End of Injector Pulse.

General Tuning Questions And Discussions
Ken
Posts: 206
Joined: Tue Dec 17, 2013 1:05 am

End of Injector Pulse.

Post by Ken »

Been away from home for some time now, so using alone time to familiarize myself with a parameter within the ls1 tunes that I find little mention of here or other forums, plus the info I've found isn't definitive.

The end of injection timing in EFILive is referenced in mS, a total of 8, and in HP tuners defined as reference periods, again in a total count of 8, leading me to think that each single unit represents 90 degrees of crank timing.
Why EFILive reference it in time (mS) makes no sense that I can rationalize as the differing crank speed means differing timing settings.

Probably spent the best part of 2 months researching this feature, one that in my opinion should be altered when aftermarket cams are used, and prior to making any other alterations in the tune, however I don't want to tune altered motors without having this option available in TunerPro.

Has anyone else here written an xdf with this feature added that is willing to share, if not, i'll keep chipping away at it til I work it out myself I guess.

My first effort at creating the entry for the 12202088 and 12225074 has resulted in editing the right area, though I'm not yet 100% sure that the math equation I've entered into the xdf is a correct representation of a feasible setting, though the EFILive example results in the correct change according to EFILive when opened with the program after editing with TunerPro.
Would love to have the tools to jack into the injector circuit and measure the differing end of timing figures myself with regards to the differing temperatures.

I have an xdf that equates the figures the same as EFILive, though I've dropped it in favour of referencing the final figures in crank degree's as in this pic below.
Thanks lads.
WH_InjTiming.PNG
WH_InjTiming.PNG (25.47 KiB) Viewed 7852 times
User avatar
antus
Site Admin
Posts: 8228
Joined: Sat Feb 28, 2009 8:34 pm
cars: TX Gemini 2L Twincam
TX Gemini SR20 18psi
Datsun 1200 Ute
Subaru Blitzen '06 EZ30 4th gen, 3.0R Spec B
Contact:

Re: End of Injector Pulse.

Post by antus »

Dimented 24x7's 2088 XDF states
Defines the desired point at which the fuel injection pulse will be completed by. The value is the position for the current cylinders next firing pulse in degrees after the current cylinders last reference pulse. By altering the position fuel is delivered, the fuel is given sufficient time to vaporize before it is ingested."
It also has it as a 16 bit value, with equation (X/256)*(720/Y).

I would suggest you use his XDFs as they are very high quality.
Have you read the FAQ? For lots of information and links to significant threads see here: http://pcmhacking.net/forums/viewtopic.php?f=7&t=1396
User avatar
The1
Posts: 4694
Joined: Mon Jan 04, 2010 10:23 am

Re: End of Injector Pulse.

Post by The1 »

provided you have sufficiently larger injectors than required then you can get some benefit, as you have to squeeze the whole injection event smaller so you can shift the whole event around in that one cycle.

i did a EOI Calculator Spreadsheet here which maybe of some reference you can modify it to suit.

viewtopic.php?f=27&t=2518
Ken
Posts: 206
Joined: Tue Dec 17, 2013 1:05 am

Re: End of Injector Pulse.

Post by Ken »

The1 wrote:provided you have sufficiently larger injectors than required then you can get some benefit, as you have to squeeze the whole injection event smaller so you can shift the whole event around in that one cycle.

i did a EOI Calculator Spreadsheet here which maybe of some reference you can modify it to suit.

viewtopic.php?f=27&t=2518
I felt first hand what you're referring to regarding larger injectors tonight.
I kept pushing the setting 4 hex bytes larger in size till it started to lose mid to top end, what it did to the bottom end was brilliant.
Since the numbers change at odd increments with one hex byte difference, I chose to edit the setting using the hex entry instead, pushing it 4 hex bytes at a time, full temp original setting was 058D, 0591 was noticeable, 0595 was better again with no loss to top end, 0599 and 059D was killing top end, though bottom end was better with every change.
Considering an injector change now to keep the 059D setting, needed less throttle than ever to pull hard off the mark.

One thing I did notice with the original induction pulse noise was that any setting higher than the 0599 setting, the induction cyl pulses could no longer be heard as individual pulses, and instead sounded like a carburettor'd constant induction noise.

Though after seeing the End of injector timing trim settings on the VX HSV, I'm considering changing back to the 059D injector timing setting, and bringing the EOIT trim setting back to 0595 to see if it affects to top end at all.

Also, according to EFILive, the trim setting isn't supposed to be greater than that of the EOIT setting, yet it is in the VX HSV tune from GM, I'm finding lots of contradictory information regarding this feature both online and within the popular tuning programs themselves.

EOIT.
VX_CS_InjTiming.PNG
VX_CS_InjTiming.PNG (61.41 KiB) Viewed 7537 times
EOIT Trim.
VX_CS_InjTim_Trim.PNG
VX_CS_InjTim_Trim.PNG (63.26 KiB) Viewed 7537 times
antus wrote:Dimented 24x7's 2088 XDF states
Defines the desired point at which the fuel injection pulse will be completed by. The value is the position for the current cylinders next firing pulse in degrees after the current cylinders last reference pulse. By altering the position fuel is delivered, the fuel is given sufficient time to vaporize before it is ingested."
It also has it as a 16 bit value, with equation (X/256)*(720/Y).

I would suggest you use his XDFs as they are very high quality.
I recently found and looked through Dimented 24x7's 2088 XDF, the "Y" variable in the equation just seems to be referenced to cylinder count, odd for an LS1 tune, whats even more odd about that is if I change the cylinder count to 6, the 720 degree limit becomes 960 degrees.
Other research has led me to thinking that the "Y" variable might be the Scalar "Target for Injector Pulse Boundary" if at all it's meant to have a "Y" variable.

Dimented's XDF:
Dimented_EOIT.PNG
Dimented_EOIT.PNG (57.49 KiB) Viewed 7479 times
The same outcome could be achieved using a math sum with either
X / 2.844444
or
X * 0.3515625

Posts on other forums have led me to believing that the boundary is a part of the math equation.
VCM Editor also shows the Boundary as being a part of the group associated with the Injector Timing options:
HP_Inj_Timing.PNG
If the Boundary is the "Y" variable, the math sum used in Dimenteds XDF gives errors when pushing the upper and lower boundary's of options, regardless of whether or not an upper and lower limit is set in the XDF, and just shows the word "inf" in the box of converted string.

However, using a math sum of "(X/256 + Y) * 45" and using the boundary as the Y variable, calcs spot on when setting the boundary to any of the upper and lower limits of 0 to 8.
This is using Dimenteds math sum and boundary as Y variable, the scale of timing ends up rather large:
My_EOIT_2.PNG
My_EOIT_2.PNG (58.91 KiB) Viewed 7479 times
Though using the equation I showed above: "(X/256 + Y) * 45"
Ends up like this below, the scale of which is much narrower, and calcs without error when using the max upper and lower boundary options:
My_XDF_3.PNG
My_XDF_3.PNG (59.14 KiB) Viewed 7462 times

Though this is all just a calculated guess on the research I've done so far, the jury's still out.
At the end of the day, are these figures meant to represent crank degrees from a 720deg scale, or 90.
Last edited by Ken on Sun Jul 28, 2019 10:44 pm, edited 1 time in total.
Ken
Posts: 206
Joined: Tue Dec 17, 2013 1:05 am

Re: End of Injector Pulse.

Post by Ken »

Looks like I reached the max pic attachments in the previous post, so had to do another for this one.

If the EOIT is meant to be a scale of 90 degrees and not 720, and the boundary is meant to be the "Y" variable, the below equation does the job.
My_XDF_4_90.PNG
My_XDF_4_90.PNG (57.85 KiB) Viewed 7462 times
User avatar
antus
Site Admin
Posts: 8228
Joined: Sat Feb 28, 2009 8:34 pm
cars: TX Gemini 2L Twincam
TX Gemini SR20 18psi
Datsun 1200 Ute
Subaru Blitzen '06 EZ30 4th gen, 3.0R Spec B
Contact:

Re: End of Injector Pulse.

Post by antus »

Internally its not degrees at all but reference counts, but the point of the XDF is to convert the data the program uses to something that makes sense to humans. So if you get the right result, and its a clear and measurable unit in the XDF which works for humans with no side effects, then its a win. I think internally its using low res timing signals. Is it 8 degrees per low res pulse?
Have you read the FAQ? For lots of information and links to significant threads see here: http://pcmhacking.net/forums/viewtopic.php?f=7&t=1396
User avatar
The1
Posts: 4694
Joined: Mon Jan 04, 2010 10:23 am

Re: End of Injector Pulse.

Post by The1 »

im not sure how other pcms and code do it i only know of ecotec pcm's and have lost most of my memory of it, it's been a while but i hope/think this is right.

The pcm's injector counter is synchronized on the first CAM pulse then Injector 5 is synced with 3x pulses, EOI then uses 3x Pulses.

So if you used X as calc that would give you the 3x pulses amount, So the scalar "Injector - Fuel Boundry Located Cal Deg After 3x REF" will be the pulses before the intake valve opens as a buffer, then add on the EOI Table amount to get the total 3x pulses when injection will finish before intake valve opens. More pulses more delay. I think that's right lol

As you have found there is a limit to how much fuel you can squeeze in each cylinders time window, theres certainly a torque gain to be had if you have the injectors for it because when the engine is hot the valves will help atomize the fuel.

The calc used in enhanced converts it to Crank degrees.
kojab
Posts: 440
Joined: Sun Mar 22, 2009 11:52 am
cars: VT V6 supercharged in a corolla
Location: Sydney
Contact:

Re: End of Injector Pulse.

Post by kojab »

First time to use Youtube so I hope you will be able to view this URL.



RTLS1 is changing the end of injection time in real time.

I am sorry about the poor quality of this video and promise that I will get it better next time.

With a few more changes to my LS1 bench harness and with the aid of a Picoscope I was keen to see exactly where the end of injection was occurring.

I recently added "Injection Timing" and "Injection Timing Trim" along with HPtuners "Injection Boundary" settings to the "RTLS1"s config folder for real time operation in our RTLS1 tuning tool and so far some of my findings are a bit of a mystery.
In the Video clip we are starting at 0 and increasing the "Injection Timing" by 100 dec (64 hex) at each step. We can step it at any rate if we wish but decided to use 64 hex to keep the video clip short.

Mystery number 1 is the "Injection Timing Trim" does nothing after moving it no matter what bench conditions we could apply to it.

Mystery number 2 is HPTuners "Injection Boundary" Timing which also gave no change to the overall end off injection timing. Could this be why EFILive did not include it in their editor.
Ken
Posts: 206
Joined: Tue Dec 17, 2013 1:05 am

Re: End of Injector Pulse.

Post by Ken »

antus wrote:but the point of the XDF is to convert the data the program uses to something that makes sense to humans.
In which my aim was to convert the used reference point to an actual output figure that reflects commonly used units of measure, crank angle was what other's have attempted to use.
I've found 4 different XDF's online from reputable members of forums, all of which use different math equations, and all of which show a different crank timing for the same hexadecimal input, one of which you yourself recommended to use.
Now I know I could just use the Hexadecimal units or math converted decimal units, and attempt to remember what works best with whatever setup I use it on, though I would really like it if I could correlate an EOIT to crank angle with different cams used.
The1 wrote:So if you used X as calc that would give you the 3x pulses amount, So the scalar "Injector - Fuel Boundry Located Cal Deg After 3x REF" will be the pulses before the intake valve opens as a buffer, then add on the EOI Table amount to get the total 3x pulses when injection will finish before intake valve opens. More pulses more delay. I think that's right lol
So the boundary is a part of the math equation then.?
The1 wrote:The calc used in enhanced converts it to Crank degrees.
So in that instance it's converted to crank degree's, I'll check that out, thanks.!
kojab wrote:I recently added "Injection Timing" and "Injection Timing Trim" along with HPtuners "Injection Boundary" settings to the "RTLS1"s config folder for real time operation in our RTLS1 tuning tool and so far some of my findings are a bit of a mystery.
In the Video clip we are starting at 0 and increasing the "Injection Timing" by 100 dec (64 hex) at each step. We can step it at any rate if we wish but decided to use 64 hex to keep the video clip short.

Mystery number 1 is the "Injection Timing Trim" does nothing after moving it no matter what bench conditions we could apply to it.

Mystery number 2 is HPTuners "Injection Boundary" Timing which also gave no change to the overall end off injection timing. Could this be why EFILive did not include it in their editor.
I wonder if the 2 settings which did nothing, were expecting to see other inputs like lambda and other sensor inputs that change when running realtime.
Also, do you happen to have a way of seeing at what crank angle the injector ends at with each change, thanks.

EDIT:
In the 3 x Ls2 tunes delcowizzid posted on the forum here, 12609099, 12612381 and 12619078, I've managed to work out the hex address to edit the injection timing, should there be a need for it.
User avatar
The1
Posts: 4694
Joined: Mon Jan 04, 2010 10:23 am

Re: End of Injector Pulse.

Post by The1 »

the raw equation as in Hex would be 3x pulses i would think, then math equation is converting it to crank degrees
Post Reply