Don't know whether its the command processing time or the actual injector closing time, but I can tell you from benchtop measurements that injectors close way faster than they open. The reason is that the fuel pressure wants to shut the injector, so it slows the opening and speeds the closing.

 

Well from a fueling point of view:

Final IPW = Calculated_Required_IPW + Opening_Latency - Closing_Latency

Which makes perfect sense. The Opening_Latency is the required delay for the injector to open whilst pumping "almost no" fuel. The Closing_Latency is the pulse width while the injector is still pumping fuel whilst closing.

So you would have to assume the "Injector Battery Voltage Latency Compensation" map is the combined Open - Closing pulse widths.

Don't mind me I am just thinking out aloud

ok so how does the actual timing of the pulse width work out? When does fuel get squirted into the intake runners? Obviously during the intake stroke, but when during that stroke?

How long is that phase at say 800, 3000 and 7500rpm? Does it matter if the fuel is squirted in at the beginning of the stroke or at the end?

 

i think that jcsb worked out the code that sets pulse start time. seems reasonable that latency should affect pulse start time, but not in the code that you've seen? i've never looked.