Injector Timing? Reference Periods? refereencing what?

[Billiumss]

What a hell of a read!!! Big thanks to all in this thread that explained in detail about EOIT, thanks for contributing to our tuning community.

This reminds me of what Greg Banish told me in one of his tuning classes back in 2007 or 2008. He called it transport delay. He mentioned that Ford had it and GM had something similar but never went on to explain it in more detail.

So basically what I got out of this thread for GEN 3 LS1 PCMs:

- By adjusting (increasing) EOIT "Normal" , you can reduce or eliminate most exhaust smell at idle when not running cats on a cammed LS1
- LS1 PCMs are limited on what you can control, not 1 value solves all
- .11 = 10 degrees of crank rotation
- Leave the "Injection Timing Boundary" at 6.50 (stock value)
- Leave the "Makeup Injection vs ECT" at (stock values)
- Don't oversize your injectors for your application
- The bigger the cam and more overlap the higher the "Normal Injecting Timing vs ECT" values
- EOIT normal values can range from 6.0 all the way up to 6.5, depending on valve events

I know people are looking at this information (this whole thread) and asking themselves, what value should I put in there for my combo. There is not one answer, every car will have a different value. Unlike the GEN 4 E38, E67, E40 ECMs that have way more control over EOIT, the best all of us GEN 3 LS1 guys can do is "get it in the ballpark" approach EOIT since all we can change is over ECT.

If any of my summary is wrong or misleading, let me know so I can correct it so people that research this in the future has something to start with.

Bill

[Billiumss]

Any updates on this?

Bill

We chose the magic stick 3 camshaft for my new 408 Tx Speed stroker. After reading about all the EOIT information on the internet and studying the books I have on tuning, it appears to me that the "Boundary" should be set approx. 22 degrees earlier than stock because of how much earlier the inlet valve opens than stock, and that the "Target" needs to be set approx. 22 degrees later than stock because of how much later the exhaust valve closes than stock. To accomplish the latter, I must increase Target by a total of 44 degrees to make up for the amount the Boundary was reduced plus the additional 22 degrees of delay from stock setting. These changes are obtained in HP tuners by subtracting 0.25 reference periods from the stock Boundary setting and adding 0.500 reference periods to the stock Target settings. Narrowing the injection "window" on both ends, which is what this will do, makes a lot more sense for a performance camshaft with valves opening sooner and closing later than stock than what I've seen in most other posts. I will have my car running around the middle of next week and will let you know how this works.

[Stuffs]

Great summary post too.

[mowton]

What a hell of a read!!! Big thanks to all in this thread that explained in detail about EOIT, thanks for contributing to our tuning community.

This reminds me of what Greg Banish told me in one of his tuning classes back in 2007 or 2008. He called it transport delay. He mentioned that Ford had it and GM had something similar but never went on to explain it in more detail.

So basically what I got out of this thread for GEN 3 LS1 PCMs:

- By adjusting (increasing) EOIT "Normal" , you can reduce or eliminate most exhaust smell at idle when not running cats on a cammed LS1
- LS1 PCMs are limited on what you can control, not 1 value solves all
- .11 = 10 degrees of crank rotation
- Leave the "Injection Timing Boundary" at 6.50 (stock value)
- Leave the "Makeup Injection vs ECT" at (stock values)
- Don't oversize your injectors for your application
- The bigger the cam and more overlap the higher the "Normal Injecting Timing vs ECT" values
- EOIT normal values can range from 6.0 all the way up to 6.5, depending on valve events

I know people are looking at this information (this whole thread) and asking themselves, what value should I put in there for my combo. There is not one answer, every car will have a different value. Unlike the GEN 4 E38, E67, E40 ECMs that have way more control over EOIT, the best all of us GEN 3 LS1 guys can do is "get it in the ballpark" approach EOIT since all we can change is over ECT.

If any of my summary is wrong or misleading, let me know so I can correct it so people that research this in the future has something to start with.

Bill

Actually the sizing of the injectors will have a large effect on the EOIT as they will relate to the time (crank degrees) it takes to complete the injection cycle. The larger the injector, the shorter the time needed to inject a similar mass of fuel. Going larger can compensate for Large cam overlap eoit issues. Too small an injector will exacerbate the issue. I agree, don't go too big but I would say sizing at 60-70% IDC or 10-12ms at 6000 rpm (19ms is 100% DC) would be acceptable.

This post was and continues to be a great source of information :-)
Ed M

[pontisteve]

Transport delay is the delay between when combustion takes place, and when the O2 sensor reads it's signal. The further back the O2 sensor is placed, the more of a delay there is. Kind of like an explosion taking place in the distance, and a few seconds later you hear the boom. If you stand next to the explosion, you will hear the boom instantly, without delay. The transport delay has nothing to do with injector timing. It has to do with offsetting the O2 sensor timing.

The bottom line with injection timing is that a cammed up car has a lot of valve overlap. So all that fuel sitting on the back side of the intake valve ends up getting tossed out the exhaust valve as soon as the intake valve opens. This gets nasty out the tail pipe, while also running the cylinder lean. If you delay the beginning of that injection until after the exhaust valve closes, then it won't get tossed out the exhaust. But it won't have as much time to vaporize the fuel on the back side of the closed intake valve either, so we don't do that unless we have to.

You can try to calculate EOIT by figuring out what BOIT is, but the injector size becomes a significant factor in how long each injection pulse is. It's almost easier to just add 10 degrees of injection timing at a time, until you feel the motor runs best. 20 to 50 degrees of additional injection timing is probably the area you will end up finding works best. GM uses these periods. We've come to understand that each period is 90 degrees of crank rotation, so adding .5 periods is adding 45 degrees of injection timing.

[yellz06]

.11 does not equal 10 degrees.

[pontisteve]

.11 does not equal 10 degrees.

Each reference period is 90 degrees. .11 x 90 = 9.9 degrees.

[yellz06]

Where do you get that information? each reference period is NOT 90 degrees.

[pontisteve]

Ok, then what is each reference period, and where do you get your information?

[anniversaryss]

With a gen 3 pcm we if we are just moving the normal injection vs ect.
Are we adding or subtracting to that table?

[pontisteve]

The table reflects the 720 degree engine cycle. Making the injector delay numbers larger makes the End of Injection time later in the engine cycle, also known as increasing the delay.

There is still much debate about what exactly the numbers mean, when the factory uses reference periods. I long thought based on earlier enthusiasts work that each reference period represented 90 degrees. So a number of 6.0 meant 6*90 = 540 degrees EOIT.

But when you dig into what GM is calling reference periods, it appears they are referring to the number of periods in the crank signal. On the older LS engines, the crank sensor is a 24x pulse, meaning there are 24 ups and 24 downs, and 24 periods in the crank signal per crank revolution.

On newer vehicles, GM sometimes uses actual crank degrees of EOIT, which would be a direct representation of EOIT in the 720 degree engine cycle. So an EOIT of 360 would mean the injector would quit spraying at 360 degrees (one crank revolution). That means the injector would quit spraying at exactly top dead center on the exhaust stroke. 0 and 720 (same thing) represents TDC compression.

Guys, one of these days I'm going to get back on this and hook my Picoscope up to a stock LS1, and start moving the EOIT around while noting when the scope tells me the injector shut off. With my in-cylinder pressure transducer, I can determine what is true TDC, and that will allow me to map out the 720 degree engine cycle, and where in relation to that does the injector fire and quit firing.

[mowton]

So, I have all that down (cam events, reference periods and even ms in some earlier Gen III) and Gen IV I understand and the 720 crank degree diagram works for me. The Gen III starting at 784 degrees is what I have always pondered. My assumptions was it was due to the 24x/1x arrangement which didn't provide enouugh resolution (within 90 degrees and some times 45 degrees) to identify cylinder position ID so they added 64 degrees (app half of 90+40=135/2 = 67.5 average error...but not exact :-( ) to the cylinder cycle to come up with the EOIT equation of -784+(Boundary + Norm ECT) or 300.5 crank degrees (cd) based on typical settings of the Boundary of 585 cd (6.5 rp's) and Normal ECT of 499.5 cd (5.5 rp's).

Gen IV EOIT at Boundary - (Norm ECT + Norm RPM) is typically 520 - (110+0) at idle which leaves us with a 420 cd EOIT.

So this is where I got the 65 degrees difference from Gen IV to Gen V based on the boundaries (585-520=65). Seeing as the cam events for both the Gen III and Gen IV stock cams are all around the same, I again assummed some how the 300.5 and the 420 EOIT's must be in the same place on the 720 degree engine/cam cylinder cycle but have never gotten the math to work out. Perhaps I need a good glass of scotch and a calculator to give it another go.

Ed M

[kingtal0n]

I'm pretty sure the answer isn't here. 26 pages and not a single person has posted which button to press and what the numbers inside do.

Why not adjust phase in realtime to get some kind of real world result. Whether a dyno number or a difference in the wideband O2 reading, go to a "better number" and use that number.
I would do this but my ECU is not realtime. I've done it this way for AEM 2jz-gte and other engines in realtime. The difference in throttle response at idle is incredible from poorly phased to just right and extra crispy.

I'm thinking about loading a realtime OS to a spare computer and doing this eventually because why not. When the wideband goes richer thats usually good fuel savings. fuel smell going away after a certain number could be just over overlap.

I'll do 1-bar disabled gate WOT dynojet run looking for power from injector phase as well. Technically, run it at the track back to back with varying numbers is another form of that.

I look at the OEM table and I see that maybe the fuel injects far earlier due to being cold. Following this logic might give an initial direction for testing.

[pontisteve]

From a driveability perspective, all we really need to do is increase EOIT enough so that at idle and low RPM the injector doesn't begin to spray until after the exhaust valve closes.

From a performance perspective, the results will probably be similar, but the dyno would ultimately determine what's best, using real time tuning.

From a mapping perspective, all of us are just guessing until we get a scope and an in-cylinder pressure transducer setup, so we can measure actual injector timing vs actual crankshaft position. I'm fortunately one of the few people that are likely to be interested in doing this test, that actually owns a Picoscope and a WPS-500, and HPT. I'll get on it one of these days when I have time and an LS1 handy.

[kingtal0n]

From a driveability perspective, all we really need to do is increase EOIT enough so that at idle and low RPM the injector doesn't begin to spray until after the exhaust valve closes.

Part of the problem is some of us can't figure out exactly how to do that. How does one increase EOIT on Gen3 ecu like a 411?

Which button do you press first. Normal? Makeup? Boundary? Still not really sure and can't really test it so....

[mowton]

Its a juggling act of Boundary, ECT and RPM on the Gen IV's and a less complex (and less control :-() approach on Gen III where we only have Boundary and ECT. I personally have a 720 degree chart that plots the stock cam and new cam events and I relate that to the stock settings and what is needed for the new cam. I do not believe, and it is supported by Greg Banish, that we know exactly what the settings are referenced to internally in the PCM. If you chart it out, you can see the differences in events and modify your inj timing to compensate. COmpensation can be as simple as moving the timing earlier/later to cover overlaps that are tolerable to moving the whole event until the exhaust valve closes. Just manipulate the variables to fit within your new scenario's. I look at idle, WOT and 3500 rpm to see that the fit is good across the entire rpm range.

Ed M

[kingtal0n]

Its a juggling act of Boundary, ECT and RPM on the Gen IV's and a less complex (and less control :-() approach on Gen III where we only have Boundary and ECT. I personally have a 720 degree chart that plots the stock cam and new cam events and I relate that to the stock settings and what is needed for the new cam. I do not believe, and it is supported by Greg Banish, that we know exactly what the settings are referenced to internally in the PCM. If you chart it out, you can see the differences in events and modify your inj timing to compensate. COmpensation can be as simple as moving the timing earlier/later to cover overlaps that are tolerable to moving the whole event until the exhaust valve closes. Just manipulate the variables to fit within your new scenario's. I look at idle, WOT and 3500 rpm to see that the fit is good across the entire rpm range.

Ed M

yeah we have 26 pages of this already and it does nothing. you are not explaining how the software works at all.
I'll try again
there are 3 buttons
Normal
Makeup
and something called boundary setting with a 6.5 in it

No clue what any of those buttons do or the numbers inside them, still has not been explained for 26 pages

pls allow me to re-phrase question so that it is not mis-understood further
Q if I add .1 to normal table What Does That do
Q if I add .1 to makeup table What Does That do
Q if I add .1 to boundary setting What Does That do

[mowton]

yeah we have 26 pages of this already and it does nothing. you are not explaining how the software works at all.
I'll try again
there are 3 buttons
Normal
Makeup
and something called boundary setting with a 6.5 in it

No clue what any of those buttons do or the numbers inside them, still has not been explained for 26 pages

Check out Post #240.....you are missing alot of good data in this Thread.....

Ed M

Quote from Chris on LS7 injection tables.


Boundary is the latest point in time that fuel can make it into the cylinder for the current injection period. It is measured in degrees AFTER TDC compression. It is also the earliest the injector can fire again after the completion of the last injection event. ie. the boundary defines the start/end of a complete injection cycle.

Normal End of Injection Target (EOIT) is the angle before the boundary (in degrees) that the normal pulse (the main pulse) should finish. The hardware calulates the start of injection time from the EOIT and the desired PW.

Makeup End of Injection Target (EOIT) is the angle before the boundary (in degrees) that the first makeup pulse should finish. Makeup pulses are extra pulses that can be injected to inject more fuel during an injection cycle rather than having to wait for the next cycle. There can be more than one makeup pulse but the makeup EOIT specifies the EOIT of the first makeup pulse.

The makeup PW minimum is a minimum PW for the makeup pulses, they cannot be shorter than this (if they are they don't happen).

So you have the boundary that defines the injection cycle, a normal EOIT that specifies the EOIT for the main pulse, and finally a makeup EOIT that specifies the EIOT for any makeup pulses that might be needed if a fuel increase is commanded during an injection cycle but after the main pulse has occurred.

Hope that helps.

Chris...

[kingtal0n]

Boundary is the latest point in time that fuel can make it into the cylinder for the current injection period. It is measured in degrees AFTER TDC compression. It is also the earliest the injector can fire again after the completion of the last injection event. ie. the boundary defines the start/end of a complete injection cycle.

Normal End of Injection Target (EOIT) is the angle before the boundary (in degrees) that the normal pulse (the main pulse) should finish. The hardware calulates the start of injection time from the EOIT and the desired PW.

Makeup End of Injection Target (EOIT) is the angle before the boundary (in degrees) that the first makeup pulse should finish. Makeup pulses are extra pulses that can be injected to inject more fuel during an injection cycle rather than having to wait for the next cycle. There can be more than one makeup pulse but the makeup EOIT specifies the EOIT of the first makeup pulse.

The makeup PW minimum is a minimum PW for the makeup pulses, they cannot be shorter than this (if they are they don't happen).

So you have the boundary that defines the injection cycle, a normal EOIT that specifies the EOIT for the main pulse, and finally a makeup EOIT that specifies the EIOT for any makeup pulses that might be needed if a fuel increase is commanded during an injection cycle but after the main pulse has occurred.

Thanks for trying to help


I did see this but it is vague and not directed at Gen3 platform. He says EOIT is in "degrees" which ours is not.

ex trying to read it: It says "Normal End of injection is the angle before boundary". SO if boundary says 6.5 and Normal says 5.5 (which is not referenced in degrees but eh) then according to the statement "normal" is the angle before boundary, 5.5 "worth of degrees" before boundary. So 5.5 before the boundary of 6.5 is 1.0, or apparently 80-100 degrees. Which is completely arbitrary, confusing, and cannot possibly be correct.

Ex2 assuming it was a typo and not "the angle in degrees before boundary" but that he actually meant "the angle of EOIT that happens to be before boundary using the same reference angle as boundary" then the normal table numbers increasing makes the injection happen later and later, approaching boundary. Which makes sense so this is how I've been doing it.

However WTF is makeup, how do we deal with MAKEUP timing? One thing I read said something about they are added together. Whereas I am inclined to think it should be set as the same number as "normal" table, or later. If you go earlier with any extra pulses we just defeated our own original plan to inject after EVC.

With a realtime computer you can just sit there with the engine running and drop the number until you get some fuel smell and the wideband starts bouncing lean etc... someone do this to the makeup and normal tables plz ty

From Below:

You would increase the numbers to increase the injector delay. Makeup refers to some pulses that are used only in certain transient situations, so don't worry about that too much. I'm not sure about boundary's role yet. Normal is what you want to increase.

The problem is we don't really know what that number represents. For the longest time, I thought it represented 90 degrees, since there are 8 "normals" max, and the engine has 720 degrees of rotation in one cycle. So 90 makes perfect sense. Also 90 seems to line up pretty well with what the factory settings are. It is typical to see (on a warmed up engine) anywhere from 280 to 460 degrees of EOIT in Fords. Usually that number increases as engine load increases, presumably because the pulsewidth must increase and they need more time to inject the fuel.

But the further I looked into it, the words 'reference pulses' relates to crankshaft pulses. And there are 24 of them to represent 360 degrees of crank rotation. Google "LS1 24x crank signal" to see what that might look like. This leads me to believe there are 24 15-degree reference periods per crank rotation, or 48 periods per engine cycle.

I'm just not sure how those numbers somehow add up to 720 degrees. The only way we will ever know the correlation for sure is with a scope on the injector, and by changing the numbers in the tune and seeing what effect they have on the EOIT. For now though, if you make the numbers in the Normal table bigger, I believe that will increase your delay.

Thank you this makes a lot of sense and pretty much where I thought we were currently. It also answers the other fellows question about what to do with the number (make it bigger )

[pontisteve]

Part of the problem is some of us can't figure out exactly how to do that. How does one increase EOIT on Gen3 ecu like a 411?

Which button do you press first. Normal? Makeup? Boundary? Still not really sure and can't really test it so....

You would increase the numbers to increase the injector delay. Makeup refers to some pulses that are used only in certain transient situations, so don't worry about that too much. I'm not sure about boundary's role yet. Normal is what you want to increase.

The problem is we don't really know what that number represents. For the longest time, I thought it represented 90 degrees, since there are 8 "normals" max, and the engine has 720 degrees of rotation in one cycle. So 90 makes perfect sense. Also 90 seems to line up pretty well with what the factory settings are. It is typical to see (on a warmed up engine) anywhere from 280 to 460 degrees of EOIT in Fords. Usually that number increases as engine load increases, presumably because the pulsewidth must increase and they need more time to inject the fuel.

But the further I looked into it, the words 'reference pulses' relates to crankshaft pulses. And there are 24 of them to represent 360 degrees of crank rotation. Google "LS1 24x crank signal" to see what that might look like. This leads me to believe there are 24 15-degree reference periods per crank rotation, or 48 periods per engine cycle.

I'm just not sure how those numbers somehow add up to 720 degrees. The only way we will ever know the correlation for sure is with a scope on the injector, and by changing the numbers in the tune and seeing what effect they have on the EOIT. For now though, if you make the numbers in the Normal table bigger, I believe that will increase your delay.