How important is fuel pressure regulator accuracy.

[Lucky Phil]

While many automotive system use a pwm fuel pump with a fuel pressure sensor most motorsport products don't (none come to mind at least); even with a return-less fuel system. Instead the regulator is incorporated inside the tank and only fuel maintained at the correct pressure flows to the rail. The intake pressure sensor is used by the ECU to compensate for differences in pressure between idle and WOT by changing the injector on-time. Been standard on many Yamaha's since 2006. 

 

I have yet to see an OEM motorsport or automotive application that operates in a choked flow condition. Maybe a DI unit (haven't really studied them to be honest), but those are still a few years away for mainstream motorsports usage.

 

Modifying fuel pressure at the regulator is a very common method for supplying additional fuel in aftermarket boosted applications, normally the regulator is just referenced to manifold pressure at a 1:1 ratio.    

 

Nice quick summary of narrow band Phil!

Ha, I get a bit wordy for sure.

Mechanical returnless systems still have an in tank regulator but the most common returnless system ( ones with a MAF sensor) and method of varying fuel pressure is via the ecu varying the fuel pump speed. 

 

Ciao

[68C]

So now it appears the accuracy of the regulator is important the next obvious question is should we check this pressure? I assume a simple 0-5 bar gauge tee'd into the pump output would do. Ah.... what pressure should it be, 3.5bar?

And what tolerance should we accept, + or - 1% ?

[gstallons]

You will need to look at some service info to see what is the specific fuel pressure .

[docc]

The workshop manual for the early Sport with the external regulator specifies 3 +/- 0.2 bar.

 

The external pump has a relief valve set at ~5 bar.

[Camn]

On mine (external regulator) the pressure was 3,2 bar when I measured it.

[GuzziMoto]

While many automotive system use a pwm fuel pump with a fuel pressure sensor most motorsport products don't (none come to mind at least); even with a return-less fuel system. Instead the regulator is incorporated inside the tank and only fuel maintained at the correct pressure flows to the rail. The intake pressure sensor is used by the ECU to compensate for differences in pressure between idle and WOT by changing the injector on-time. Been standard on many Yamaha's since 2006. 

 

I have yet to see an OEM motorsport or automotive application that operates in a choked flow condition. Maybe a DI unit (haven't really studied them to be honest), but those are still a few years away for mainstream motorsports usage.

 

Modifying fuel pressure at the regulator is a very common method for supplying additional fuel in aftermarket boosted applications, normally the regulator is just referenced to manifold pressure at a 1:1 ratio.    

 

Nice quick summary of narrow band Phil!

To go further into the narrow band O2 sensor. It does not measure a static reading, it lacks the accuracy for that. What typically happens is when the motor is running in closed loop mode the ECU is cycling the mixture between rich and lean. It counts the passes between lean and rich, as I recall 1 volt is rich and 0 volts is lean. The cutoff point is right around 0.45 volts, above that (rich) very quickly becomes 1 volt and below that (lean) very quickly becomes 0 volts. A narrow band O2 sensor really lacks any kind of accuracy, especially in the region we want. It goes very quickly from rich to lean, near 1 volt to near 0 volts. The ECU uses this to find that cut off point between rich and lean and cycles the fueling so it is toggling back and forth from rich to lean to rich to lean. It cycles back and forth to make sure it is at the right mixture, the difference between rich / 1 volt and lean / 0 volts is pretty small. So while it is going back and forth between rich and lean it is never far from the correct mixture. And supposedly that cycling back and forth also happens to work better with the catalytic converter. But that last part may be outdated as many auto companies are moving towards wide band sensors for more accurate fueling.

[68C]

If all V11 fuel pressures are in that region I am surprised at the wide tolerance, could be up to 13% difference and still be in tolerance. I would have thought varying the fuel map by that much would be noticeable.

 

That is unless I have screwed up my maths, + or - 0.2 is 0.4, divide by three is 0.133r, which I take to be 13%.

 

So I guess the fuel pressure is not super critical or a narrower tolerance would have been specified.

[Kiwi_Roy]

When I say the accuracy is not that important I mean one or 2 psi is not going the make the difference between a motor that will run or not.

Fuel flow through a given orifice is proportional to the square of the pressure drop (to double the flow you have to square the pressure difference) so one psi makes just a tiny change.

I think injectors run under choked flow so that the manifold pressure doesn't have any effect, I don't know that for certain it's just a theory I came up with.

Early FI systems tubed the regulator reference fitting to the manifold, this must have resulted in a fuel pressure all over the map

[emry]

I won't quote to keep this simple.

 

Phil, I actually meant it. Your description of how a narrow band works was solid.  No sarcasm was implied. 

 

Gstallons, DI = direct injection

 

GuzziMoto, quite right, in addition modern ECU's that still use a narrow band measure also measure the time it take for a "switch" (cross of the 0.45v) to happen. This works better on multi cylinders than twins. If the switch was quick, then the change in injector pulse was close, if it was very slow then the injector pulse was way off and will be corrected by a larger margin on the next cycle.

 

68C, fuel pressure is important but it is an area of diminishing returns. 13% of idle injector pulse is about nothing in the real world, math is important but due to the other variable involved with a running production engine it ends up being the difference between "good" motor and "weak" motor. At WOT that can make a bigger difference. Quality control and accuracy vs. production cost quickly become a fight, cost wins every time.

 

KiwiRoy, systems that use a manifold vacuum referenced pressure regulator are still common, particularity on multi-cylinder models. Large twins intake pressure generally vary to much to be useful and are accounted for digitally for this reason. There are 4 cylinder models that reference 3 cylinders to fuel pressure and then only the 4th only to the ECU, this is used for both D-electronic fueling, finding TDCC, and acceleration compensation. Needless to say the response of modern air pressure sensors is very good. In the scheme of things the response of a fluid system is pretty slow to react to changes to intake pressure, there is lots of mass involved in proportion, so it ends up working pretty well even when on a theoretical level it doesn't.