I've been reading alot about how Speed Density works. Basically, it uses the ideal gas law to calculate the density of the air going into the cylinder:

density = Pressure / R / Temperature

where:
Pressure is linearly proportional to the voltage from the MAP sensor
R is the ideal gas constant
Temperature is measured by the IAT sensor

Then, since not all the air is trapped in the cylinder, the computer multiplies "density" by VE%, which is programmed into a table. Note that the density equation and R are not accessible or adjustable. The computer then calculates the injector pulse-width required to reach the desired A/F ratio.

Now, assume the MAF is disconnected and the closed loop is disabled. If you were to simply replace the 1 bar MAP sensor with a 2 bar, then you would have half the voltage for any given pressure. Half the voltage means that the density formula will yield half the actual density and the computer will supply half the fuel (also adjusted by VE, so it will actually be less than half). So, to fix this situation, you half the injector constant, which doubles the pulse width, then condense the VE table to the left half.

I understand that there are some other tables that are affected, but could it be this simple to get the motor running reasonably well?

Mod's, sorry for the double post.

 

Subscribing and a bump Mike.

 

Wouldn't this reduce the fidelity of the computer's control over the injectors? By halving the injector constant, it would effectively double the minimum pulse width that the ecm is capable of commanding. Say the computer can normally command pulse widths (these are made up/innacurate) 0.1ms, 0.2ms, and 0.3ms. With the new injector constant it is as if the the computer now commands 0.2ms, 0.4ms, 0.6ms widths.

It seems like this setup would allow you to provide adeqaute fueling for a boost application, but you would have to give up the ability to fine-tune the AFR.

 

I thought that the ECM could command pulse widths down to the 3rd decimal, like 1.234 ms.

 

Wouldn't it be easier to make a 2 bar map work by leaving the injector constant alone? You could then adjust your VE tables and Timing tables for the New Map sensor output voltage. You would lose some resolution but this sure seems alot easier.

 

You would have to half the values in the VE table. In other words, 90% would have to be 45%. To me, it seemed easier to just half the injector constant.

 

The only problem with changing the injector constant is that it is tied to alot of other fueling functions that cannot be changed with Tunercat or Lt1 Edit. For example all the little fueling adjustments that the computer makes in relationship to the TPS and opening/closing of the throttle.

 

Hmmm. . .that could actually solve my very lean tip-in problem.

Mike

 

Ya, you won't have any kind of lean problems if you cut the injector constant in half.LOL

 

the most important aspect about halving the injector constant IS that you can "command" values over 100% on the VE Table... where 50 would equal 100%.

That is really needed.

I will be using the 2 bar probably only for spark calculations.. but i will still use the MAF for fueling... so we'll see how it goes.

 

The Key would be to had a muliplier in hex to the VE result. Or remove the soft limits of 100% to the VE table.

Unfortunately there are too few people working on LT1 hex development.


It would be neat to also include some more complicated equations to help with fuel delievery. For instance, what happens to VE as a result of lambda, or temperature.