Grand rapids FLOCAL setting with SDS EFI.

In my EIS 4000, there is a calibration setting called FLOCAL for fuel flow. I am curious what settings others have found to be accurate. I am using the fuel flow setting where the EIS 4000 intercepts the signal to one injector. I am using ross's cylinder mount EFI injectors (40 lb per hour?) And 40psi fuel pressure static. From test runs on the ground, looks like aboit 230 is roughy close. Anybody have a calibrated value?

The EIS grabs a signal to one injector vs. the signal from the FF transducer? That certainly isn't how its done on the EIS on my -8... Is this a holdover from the Subbie installation?

Why not use the FF output of the EM-5 that emulates the signal output of the flowscan/red cube transducer? Seems like that would be a more conventional setup.

As Mike said, If your ECU is so setup for fuel flow output, you can connect your totalizer to pin 2 on the white ECU connector.

For others not familiar with this feature, this eliminates the mechanical flow meters (Floscan or Red Cube) entirely, as well as their drawbacks of cost, weight, plumbing and potential for leaks.

rv6ejguy said:

...For others not familiar with this feature, this eliminates the mechanical flow meters (Floscan or Red Cube) entirely, as well as their drawbacks of cost, weight, plumbing and potential for leaks...

Click to expand...

...and potential blockage.

Hmmm. Does the fuel flow output from the em5 come standard Ross? I thought my ecu wasn't equipped with it. Yes, the old Subaru was set up that way(tapped off one injetor), with a scale factor of 210 or so.

Caleb

Norcalrv7 said:

Hmmm. Does the fuel flow output from the em5 come standard Ross? I thought my ecu wasn't equipped with it. Yes, the old Subaru was set up that way(tapped off one injetor), with a scale factor of 210 or so.

Caleb

Click to expand...

Looks like you did not order it at the time but you have V29 software now after the reflash.

Ross, is FF output standard starting with V29, or is it an option hat has to be purchased?

Toobuilder said:

Ross, is FF output standard starting with V29, or is it an option hat has to be purchased?

Click to expand...

Must be purchased. Took quite a while to write the software and do the bench testing. We feel it's good value compared to the cost and installation time of mechanical flow meters.

I would think it would be tough to get an accurate calibration, depending upon the variability of your flights. The FF sensors work by sending an electrical pulse each time a pre-defined, amount of fuel flows through it. The EMS simply counts the pulses and turns that into a flow rate, based upon the # of pulses per second. The calibration is used to tweak the association between electrical pulses per second to actual GPH displayed/recorded.

An EFI, adjusts fuel flow by changing the amount of time that the injector is open (the amount of time, in microseconds, that voltage is applied to the injector) . If you assume a constant fuel pressure, you can calculate fuel flow based upon open time. However, it is not constant. During the initial time that the injector is opening, the flow is different than when it is fully open. The bigger issue is that the EIS is using the number of electrical signals per second to create a fuel flow. However, in the case of an EFI, each electrical signal from the injector is not representative of a static fuel flow. You have to measure the length of the electrical signal to capture fuel flow. I am sure this is why it took Ross some time to write the code for it. It is not a simple activity.

Larry

lr172 said:

I would think it would be tough to get an accurate calibration, depending upon the variability of your flights. The FF sensors work by sending an electrical pulse each time a pre-defined, amount of fuel flows through it. The EMS simply counts the pulses and turns that into a flow rate, based upon the # of pulses per second. The calibration is used to tweak the association between electrical pulses per second to actual GPH displayed/recorded.

An EFI, adjusts fuel flow by changing the amount of time that the injector is open (the amount of time, in microseconds, that voltage is applied to the injector) . If you assume a constant fuel pressure, you can calculate fuel flow based upon open time. However, it is not constant. During the initial time that the injector is opening, the flow is different than when it is fully open. The bigger issue is that the EIS is using the number of electrical signals per second to create a fuel flow. However, in the case of an EFI, each electrical signal from the injector is not representative of a static fuel flow. You have to measure the length of the electrical signal to capture fuel flow. I am sure this is why it took Ross some time to write the code for it. It is not a simple activity.

Larry

Click to expand...

Fuel pressure is held at a constant differential above manifold pressure which is what really counts. We look at pulse width and frequency minus dead time (time it takes the injector to open) vs. voltage which also affects the time to open, in order to generate our own pulse train to emulate the signal of a traditional flow meter.

Bench testing on an ECU running an actual injector flowing fuel into a graduated cylinder allowed us to tweak the dead time and flow calibration number to close agreement to actual flow.

We eventually varied rpm and pulse width over a 5 minute test to confirm accuracy. We got this to repeat within about 1.2% time after time. We don't measure fuel temperature which affects density and viscosity to some degree but I suspect mechanical flow meters also have some error due to these effects.

Dave Anders found the feature pretty accurate in his flight testing to date once properly dialed in and he removed his mechanical flow meters.

The dead time and flow calibration are both user programmable in a setup mode through the panel mount programmer.

We also correct the pulse train for the different trim amounts on each injector.

So yes, bit of thinking, some math, some software and testing to validate and arrive at calibration numbers. Well worth the trouble though IMO to be able to remove the flow meters and just hook up a single wire.

Switched over to the SDS pin 2 Fuel flow output today. Going to take a few flights to get it dialed in, but looks like a value of about 1400 should be close with the 46lb injectors

i've had a few great days for testing and this is fun. you can dial the AFR you want at idle, cruise, and WOT full power. as for the FF accuracy. it's very good in my 4. i'm still dialing it in for the 46 lb injectors. however it's still so close to the gauges in each tank no one would what to run down that low. right now i am within about 0.5 gals or so of my computer agreeing with the gauges when i'm taking around 25 gallons. i can get it closer but have been doing other things. the SDS FF is great, don't waste your time or money on the cube or flowscan.

Confirmed after a few flights:
a setting of 1400 Is within 1/10th of a gallon for me after a full tank.

Caleb

Norcalrv7 said:

Confirmed after a few flights:
a setting of 1400 Is within 1/10th of a gallon for me after a full tank.

Caleb

Click to expand...

My software guy will be happy with that! Thanks for the confirmation Caleb.

Ill come back with some updates after some more hours on it. Curious to see if the variance in fuel pressure regulator will effect it.

Caleb

No More Red Cubes

I'm updating this older thread after seeing a number of folks posting about issues and failures with their Red Cube fuel flow transducers. If you're thinking of going with SDS EFI, you'll be able to rid yourself of these outmoded devices, their associated plumbing, weight, expense, restriction etc. and go digital to digital with a single wire from the ECU to the engine monitor.