How to calibrate the fuel flow indication in the 3500?

[ao.frog]

Hi.

I Live in Norway and have a 3500 EFIS in my RV-7.
I've noticed that the fuel flow probably is showing too high.
Reason is that when I refuel and fill the tanks completely, the amount of fuel going into the tanks are always LESS that the "fuel used" number on the EFIS.
For example:
"Fuel used" showing 120 litres and the tanks take on 110 litres.
The difference is usually beetween 5 and 15 litres on a 100 litres refuel.

I checked the K-factor and it's set at 180 (it was delivered to me this way)
Can I adjust the K-factor alittle so to get the "fuel used" numbers more accurate?
If so; how much? What does the number "180" mean really?

The fuelgauges are within 1-5 litres usually so the error is in the FF I think.

And like to add that I'm VERY happy that I choose AFS! what a great product and company!
And I've already decided to install a 4500 in my second RV-7...

Thanks

[Trevor Conroy]

What type of fuel flow sender are you using? It is the "red cube" from Electronics International or the "gold cube". If it is the red cube, the K-factor is 680.

[ao.frog]

I have the transducer which came with the 3500: the FT-60 Red Cube.
So you're saying I schould change the K-factor from 180 to 680? That was quite a change...?

Here's my calibration data: (sorry for the fuzzy pic)

http://img194.imageshack.us/img194/6622/ffdata.jpg (http://img194.imageshack.us/i/ffdata.jpg/)

[Trevor Conroy]

180 seems really off. What type of engine do you have? Where is your FT-60 mounted? On an injected engine, it should be mounted between the throttle body and the distribution manifold. Per our manual there is a calibration procedure.

Calibration
"The accuracy of the fuel computer is affected by the value of Counts per .01 gals (K Factor). The Counts per .01 gals (K
Factor) sets the calibration of the instrument to match the flow transducer and the variations in the installation. After
running a tank of fuel use the following formula to adjust the accuracy.
The Counts per .01 gals (K Factor) is adjusted from the Fuel Flow/Computer page in Instrument Calibration.
New Counts per .01 gals = (Old Counts per .01 gals) x (Disp GAL USED/PUMP GALS)"

[ao.frog]

I have an Extreem IO-360 from American Engines.
The FT-60 is mounted beetween the engine driven fuel-pump and the Silverhawk injector. (see pic)
That was the only place it could be mounted with straight lines running in and out from it.
It is not bolted to the engine, but "hanging" on the fuelline and both the fueline and the FT-60 is covered in firesleeve. (after the pic was taken, the edges of the firesleeve was covered with high temp RTV)

Forgive me for a possible stupid question, but is it different K-factors for litres and US-gallons? For example 680 for US-gls and 180 for litres?
I have not changed the K-factor but maybe it changed automaticly from 680 to 180 when I set the 3500 to display litres? (no use for quantity-display in USGLS in Europe. It's litres almost everywhere)

http://img413.imageshack.us/img413/186/210u.jpg (http://img413.imageshack.us/i/210u.jpg/)

[ao.frog]

Hi again.

Just a follow-up to my post above:

I got info from Trevor that the K-factors indeed are different: 680 for US GLS and 180 for litres.

1 US GLS is 3,7854 litres so if I multiply 180 with that, I get 681,372 ......Voila!
So I guess the 3500 by itself changes the K-factor automaticly when I switch beetween litres and US GLS. What a smart little machine!

Now, why dind't I think of that?? yeah; I know... I'm not smart enough..

Anyway; now, with my mind at ease, I can adjust the K-factor and get the FF-readings to be as accurate as the tank-readings.
You gotta love these smart EFIS'es...
__________________

[rshannon]

What K factor value should be used as the initial pre-calibration starting point for the FT-90 Gold Cube transducer in a gravity feed system?

[Trevor Conroy]

Hello the K-Factor should be 338 for the gold cube.

[ao.frog]

Just wanted to post a follow-up on the tread above:

A while after this tread was written, I adjusted the K-factor from 180 (the std setting for litres) to 202.
That was the K-factor I figured out after filling a few tanks to full and calculating the new K-factor as described in the AFS-manual.
With this new K-factor, the fuel-calculations on the AFS-3500 became more accurate.

Since then, I've recorded the amount of fuel needed when I've refueled to full tanks, a thing I rarely do (fly with full tanks I mean)

Last week, I re-calculated the K-factor, and my new one is 196.
This schould make the fuel calculations on the 3500 even more accurate.

For those of you using something other than litres, you can adjust your K-factor the same way, even though the K-factor value is different (It's 680 for US GAL for example)

So... this is just an update if any of you need more accurate fuel calculations than you get with the std K-factor.

NB: you might not need to adjust it. This depends on your installation.

[johnb]

In my low-wing 4500 installation (red cube, EFI) I dialled in the FF a little bit differently. I plumbed a temporary Tee into the pressurized return line and fitted a needle valve. Turned on the pump and adjusted the needle valve to bleed 21.3 oz/min (10 gal/hr). Then adjusted K factor until gauge read 10 gph. Eliminates the variable of "full" tank when determining actual fuel burn. If the tank is a wing tank with a large surface area, minor differences in fuel fill levels over a large area can result in significant fuel volume deltas.

YMMV :)
John

[Rob Hickman]

If the tank is a wing tank with a large surface area, minor differences in fuel fill levels over a large area can result in significant fuel volume deltas.

YMMV :)
John


So how do you know how much fuel you have in the plane? I seriously doubt that it is a significant amount if you fill both wings to the top of the filler.

When you adjust the needle valve to 10 GPH what are you using to measure it?