hi marek , thank you for this , it validates my method
the reason I was experimenting with the 4 sq was to do what you are describing adjusting the dead time until the engine replicated the 2sq running ,and use that to determine the correct dead time .it didn't work becaue the inj are too big for that tiny pulse width on 4sq .
I had already adjusted the gas pressure to give the highest pressure compatible with smooth 1k rpm tickover ,this worked out to be 16 psi .
I am using a tech edge wide band ,this is connected to the msqt and is used for the live closed loop mapping(autotune) and ego running correction .
I am at 4ms at idle now .and its nice and smooth , I have a very tiny tip in accel stumble I want to work on .maybe with the accel map map .at the moment I have only used the accel throttle pos map.
I may try the 1 squ , but im on a bit of a time critical process where I have to have it reliable for a airport pickup on sunday ,as my girlfriend flies in from prague !
if I can get it smooth ,and repeatable and starting well,i still have to sort cranking pw , not really sure which way to go there my instinct is saying something like 1.5 times idle pw , then I will use it and refine as I go .
If you have a wideband lambda sensor hooked up, you can do a test to determine the injector "dead" time:- holding it at a constant medium to high rpm, run it at 2 squirts per cycle and note the lambda reading. Switch to a different numbers of squirts per cycle (e.g. 1) and note whether the lambda reading changes. Adjust your injector deadtime until swapping the number of squirts per cycle makes no difference - this is the point where the deadtime must be correct. Once you have deternmined this for your setup, don't alter it.
If you wish, you can then adjust your "Reqfuel" to then let your fuelling table have a wide spread of values to make tuning adjustments easier. At the end of the day, the two are simply multiplied together anyway, so absolute numbers aren't important.
From your post last October, correct deadtime was 3.3-2.2 = 1.1mS. This also tells you that any pulsewidth below 3.3mS will deliver no fuel and that the minimum amount of fuel delivered is always at least 2.2mS worth. No doubt the resistors and the voltage will slightly alter these times.
You'll want to adjust the gas pressure (or jet size) so that an acceptable idle (i.e. minimum usage) can be achieved with just over 2.2mS of fuel (assuming 1 squirt per cycle) and then look to see whether the highest rpm achievable still allows enough time for the injector to deliver enough fuel at this peak demand level.