Deep into SOIT/EOIT, looking for a bit of advice

[Allen Vos]

Been spending a bit of time with SOIT/EOIT and have a few questions.

Sorry if the post is a bit lengthy but lots to cover.

I had someone here help me setup my SOIT\EOIT a couple months back, now wanting to just optimize it a bit in the transition area from the off idle area(1500-2000RPM) to 4000 RPM, if its even necessary as the car is running quite well.

I'm not really experiencing anything negative as far as drivability goes but when making adjustments like advancing the injection timing a bit later the car feels a wee bit lazier going into the max injection advance area.

I've included my tune, screen shots of graphs for EOIT/SOIT timing.

Here are my Cam specs for SOIT/EOIT

282/297 advertised duration (.006")
IVO - 32 degrees BTDC
IVC - 70 degrees ABDC
EVO - 83.5 degrees BBDC
EVC - 33.5 degrees ATDC
Overlap is 65.5 degrees

The boundary angle is set to 470 and the higher RPM is 580, with the transition starting at 2500-4000 Here is the reasoning and calculations behind this

EOIT sets the Boundary at idle to 470 so that no matter what, fuel is not sprayed "in the danger zone" of getting reversed. IVC is 610, BDC is 540, 610-540 is 70. So 540-70=470.

EOIT at higher RPM sets the Boundary angle to occur 30 degrees before the IVC. IVC is 610, so EOIT or the Boundary angle is 580.

Make up mode has been turned off.

I have played with SOIT at idle and have determined that 389 degrees is the "sweet spot" with the richest STFT.
389 degrees is 4.5 degrees before the EVC. This really cleaned up the raw fuel smell as well.

As far as the boundary angle of 580, the car pulls extremely well and hard above 4000 RPM, so I am only assuming the 580 boundary angle is correct.

EOIT-SOIT Tables.jpg

So for right now I am playing with the transition from say (1500-2000RPM) to 4000 RPM and I am just wondering If I am on the right track.

I have not modified the Boundary table at all, as I am only shifting the Normal RPM table right or left to either advance or retard the SOIT transitioning from the idle area to 4000 RPM.

I have to be honest, I have started the transition as low as 1400 RPM and as high as 2900 RPM and the only thing I notice, is if I advance it at a later RPM the engine just feels a bit lazier transitioning into higher RPM.

I fully expected to feel bucking and jerking or hiccupping in the the lower RPMS, but felt nothing.

I have about 10 different settings I tried and flashed to the ECM and have included them in an EXCEL spread sheet, which I have attached here.

I made sure my Normal RPM tables values were changed to have a smooth transition of SOIT values going from low RPM to High RPM.

The SOIT formula is SOIT=SOIT- NORMAL ECT- NORMAL RPM - MAKEUP ECT(if used) - MAKEUP RPM(if used)

I belive the Injector PW value is added to the SOIT after the value is found.

From what I understand you can use any combination of numbers on your boundary table, Norma ECT and Normal RPM table to achieve the actual SOIT value.

I know there are many different opinions on SOIT/EOIT here, but I am not trying to start a thread on opinions about that

I know I was told to watch the STFT values going from low RPM to High RPM, but I could not get any kind of a stable reading to make any kind of call using STFT's as I am only doing street tuning and don't have a DYNO

I have used Cringers EOIT program and plugged all my values into including injector PW and I do understand how his program lays everything out.

EOIT Assistant.jpg

I have also got my hands on the Injector Timing tool by black charts and put all my data in there. Now it seems a bit more complex than Cringers EOIT tool, but I am not sure how to properly read the SOIT and SOIT-FRC values. The SOIT and SOIT-FRC(green Lines) drop to the bottom of the graph from idle to 3500 RPM but then spike to the top of the graph by 4000 RPM. I am not seeing the pattern of this in the EOIT assistant program.
I have included a screen shot of the graph and the excel spread sheet.

Injection Timing Tool Graph.jpg



I do have a basic understanding on how the SOIT/EOIT works with valve timing along with the boundary angles what I am struggling with is the SOIT/EOIT transition from low to high RPM. I am just looking for some guidance on this as I cant really notice much in the way the car behaves during the transition, other than the bit of laziness on the later injection timing when the transition RPM is raised.

I am tempted to just leave it alone and have the transition start at 2000 RPM as it is in my tune, unless any of you guys spot something obviously wrong and are able to explain why its wrong and what I should do to correct it?


INJECTOR TIMING TOOL GEN IV - Allens Vette.xlsx
SOIT-EOIT Work Sheet.xlsx
Current Tune.hpt

[Cringer]

In general, no one is going to agree on the correct or best way to handle SOI/EOI. The OEMs are going to focus on Emissions, Drivability, and Power (in that order). And as for aftermarket tuners, most won't even touch EOIT at all, even if you give them the cam specs. Count yourself lucky if your tuner adjust the Normal ECT table in order to clean up raw fuel smell at idle.

The other fact of life is that there is nothing you are going to be able to log in the scanner to objectively tell you if your SOI/EOI settings are good or bad...well except for fuel trims/wideband...which is your only objective guide. The other guideposts are idle smell, seat of the pants drivability and a dyno (which most of us peasants do not have at our disposal).

I have found a few spreadsheets floating around for this stuff and was underwhelmed by them. This is why I created my program to help with this and make it as powerful as possible. I created a video to arm everyone with the background knowledge needed to understand the HOW and the WHY to tune SOI and EOI. Of course not every "expert" out there will agree with my thoughts on this, and that is fine. And you know what? I am totally happy if they share what they know with me/all of us and we all get better. I am always willing to put away my ego for something better for the sake of learning.

The previous tune I revised for you on that thread a long time ago is exactly how I would set up the initial EOIT stuff if it were my vehicle. And this is a *starting point* for you to advance and retard SOI for the low/mid RPM stuff to find the best fuel trims, and it sounds like you did for idle. Making any improvements here is going to be a slow and arduous process. You are going to have to make small adjustments and analyze the long term fueling trends. I am not talking about LTFT, I am referring to getting familiar with you fuel trims and being able to identify 1% swings in the right or wrong direction. Part of this is being consistent in the way you drive. These are 2D tables that are setup in RPM increments...they are not setup in a 3D format with RPM and MAP (which I would prefer). So in other words, if you use more throttle you are spraying more fuel at a given RPM...so be consistent in how you drive/log/tune. More throttle = longer pulse width = a later EOI during closed loop.

Additionally, if you are finding the car is getting a little lazy, throw a couple of degrees of timing in that area to see what happens. Optimal fueling is going to reduce the chance for knock anyway. And as you will find as you make changes in one area, it is likely to impact multiple other areas as well. MAF/VVE, O2 voltages, rich/lean PI settings, EOIT, and timing are all related.


Just to clear the technical stuff up, the correct formulas are:


1) For normal low RPM/Closed Loop operations with short IPW
1a) If Make Up Mode is enabled (set to Single or Multiple)
SOI = Boundary Angle - Normal ECT - Normal RPM - M/U ECT - M/U RPM

1b) If Make Up Mode is disabled (set to None)
SOI = Boundary Angle - Normal ECT - Normal RPM


2) For high RPM/PE operations with long IPW
You have to understand that SOI is moveable and floating while EOIT (or the Boundary Angle) is not. SOI will be forced to moved when IPW becomes excessive. The definition of excessive is when the engine RPMs are high enough that the window (in milliseconds, not degrees) to spray all the fuel needed is not long enough. So a new SOI has to be calculated.

2a) SOI = Boundary - IPW (expressed as degrees)

My approach is to spray as early as possible (and disabling Make Up mode goes hand in hand here) for closed loop. So long as:
1) you do not short circuiting the fuel
2) do not cause any drivability issues

Other people like spraying late as possible. This is accomplished by setting the Boundary Angle to whatever number you want (520 - 580 or whatever) and then zeroing out the Normal tables. This then forces the ECM to back calculate the SOI every time using the IPW as degrees. I do not like this method personally. Feel free to experiment and report back. This is the same as formula 2a (however, these people also probably have Make Up Mode enabled, so add the default 30*).

So in conclusion:
1) For low RPMs, think of the spray window in degrees and focus on a specific SOI that will be driven off valve events.

2) For high RPMs, you have to ignore SOI since it is floating and you are going to spray on the back of the intake valve and thus some fuel will be short circuited, and there is nothing you can do about it (unless you get bigger injectors and sacrifice idle quality), so focus on EOI which will be best found on a dyno.

The quick formula to convert MS to Crank Degrees:
[RPM] * [0.006] * [INJECTOR PULSE WIDTH IN MS]

The quick formula to convert Crank Degrees to MS:
[CRANK DEGREES] / (RPM * 0.006)

Note we use crank degrees here and not cam degrees since the Boundary Angle and all the EOIT tables are based on 720? for two rotations of the crank for a 4 cycle engine and the cam shaft's valve events are also listed as in relation the crank (TDC and BDC).

I am hoping other folks chime in here as I would like to learn their viewpoints as well.

[GHuggins]

Very well said Cringer.... The only thing I don't know for sure is how the ecm coding handles the rpm setting. Does it continue to use the last known value, interpolate or cut it off? I would think if it's using the last known, then the black spreadsheet is right. If it's cutting it off then your spreadsheet is showing it correct. If interpolating like some things do, then they're both wrong.

All in all when all else fails, always go by trims and afr errors. The richest is always the best whether it be at idle or at rpm, but getting some people to listen to that is a different story. With rpm you have to find a happy median as well to keep soi happy as trims CAN greatly change with light throttle high rpms vs full load. This is why I do the idle/stomp test and then adjust accordingly as things proceed... You can know in 3 test how it needs to be shaped...

[edcmat-l1]

I did quite a bit of testing many years ago (2008ish) before anyone had any spreadsheets or calculators for this. One of my test mules was a GEN3 stroker motor, big cam and big injectors. It had a FAST manifold on it so I took advantage of that and put a low pressure transducer in one of the ports (nitrous boss) so I could "see" the intake valve opening. Using this setup with a 4 channel o-scope and 5 gas analyzer I was able to move the injector timing and change substantially the left over hydrocarbons. Keep in mind these aftermarket lumpy cams can have as much as 8,000+ PPM HCs running at stoich so even if you cut them by 30% it's still gonna stink pretty good but hey, it's a 30% reduction.

I had some scope screen shots but that was several laptops ago and I did not take good care to save a lot of my scope screen shots over the years. Shame on me.

What I will say, you'll find way more quantifiable results the bigger the injector. Smaller stock size injectors you'll see some results in the shape of the VE down low, the change in injector "on time" (before and after) and smell. Smell is going to be your biggest difference and your best measuring stick unless you're using real diag. tools to measure stuff.

Once you get close, the difference of moving your timing one way or the other by a few degrees is negligible. My opinion, if you've already noticed a substantial difference in the smell, you're close enough.

If I get some free time in the near future I'll do some more testing. I have a couple good candidates in the shop right now just need to find a half day to hook up some equipment and run some tests.

[Cringer]

I forgot, there is one other nice indicator of proper EOIT. CLEAN cylinder head intake runners, intake manifold runners, intake manifold plenum, and throttle body. Once I got SOI/EOI sorted out there was no more goo and black nasty carbon hanging around anymore (of course you have to clean the old residue off first to remove it).

[Cringer]

If I get some free time in the near future I'll do some more testing. I have a couple good candidates in the shop right now just need to find a half day to hook up some equipment and run some tests.

That would be great to see the results!

Edit: if you can specifically test to see how and when the Make Up Mode fires, that would be most excellent!

[SiriusC1024]

I've found that EOIT adjustments lead to more accurate airflow models even at the top end. The transition in applied corrections between fuel trims and wideband isn't as abrupt.

[hjtrbo]

I'm budget. I just blindly add 20 or 30 deg on the boundary idle cells for idle smell and walk away. What we really need is bi-directional control of the boundary table.

[Alvin]

and a dyno (which most of us peasants do not have at our disposal).

I moved one +/- 20, and 40 degrees.. So -40, -20, 0, 20, 40 from stock just last week.

The graphs overlaid each other on the dyno. zero difference at all.

[SiriusC1024]

What about IPW and DC?

[GHuggins]

I moved one +/- 20, and 40 degrees.. So -40, -20, 0, 20, 40 from stock just last week.

The graphs overlaid each other on the dyno. zero difference at all.

It's most apparent with boosted applications and doing the pulls starting from idle. I've seen 100 rwtq gains with turbos below 3000 rpms, which you would think would make the tires break loose easier, but because the power and torque come in so smooth vs the abrupt boost hit, it just makes them pull that much better.

I also saw gains at higher altitudes changing it even on stock cams if that's worth anything. The dyno I did most of my research on was at 3400 feet Granted it was mainly from the added ignition timing it'll let you run, so you might want to play with ignition timing a touch while moving it around as well. More retarded seems to allow more timing, to a point of course and you need to ramp it in and not just add it everywhere.

I'd have to go do some digging for the dyno pulls, but it can make a substantial difference...

[GHuggins]

OK, it wasn't 100 hp, it was 100 tq and 50 hp. I never can remember that part for whatever reason. All I did was go from factory injection to retarded 15 degrees as a test. I later dialed it in much better for all of the rpms because as you can see retarding it actually made it lose up top, so I kept it advanced and changed it up slightly for that. Even though it doesn't show it, both pulls were started from idle as that's how I always do dyno pulls on manuals. I did keep this one advanced with rpm until they started spraying nitrous at which point I retarded it for ignition timing safety... This is the one car in my sig. It was a shop car that I experimented on.

Injection Timing Change TT.jpg

[edcmat-l1]

OK, it wasn't 100 hp, it was 100 tq and 50 hp. I never can remember that part for whatever reason. All I did was go from factory injection to retarded 15 degrees as a test. I later dialed it in much better for all of the rpms because as you can see retarding it actually made it lose up top, so I kept it advanced and changed it up slightly for that. Even though it doesn't show it, both pulls were started from idle as that's how I always do dyno pulls on manuals. I did keep this one advanced with rpm until they started spraying nitrous at which point I retarded it for ignition timing safety... This is the one car in my sig. It was a shop car that I experimented on.

Injection Timing Change TT.jpg

First off, I don't know anyone who starts dyno pulls at 1000 rpm. Second, you didn't start both those pulls at the same RPM. The difference could be spool speed and without seeing both data logs there's no way to really determine if the gain was due solely to injector timing. The fact that the lower HP pull was started almost 1500 rpm LATER than the more powerful pulls indicates that a difference in spool speed is most likely the difference. The more powerful pull you had a 1500 RPM head start on the less powerful pull, allowing the turbos to gain momentum so by the time the engine was to the same point in RPM the turbos were well into their spool.

The fact that this is a turbo application, plus the fact that both pulls are not identical, is not a good comparison.

Not to mention there's no AFR. The difference could be in AFR. A richer mixture will cause the turbos to spool easier, quicker, than a leaner mixture will. Retarding the injection timing SHOULD richen it up especially down low.

It's just not a good back to back comparison where you can say "Yep the power difference is soley due to injector timing".

[GHuggins]

Well, I stated in there that all the pulls I do on manuals are from idle. To clarify, idling in 4th gear then stomped to wot. Might be surprised what you can find out and how well your tunes are dialed in doing it that way

Otherwise, found a couple where they were perfectly over one another. Well nearly and as to spoiling, well that's the whole point. Changing injection, changes how the turbo's spoil comes in, thus why the added power and torque down low. Less wasted heat energy....

Injection Timing Change TT 2.jpg

I also have supercharged with cam big hp gains where I only changed injection and minor gains with cams where I only changed injection. I thoroughly researched this back in the day... I would have to do a lot of digging for those and don't have the time for it right off, but even the drop off up top should show injection changing things either way.

[edcmat-l1]

Yeah I'm not into full load on an engine at 1000 rpm. That's just dumb.

And the second log you posted has a 400 rpm start difference. Not to mention the files are 5 runs apart.

Post 3 consecutive runs on one file, then 3 consecutive runs on the change file, dyno graphs and data logs.

Anything FI is much less repeatable than anything NA, especially turbo cars. If there's 1 thing I'm REALLY good at its running a dyno and getting extremely repeatable results. Damn near lab grade results.

The way you strap a car, the way you place it on the dyno, start RPM, coolant temp, air temp, airflow on the dyno, so many things influence the outcome. Most guys, even really experienced guys, ain't doin it right.

[GHuggins]

Yeah, that was over 10 years ago. Don't have any of that. I did do 3 consecutive pulls every time though. Average results and all. Whole reason I had those identical other rpm start point runs to show you.

Just dumb eh

The WHOLE reason I started doing them that way on the dyno is because I "looked dumb" at the drag strip doing it the usual, what I would now call "dumb way". I had a cammed zl1 with pulleys and a 108 tb car that dynoed great doing 2500 to 3000 rpm start pulls, but guess what, nobody does that at the drag strip. In fact show me one drag strip where they'll let you start rolling, get up to rpm then stomp it LOL.... The zl1 once equipped with good hooking tires would stumble and spit and sputter launching at anything less than 2000 rpms. The only way we could take off was to rev it to 2000 or 2500 then dump the clutch.

Wound up being injection timing in the end on that one too, FWIW... It's all about keeping fuel in the chamber with boost or anything as far as that goes.

That twin turbocharged car combined with a sawmill workers research believe it or not, is the whole reason I started investigating this way back when. The sawmill worker - and I wish I could still find his research on the subject - found that by retarding injection timing torque would be increased in the higher rpms allowing his engines to perform the work easier and consume less fuel. However, they wouldn't start or idle worth a damn. I took that combined with the fact that we can change injection with rpm and tried it out on that TT ls3 car. I then further explored it on other types of Boost and NA platforms. The OE changes it for a reason even if it is "to keep fuel in the chamber" for emissions and economy.

Another thing I would like to point out in case you didn't notice it, that car struggled bad with Desoot as this was before that was discovered. Greatly skewing the PE was the only way I could work around it at the time. Changing injection also made that not as noticeable as a nice side affect

As what I consider to be a very great man in the industry once said. The dyno is where you should put it through the paces and know everything before it ever hits the track. Or at least Gale said something almost identical to that.

I do the things I do for a reason and usually because at some point bad things happened to me.

I also really believe in what Cringer has in his signature. What I lived by in the tuning industry from the beginning and why I've always kept digging and digging.

[edcmat-l1]

OK, it wasn't 100 hp, it was 100 tq and 50 hp. I never can remember that part for whatever reason. All I did was go from factory injection to retarded 15 degrees as a test. I later dialed it in much better for all of the rpms because as you can see retarding it actually made it lose up top, so I kept it advanced and changed it up slightly for that. Even though it doesn't show it, both pulls were started from idle as that's how I always do dyno pulls on manuals. I did keep this one advanced with rpm until they started spraying nitrous at which point I retarded it for ignition timing safety... This is the one car in my sig. It was a shop car that I experimented on.

Injection Timing Change TT.jpg

Well, I stated in there that all the pulls I do on manuals are from idle. To clarify, idling in 4th gear then stomped to wot. Might be surprised what you can find out and how well your tunes are dialed in doing it that way

Otherwise, found a couple where they were perfectly over one another. Well nearly and as to spoiling, well that's the whole point. Changing injection, changes how the turbo's spoil comes in, thus why the added power and torque down low. Less wasted heat energy....

Injection Timing Change TT 2.jpg

I also have supercharged with cam big hp gains where I only changed injection and minor gains with cams where I only changed injection. I thoroughly researched this back in the day... I would have to do a lot of digging for those and don't have the time for it right off, but even the drop off up top should show injection changing things either way.

OK, to be clear, you posted 2 dyno graphs with 4 pulls and only 1 out of 4 was a pull from 1000 rpm. The other 3 were from a reasonable 2000ish RPM.

Trying to say someone should check how good their tune is by starting pulls at 1000 rpm at 1:1 is just idiotic. Definitely the dumbest thing I've seen you say on here.

Take the OP for instance. His vehicle is a LS3 C6, manual trans with a stage 2 or 3 cam. The cam kills low end torque. Why would I or anyone else start a pull on this vehicle at 1000 RPM?

Take his vehicle totally stock, strap it down. Put it in 4th gear, start a pull at 1000 RPM. What happens? On a stock calibration it pulls all the timing out of it to keep from detonating the tops of the pistons off. What do we do, try adding timing so can pull itself out of that hole?

That's diesel shit. You don't start gas powered anything at 1000 RPM. You telling me you're going to put a LSA or LT4 or Hellcat in 1:1 and womp it wide open at 1000 RPM? That's just asking for problems.

Show me some really precise dyno graphs that start +/- 50 RPM. Multiple runs overlaid that look exact preferably. Bonus points for .drf or .wp8 files with accompanying data logs.

You have to remove all the variables. As long as there's variables, there's no A vs. B.

[GHuggins]

OK, to be clear, you posted 2 dyno graphs with 4 pulls and only 1 out of 4 was a pull from 1000 rpm. The other 3 were from a reasonable 2000ish RPM.

Trying to say someone should check how good their tune is by starting pulls at 1000 rpm at 1:1 is just idiotic. Definitely the dumbest thing I've seen you say on here.

Take the OP for instance. His vehicle is a LS3 C6, manual trans with a stage 2 or 3 cam. The cam kills low end torque. Why would I or anyone else start a pull on this vehicle at 1000 RPM?

Take his vehicle totally stock, strap it down. Put it in 4th gear, start a pull at 1000 RPM. What happens? On a stock calibration it pulls all the timing out of it to keep from detonating the tops of the pistons off. What do we do, try adding timing so can pull itself out of that hole?

That's diesel shit. You don't start gas powered anything at 1000 RPM. You telling me you're going to put a LSA or LT4 or Hellcat in 1:1 and womp it wide open at 1000 RPM? That's just asking for problems.

Show me some really precise dyno graphs that start +/- 50 RPM. Multiple runs overlaid that look exact preferably. Bonus points for .drf or .wp8 files with accompanying data logs.

You have to remove all the variables. As long as there's variables, there's no A vs. B.

I never had a problem with it... AND yes, even though the dyno didn't always record for whatever reason - it was touchy - I started all manual car pulls from a rolling idle. Call it idiotic. Call it what you want. Pretty sure all car owners will stomp them from a light at some point??? If your tune can't handle that, there might be other problems.

[hjtrbo]

Must be something in the water with Allens posts, they always involve some sort of shit fight.

[GHuggins]

Must be something in the water with Allens posts, they always involve some sort of shit fight.

Yeah, I was honestly going to stay out of them too, but due to what I've seen over the years with injection timing I thought I would show some defense for it.

The dyno actually recorded all 3 for this one Ed. This is that zl1 that I had to start dynoing things differently for Depending on where in the rpm band you look, I wound up running 15 to 100 degrees more retarded injection timing over the factory 520 setting.

These dyno's are after it was changed to a 416 stroker, but just wanted to show the pulls starting at a rolling 4th gear idle. No engine problems. No dieseling and one of the most dependable back to back running cars I've ever seen. You can however still see a bit of a lean tip in.

zl1 dyno.jpg