05-18-2020 03:34 PM

Mine is not as severe.
 
Interesting problem and solution. I have a 7 and the cube is "soft" mounted at the sump attachment under the M1B spider. Schematically, after the servo and before the spider. I notice over 10k that as leaned the fuel flow wanders with 0.6 GPH above and below a mean, the engine can begin to stumble on the low side, so the change appears real. I have measured fuel temperatures and found the highest rise (for my installation) was between the boost pump exit and the inlet to the mechanical pump. In the tunnel seems to be pickup the heat.

I am clearly approaching the issue as a temperature problem, not a cube vibration, but the 14s seem to have a flow inaccuracy I don't have.

This gives me another avenue of thought, thanks to the OP for launching the conversation.

bjdecker 05-18-2020 06:11 PM

Now with data...
 
...I am really kicking myself for not looking at this sooner and ignoring the data; I was too locked up in the "it used to work this way..." train of thought.

Attached pic is the fuel flow from a recent flight. The RPM is 2690 and the MAP starts at 29.2" and decreases to 24.4". Everything else is held constant, except for the propeller blade angles are changing to hold 2690RPM. The ascent profile is 120Kias, for about 1400FPM.



The steep increase in fuel flow is take-off and the steep decrease is top of climb and leaning out for cruise. All of the little "jaggies" are +/- .1 gph swings.

Too me, this looks like a vibration and/or pulsing, causing the fuel sender to report inaccurate rates. These variations over time add up. It also explains why the error is not consistent and varies with altitude, power profile, decent profile, OAT, DA, etc. etc.

Waiting on hoses...and drinking a beer...

B

Gash 05-18-2020 06:51 PM

Before and After Red Cube Relocation
 
Here's my data from before, and after relocating the red cube. This data is from the same route flown between Arizona and Wyoming on different days. This sample is from the exact same altitude (12,500 MSL) and power settings (WOT and Mixture LOP at 8.0 gph). Note that during this time, the throttle and mixture were not moved. Based on my data, the red cube fuel flow transducer in my installation should not be mounted so close to the engine per Vans RV-14 plans. I don't know if it's a temperature or vibration problem, but the erratic FF readings were solved by simply moving the red cube to the engine mount location (see my photo in post above). It may not be accurate to generalize this conclusion to include all Vans RV-14s. All I can say is that in my own installation, it solved the problem. As the saying goes, "Your mileage may vary."

BEFORE relocation


AFTER relocation (FF is the blue line)

RicoB 05-18-2020 07:53 PM

I have mine mounted to the firewall and I keep chasing the tail on this. gets accurate for couple of flight than it wonders off. running out of resources other than could the unit be faulty ?

Bavafa 05-18-2020 08:15 PM

And in my RV7A, I had it mounted between the electric and mechanical fuel pump which is suppose to be a NO and I had a great and consistent result.

My current RV has it mounted in the standard location for a 14 and also have a great and consistent result. Of course the accuracy is within 0.2, 0.3g but I never fill the tank to the neck to have an exact fill up.


All times are GMT -6. The time now is 04:58 PM.