BSFC Calculation

[Briggs09hp]

I am wondering if anyone can tell me how exactly to calculate Specific Fuel Consumption.

I know BSFC = Fuel rate/Power.

I know what our power is from the dyno.

What do i need to calculate fuel rate in grams per second? I know the fuel injector pulse widths from the ECU program. I know what the fuel pressure is.

What else do i need? This seems like it should be simple but something just isn't clicking.

Do I need the injector flow rates? How can you measure flow rates? Wouldn't an injector flow differently at different fuel pressures? I am using the stock fuel injectors from a 2001 CBR600 F4i.

I am trying to fill out a spread sheet about our engine. For example, i need to know the Specific fuel consumption at 7500 RPM.

Thanks,

Matt

[Mikey Antonakakis]

I'm pretty sure off the top of my head that the F4i injectors are spec'd at 230cc/min, although it's probably best to measure them to be sure. You also want to test at the same fuel pressure you'll be using on the car. Also, the injector open and close events mess with that PW*flowrate calculation, because they don't just take a step from 0cc/min to 230cc/min. So that will make it a little tricky to calculate.

If you know a volumetric efficiency curve for your setup, you can get a theoretical number from that too. Use your target AFR for that RPM and calculate the flow rate for whatever your VE is at that RPM. Something like air density*VE*RPM/2*displacement/AFR (I really didn't think too hard for this, wasn't trying to get a perfect equation so if I left something out I'm sorry, but I think you get the picture).

[Pete Marsh]

Put your dyno fuel tank on some acurate scales, run the engine at the speed and load in question for whatever period of time gives a reasonable result for you scales. 1 min should do it.

Pete

[Faterooski]

What we use at work here is a similar to what Pete is describing. You are running the engine at a steady-state condition where you know the exact RPM and torque output and measuring everything as you burn through a set amount of fuel (typically 1 lb, which doesn't take too long at 350 HP). Since nothing is 100% steady state, take an average of your HP output to get a more accurate reading. Here at work, a .010 drop on BSFC is a huge deal. For an FSAE application and your dyno setup, you probably won't be able to achieve (and won't need) .XXX accuracy, but should be able to get useful numbers. What you think are small tweaks in fueling and timing will make pretty big changes in your BSFC if you can measure it. Make sure when comparing tunes that you keep the # of variables as low as possible (ambient air temp, coolant temp, exhaust backpressure, fuel temp, etc.) or your comparisons won't mean much.

I know Innovate Motorsports' data aq software has a rough BSFC calculation given injector ratings, fuel pressure, and injector dwell time. Never tried it out though. I wouldn't think making your own channel in your data aq system and calibrating it enough to get useful numbers would be too big of an undertaking.

[VFR750R]

Do what ever you have to do to get fuel flow in lbs/hr and then divide by uncorrected HP (real HP, not corrected to some other operating conditions).
The most accurate way is with a scale as the others have said. I wouldn't trust doing math to calculate it as there will be errors in the fuel pressure, error in the pulse width, error in the manifold pressure, etc. A scale will measure lbs/time directly.
You could also get a fuel flow meter, which are fairly accurate, but if there is a fuel return line from your fuel rail, you'd have to subtract that flow making this a less attractive option.

[Kirk Feldkamp]

There are sensors for fuel flow rate available. The dyno we use has one, you just plumb it inline (making sure to account for the fuel pressure regulator dump flow). I think they're kinda pricey, but they're out there! You can then just log the output and combine it with the data that's coming off your ECU and dyno.

-Kirk