injector Timing {B3702}.... when to tune?

[joecar]

Thanks, that's a great site, I added it to our scattered tutorial section: showthread.php?t=2990

Interesting topic, I'm all ears...

[SSpdDmon]

Ok... um... I know i shouldnt But I have to post my results. <<GEEK ALERT GEEK ALERT>>

I went back and retested before and after and came up with the same results. I flly warmed up my car to 80*c. I monitored AFR using a gauge on the dash page... because you can enable the min and max needles. With the stock injector timing settings, copied from the stock 20002 FBody LS1 tune, my AFRs bounced around at idle (like they always have) from anywhere between 13.8 to 15.8, giving me an average right about where it should be. I then used my logic above, and altered the timing to essentially fired the injector earlier. I subtracted.334ms from the injector timing because this equates to 2 degrees on the crank. I reflashed and restarted the car. I let it get to 80*c again and started my monitoring. Now, my AFRs bounce between 14.2 and 14.8. I thought I was seeing things, so I put the setitngs back and tested again... back to the way it was. Readded timing to the offset of .334ms and sure enough the AFR's are back to 14.2-14.8.

Hmm... it occurred to me that that 2 degrees on the crank is 1 degree on the cam, the injectors are fired off the crank sensor, so I should actually use more time... so for the hell of it I am going to try to add .668ms and see what happens. I really dont have an explanation.. but I thought I would share. My cam specs are 242/248 .613"/.617" 112+2. This cam with my 11:1 (or higher) compression ratio makes for a rough idle.

UPDATE: I tried lowering the timing offset by another .334ms and see the same result. Maybe slightly more stability but still definitely different that stock timing. Just to make sure, I put stock timings back again and saw the same bouncier AFRs. It is a good Friday.

Interesting. I love being proven wrong.

I wonder how this would affect my setup. I'm still on the stock cam...but, I'm running Motron 60's.

[Goldfinger911]

Well, nothing is 100% yet... I just enjoy the learning and breaking the code of systems like this.
Regarding you car.. the stock cam should be optimal for the stock injector timing.. however I would assume the Motrons do not react when charged exactly like the stock 26lb'ers. So.. maybe some compensation for that is needed. The question is.. are they faster or slower than stock injectors? Disc injectors are usually faster, but I dont know.

I am working on a way to use the Export Log function of EFILive Logging to look at the stability increases and decreases of MAP and of AFR with different injector timings.

I have a cam question... or two... does anyone know what the stock LSx cam's intake lobe centerline is? Would it be the same for all LS cams (aftemarket and stock)? Where I am going with this is... lets say the stock 2002 LS1 cam has a intake duration of 197 deg. Then I install an aftermarket cam with 242 deg at the same lift. Does that mean that my intake valves are not not just staying open later... but opening sooner by the exact same degree? Like this:

242* - 197* = 45* total increase in duration
then 45* / 2 = 22.5* meaning the intake valve opens 22.5* earlier and closes 22.5* later, assuming the ICL is the same?

Given this huge assumption... it is quite possible that the stock injector timing could be spraying fuel on an intake valve that is not fully closed?

[swingtan]

Center line of the cam lobe will depend on both timing and any ground in advance. A longer duration cam will nearly always result in the valves both opening earlier and closing later. Think of it in terms of overlap and the problems of big camshafts having both the exhaust and inlet open at the same time. FWIW, Hymey and I have spoken a bit about this.

We looked at injector timing vs cam timing and what happens with bigger cam shafts. The theory was that for big cams with lots of overlap, you could delay the injector timing to fire when the exhaust valve closed to minimise fuel being lost out the exhaust at low RPM. Then ramp the advance up as the RPM climbs to give the fuel time to fully vaporise for combustion. The issue was that it will probably effect transient fueling a bit and that would need a lot of work to correct for sharp throttle movements. It certainly needs lots of WB use when playing.

In latter tunes, I think GM have gone for very high injector timing to take advantage of the heat in the head / valve to vaporise the fuel in an effort to increase fuel economy. However that also means that the intake charge warms up a lot while sitting in the intake. I think that dropping the timing may allow for some "chemical cooling" of the charge, allowing that charge to be used before it's taken on too much heat from the head. The risk is though, that some of the heavier components of the fuel may not vapourise and remain in liquid form. So the timing needs to ramp up again as the RPM climbs.

Simon.

[Goldfinger911]

11/23 UPDATE: Well, I have been driving, logging and tuning of the past few weeks. The injector settings in my latest pump gas tune are the ones that have given me the best overall performance of my Lucas 42#'ers. You can check out my tune in my signature below.

I have been experimenting with injector timing, IFR, Pulse Width Adjustment, default and minimum pulse width. These settings given me the more stead AFR's while Auto VE tuning and under decel, my AFR's stay around 14:1. Please note, my car has no MAF and runs a full OLSD tune. Questions, let me know. My IFR is set up for a vac referenced reg at 64psi of fuel pressure.

[redhardsupra]

what's your methodology?

[Goldfinger911]

It is actually a combination of a few methods I used.

I started playing with injector timing because my the advance in my new cam, and the erratic nature of my AFR at idle and at light throttle compared to the old one. I calculated how to alter the injector timing to account for x number of degrees advance in the intake valve opening event. This made a definite improvement in the swing/range of AFR's. Then I took some data from AEM on the Lucas 42# injectors that RC Eng resells and incorporated the values from AEMPro/AEMTuner into my Pulse Offset table. (I used to be a big AEM/2JZGTE guy in a prior life) Lastly, I talked to RCEng regarding the minimum pulse of those injectors and dropped my min and default pulse accordingly. On top of that, a lot of logging (I have a folder with 94 log files) and tweaking with driveability being my #1 concern. I know some people laugh at the word driveability when running a big cam but I think it can be done.

My cam swap went from 230/234 .598/.601 113lsa to an XFI lobed 242/248 .613/.617 112+2 and now I have the new cam idling and low-speed driving like the old cam. (until I get to 4000rpm, then the fun starts.

what's your methodology?

[jfpilla]

I think this question is in line with the thread.
Can someone explain the attached tables and how they differ in affect? My car has an LS7 engine. I've converted whatever tables I could to the LS7 values. I don't understand these and if I should or how I should change values?

[joecar]

Joe,

B3702 is the absolute timing of each injector wrt to some crank position referenced event i.e. BDC or something (since IVO changes with cam)...
I'm not sure what exactly this event is.

B1205 sets the timing based on ECT.

(I'm sure you already know that... <- you, <- me)

What I don't understand is that the 1st table is in time (ms) and the second is in angle (degrees [is it CKP or CMP degrees...?]),
and it both cases wrt what crank position/event...?

I don't know how to tune those... spraying/atomiztion must occur by the time IVO occurs...
might need to plot some graphs correlating crank position and IVO/IVC.

Joe

[jfpilla]

Joe,
We both don't understand the same thing.
Each table references ECT. The LS1 table uses MS and the LS7 table uses, what appears to be, some crank position value?
Trying to back into the thought process, the LS1 values increase as the engine temps go up. Do the injectors open sooner or later when the engine is cold? Which ever it is the LS7 values are highest at colder temps. It would seem to make sense that the values are upside down if the LS7 table is CP. But that doesn't help in translating from the ls7 to the ls1 table.
It's interesting to note that if you turn the LS7 table upside down and put a decimal after the first digit the numbers are close.