Anyone have accurate data for Siemens 107961 (630cc | 60lb) + questions about injector data accuracy and it's effect on VE tuning

Engine is RB25DET with Link G4X ECU, fuel pressure is 3bar.

I've been playing around trying to tidy up some areas of my VE table and noticing that trying to use measured/over desired as a correction factor overshoots the correction by a lot - very small changes to the number seem to produce much larger changes than expected to the resulting measured AFR. This has had me questioning the accuracy of my injector data - would I be on the right track here?

I have been struggling to find "official" data for these Siemens injectors - with online reports of dead time at 14v ranging anywhere from .3 to 1.1

The data I've come across that "looks" the most reliable is here https://www.fiveomotorsport.com/107961-siemens-60lbs-fi114961/

A previous owner of my Engine and ECU had configured the dead time numbers about 2x higher than these but with only .5 minimum pw compared to 1.5 shown above.

With the "Original" high dead times and low min pulse width and flow rate 630cc - I'm seeing VE of around 50 in the idle area - and in high RPM at 1.5bar boost it's only in the low 90s and running rich as hell (.65 - .6) so probably should be only in the high 80s, would numbers over 100 not be expected in this area of the map?

With the numbers from the link shown above including flow rate 646cc and min PW 1.5ms - I found I needed a VE of 75 at idle so that doesn't seem right either.

Any tips would be greatly appreciated.

I'm sorry I don't have what you're after.

I'd have them swap to IDs so you're working with high quality injectors, well matched, with full accurate data and can execute the calibration in the optimal way.

Without it you can only fudge the VE table so much because RPM/Load are not dead time or low pulse non linearity. Those things need to be accounted for in their proper place for the whole model to work properly.

Thanks for the reply Mike,

Yes - absolutely, high quality injectors with accurate data would be ideal but I'll have to make do with what I have for now - I've been out playing around doing some road tuning today and finding the dead time numbers from that link appear to lead to more accurate VE numbers that what I had originally as long as there is some load and that where I'm seeing abnormal high spikes in the VE at very light load/idle is where the pulse width being logged is very low (1.25 - to 1.5ms) so probably the zone where the flow rate linearity is dropping off rapidly. I also experimented with adding/changing numbers in the short pulse width table, in the operating zone around idle and was surprised to see how much of a difference it can make.

So I think basically the dead time numbers I have now are at least an improvement and I will just have to live with the abnormal light load VE spikes unless I can obtain/come up with some accurate values for the low pulse width adder table.

So consider logging what the pulse width is at the "spiked VE" in the table, and repeat replacing the "spiked VE" value with a value that follows the trend. This difference should tell you where (and how much) to change the short pulse width adder, and you can just tune that to get the fueling on target.

That's a great idea David, - I've added the pulse width related values to my ECU logging so I can gather some more data in this area.

Today I made some some further crude changes to both my dead time tables and low pulse width adder based on graphs in another data sheet I found online and it seems to be an improvement, after a road tuning session I've managed to dial in the cruise area of the table really nicely now, the numbers seem a bit more sensible and the car is running smoother than ever.

It's been interesting learning experience experimenting with these injector tables and seeing how much of an effect these numbers have on the amount of fuel being delivered - it's much more significant that what I first thought but makes perfect sense particularly at lower pulse widths.

That's true learning -- not just searching and copying -- but actually understanding.

Good job putting in the work, and studying the data.

Congratulations on recognizing and solving your own problem.