Camshaft and Injector Timing

[Proxses]

Injection must start before the intake valve closes. Not enough time otherwise.

[Higgs Boson]

Injection must start before the intake valve closes. Not enough time otherwise.

In a port-injected engine the fuel can be injected over much of the four-stroke cycle, with injection ending as the intake valve closes. At, say, 10,000 rpm this injection period may be something of the order of 12 ms. In a DI engine, injection can’t even start until the inlet valve opens, and then only when there is no chance of the fuel escaping through the exhaust valve during valve overlap. Start of injection will therefore be at or close to exhaust valve closing time.

www.formula1-dictionary.net/direct_injection.html

[Higgs Boson]

A GDI system also has more flexibility regarding when in the combustion cycle the fuel is added. MPFI systems can only add fuel during the intake stroke of the piston, when the intake valve is open. GDI can add fuel whenever it needs to. For example, the some GDI engines can adjust the timing so that a smaller amount of fuel is injected during the compression stroke, creating a much smaller, controlled explosion in the cylinder. This so-called ultra lean burn mode sacrifices a bit of outright power, but greatly reduces the amount of fuel used during times when the vehicle requires very little grunt (idling, coasting, decelerating, etc.).

GDI engines also react more quickly to these changes in timing and amount of fuel addition, increasing driveability. Additionally, the vehicle is able to more quickly adjust based on inputs from sensors located downstream from the combustion chamber, keeping the dirty emissions blowing out of the tail pipe in check.

Some automakers have even experimented with using GDI to fire an additional burst of fuel into the cylinder to create secondary explosion during the combustion cycle, resulting in potentially even more power and efficiency.

http://www.cnet.com/news/whats-so-gr...s-of-car-tech/

[Higgs Boson]

Ok so now that I slept, I am sure start of injection occurs during the intake stroke and ends during compression.

Based on some light reading, it seems like it would be a good idea to start injection at WOT as early as possible (when exhaust valve closes) and inject fuel as late as possible at idle and decel.

any opinions?

so if this is the case, why can't we lower the idle values for a later injection? why are they 310ish when WOT is 340 and decel is 275? what happens if we make idle 275?

if we change the SOI to something like 180 or 160 does that constitute ultra lean burn / stratified?

finally, how closely will the SOI Base table be followed? is the pulse width calculated inside the tune based on the SOI value? does it spray harder, for instance, with a later SOI or just less or just right through combustion?

[Ben Charles]

I have some new info I have been working on, when I get my mind wrapped around it a little more Ill post up!!

[J-CUE]

Ben,
Any updates yet?

[Higgs Boson]

Since there are now several threads that come back to injection timing I figured I would bump this one and see if there is any new info out there.

I've been doing a little browsing around. It looks like later injection will prevent wetting the piston/cylinder walls and can give more cylinder pressure while early injection more completely burns the mixture. Thoughts?

Now I wonder why not inject as late as possible at stoich related conditions and early during PE related conditions.... Thoughts?

With that said, when we inject late, and we use Stecks SOI Calculator, is it possible to start spraying at a value less than 180? Would that not be stratified injection since it will be during the compression stroke? These cars only do homogenous (intake stroke) as I understand it.....is that hard coded or is that just the settings in the factory tune? Can we command an SOI of 160?

Anybody got anything?

[Ben Charles]

Ill see if I can find some of the SAE papers I have read on SOI and pressure and I will post up.

Now I wonder why not inject as late as possible at stoich related conditions and early during PE related conditions.... Thoughts?

I think that would be a great idea, on my setup I have made injection early at .68 to 1.10 g/cyl and left alone the other cells for the time being. I plan to start tuning tomorrow, so I should have more info.

[Higgs Boson]

Bumping this thread again......

It's funny to read back and see us discovering this, makes me feel dumb, lol....


Anyways, I want to revisit the SOI vs EVC topic specifically. When my engine was NA, I didn't need to Start Injection anywhere near EVC so it was pretty much irrelevant. A lot of tunes are not Starting Injection before EVC even with FI applications, lower powered ones in the 600s mostly that I see.

With my 416 and 8 or 9 psi, LT4 HPFP, LT4 injectors, 24 Mpa, and an extra lift pump on the tank, I am still running out of fuel in the midrange. My EVC is 5 degrees ATDC so ideally I want to SOI at 355.....but it's not even close to enough time to complete the spray.

Does anyone have any good input on how early it's ok to Start Injection? Can I spray 70 or 80 degrees of open exhaust? That would be during the exhaust stroke, won't I just lose it all out the exhaust?

With port motors all the fuel came in with IVO and some is lost during overlap.....but there was an open intake valve as well and the air maybe helps retain the fuel? So the earliest I want to spray would be at IVO, which only gives me 362.....

Using DSteck's calc I need around 450. If I am blowing fuel out the exhaust valve for 95 degrees I would think my wideband is going to feel it's rich....and will any of that fuel contribute to combustion anyways?

Any ideas or advice? If there were bigger HPFP or injector options I would buy it right now!

I guess the solution is METH, which I was trying to avoid in favor of E85. I guess I could run both.

Unless at higher RPMs/loads everything is moving so fast that it doesn't matter......what RPM does that start to take effect?

[757performance]

Any ideas or advice? If there were bigger HPFP or injector options I would buy it right now!

Unless at higher RPMs/loads everything is moving so fast that it doesn't matter......what RPM does that start to take effect?

Curious, I know COMP has a few grinds which change the number of lobes on their GenV cams, would the extra strokes be enough to increase rail pressure to at least assist you in fueling if not supply all you need, or do you run into the same problem as the diesel guys and hit part/software limits too quickly?

[Higgs Boson]

my rail pressure doesn't go down until my injector duty cycle shoots up.

i used to be able to watch my fuel pressure drop and IDC rise. now with two lift pumps and LT4 HPFP pressure is good. in fact, my rail pressure is still close to stock values (13-14 Mpa) even with IDC max'd out.

[757performance]

Nice to see the HPFP and lift pumps are keeping pace, I just hope more injector choices come out soon. I'm not a Bosch dealer so I may be out of the loop, but I have yet to see any larger DI injector choices other than DeatschWerks' Ford line. I can only imagine the issues people will be having if injector data isn't damn near perfect... but once larger ones are available I'm sure it's only a matter of time until we see a belt-driven HPFP setup... drool.

[andrewzpsu]

Your wideband won't read rich if the fuel is unburnt. Remember, it detects oxygen, not fuel. This is the same principle when your car dead misses (like bad spark) and you can see the lean blips in the factory o2 sensor mv's.

[rjw]

bumping for further discussion

[rjw]

Anything new on this Ben, Higgs , anyone?

thx

[Higgs Boson]

Leave it stock except where you need more injection time. You need more injection time when pulse width starts going over 4.5ms, when it goes over about 6ms, it's like being out of fuel/lost control of your fueling. I don't recommend injecting when the exhaust valve is open. A stock cam closes the exhaust valve before TDC at .050 and maybe a little after TDC depending on the .005 specs, which means I like to max out my SOI around 355-360. A bigger cam will give you even less room as the exhaust will close later.

Some tuners will go up to 415, 430, etc but that's injecting with maybe 90 degrees of open exhaust valve. If you look at it like a port motor, you could start injection at intake valve open which could be in the 390 range and with enough overlap/reversion I guess the fuel could be pulled back into the cylinder to contribute to combustion but typically these are boosted applications that need so much lead time and I am not so sure how efficient that will be.....which is also why you don't want lots of overlap on boosted engines, everything blows out at overlap.

Personally, I would keep SOI at a limit of EVC on boosted engines and could possibly go as early as IVO on NA engines, although it wouldn't be needed I don't think.

[rjw]

I will be doing some testing a bit later. Old school thinking had me taking timing out for safety reasons, resulting in some random knock retard. when I put all the timing back in, it went away. Not sure if it was false knock or what? Hope to have more later.

thx

[Higgs Boson]

I will be doing some testing a bit later. Old school thinking had me taking timing out for safety reasons, resulting in some random knock retard. when I put all the timing back in, it went away. Not sure if it was false knock or what? Hope to have more later.

thx

injector or spark timing? this thread is about injector timing, not spark lead.

regarding spark, log Knock Retard, not Total Knock Retard. Also disable Burst Knock in the tune. After that, you can log all the spark pids and see what is pulling timing and see if you want to do something about it or not. it might not be knock, it might be TM.

[rjw]

so far with burst knock out of thepicture, there is no Knock retard, but something is still taking away timing at points, but it is getting better.....thx

[rjw]

injector or spark timing? this thread is about injector timing, not spark lead.

regarding spark, log Knock Retard, not Total Knock Retard. Also disable Burst Knock in the tune. After that, you can log all the spark pids and see what is pulling timing and see if you want to do something about it or not. it might not be knock, it might be TM.

I thought that ignition timing should be somewhat in sync with injector timing?