white2001s10

iTrader: (4)

For those that really know:
Does the PCM switch from sequential to batch fire at a certain RPM or DC point?
What are those points?

Most importantly, when/if it switches, does it go to true batch fire doubling the injector firings? or does it simply just extend the PW on the original firing and call it batch once it extends over 360* of crank rotation?

I'm doing some investigative work to figure out how the PCM keeps delivering more fuel even after (reported) 100% DC on the injectors is exceeded.

macsperformance

iTrader: (2)

I assume you are talking about your factory ECU so I am sorry but I can not address your first question. To be honest I have never heard of any ECU changing the way it fires the injectors like going from sequential to batch fire like you stated in the middle of operation.

But the last two I can tell you with much confidence.

No injector on the market fires twice per engine cycle (2 engine revolutions). If additional fuel is needed the injector just stays open for a longer duration (duty cycle). Duty cycle is measured over one complete cycle (2 revolutions or 720 degrees of engine rotation on a 4 stroke engine).


Now for the next part I have seen some debate about this on the forum. And the very simple truth is if the ECU shows a duty cycle of 100% it means they are open 100% of the time and closed for 0% of the time of the measured engine cycle.

Think about this everyone. If your injectors are static (100% duty cycle and it is staying open for 100% of the 720 degrees of engine rotation) it can not be open longer that 100% of the measured engine cycle period. Just because an ECU reports duty cycles longer that 100% it does not mean the injector can physically stay open for longer than 100%. It just means this is the percentage over 100% it calculates it needs to achieve the target A/F it is programmed for. On some FMU cars the duty cycle shows well over 100 % on some factory ECU's and the reason for this is that most of the time the fuel system can not keep up with the required fuel needs so the computer calculates that it would need to be open longer than all the time to supply the needed fuel for that programmed value.

If engine speed dictates that you only have 10ms for the entire engine cycle to take place then 10ms would represent a 100% duty cycle. If the ECU needs 11ms to get enough fuel into the motor then the ECU will show a 110% duty cycle but in reality the injector can not supply fuel for that cycle any longer than 10ms. So it is held open 100% of the time that cycle and 100% of the next cycle and so on. If everything else were constant fuel pressure would be the only way to get more fuel into the engine in those given 10ms.


I hope this helps you friend
Chris Macellaro

Zombie

iTrader: (2)

macsperformance, you are very close. The injectors can be open for longer than 100% of the engine cycle. I've also verified it on my scopemeter.

The way the ECU works is it commands a pulse length for the injectors to stay open.

For example, I'll say that the ECU is commanding a 20mS pulse. At 4000 rpms a 20mS pulse is 75% duty cycle. Fueling needs change very little with rpms, so as the rpms climb to 6000 rpms, the ECU is still saying I need a 20 mS pulse and thats how long the injectors are staying open for (they never actually go static).

Once we pass the 6000 rpm mark our duty cycle is over 100%, the ECU doesn't know this though since it's just chugging away saying I like 20mS pulses. At 7000 rpms the ECU is still commanding 20mS pulses, but 100% duty cycle at 7000 rpms is 17 mS, so 3 mS of our 20mS pulse runs over into the next cycle and it is now calculated as 117% duty cycle.

I've personally found I was able to run 120% duty cycle with 42lb injectors on my turbo car with no i'll effects and it maintained it's 11.2:1 a/f even on long highway pulls. Not the smartest thing to do though.

I do not reccomend running over 100% duty cycle, the reason being is the injectors are still pulsed, they never stay completely on. What this means is that you will be supplying less fuel to your car when the injector turns off for about 1mS during one of it's cycles. If your car is at 6000 rpms and needs 20mS of fuel (100% duty) it's ok for the first pulse, but on the 2nd pulse it's robbing the fueling cycle of about 1mS so now your 2nd cycle that is wanting 20mS is only getting 19mS of fuel due to the injectors dwell time.

Hope all of this makes sense, it's a little hard to explain without drawing the diagrams.

macsperformance

iTrader: (2)

[QUOTE=Zombie]For example, I'll say that the ECU is commanding a 20mS pulse. At 4000 rpms a 20mS pulse is 75% duty cycle. Fueling needs change very little with rpms, so as the rpms climb to 6000 rpms, the ECU is still saying I need a 20 mS pulse and thats how long the injectors are staying open for[/sQUOTE]




Hello Zombie

You may have misread my prior post.



The information I stated concerns the ability of the injectors pintle to stay off of its seat for the entire duration of the engine cycle 720 degrees of engine rotation.

Pulse width and Duty Cycle are two different things all together when refering to injector operation.

Duty cycle is defined by the amount of time the injector's pintle is off the seat expressed as a percentage as opposed to the amount of time it is at rest.

If you need 20ms of fuel but there is only 10ms of time to cover 720 degrees of engine rotation the computer would estimate a 200% duty cycle.


The injector wave patterns I have seen on a scope will take (depending on the system) can take around one millisecond of response time from the ECU through the harness and then to open the injector. This is set and sometimes definable depending on the system. This is the amount of time that is took into account for the injector to actually open after the ECU commands it to open. This is the case under 100% duty cycle at WOT and a 100% duty cycle condition the ECU has already opened the injector and is keeping it on so there is no delay after the initial opening of the injector.



I hope this helps and I enjoy the tech talks
Chris Macellaro

white2001s10

iTrader: (4)

Please note that I didn't ask if fuel delivery increases past 100% D.C.
I am asking why fuel delivery increases after 100% D.C.

I'm not talking about a little bit either.
Here's just one example.
Car shows 85% D.C. naturally aspirated WOT with 11.8:1 AFR
Car shows 110% D.C. with 100-shot N2O reading 11.2:1 AFR
Car shows 150% D.C. with 200-shot N2O reading 10.4:1 AFR

After seeing it rich with the 100-shot of nitrous, I figure just spray more in (dry) and it will lean out some. Having stock injectors and stock pump, I didn't expect it to go richer with the added N2O, but it did. It's not a testing accident either. The car picked up a lot of power and showed no signs of being lean at all. It ran 11.0 @127 and 30* advance so it wouldn't have survived multiple passes if it had went lean. That's showing about 500hp. I was testing with a LM-1 wideband.
I've seen the same exact thing happen on other cars as well.

I'm very curious as to why/how this happens, though I am pleased that it does. I'd like to understand injector operation in the greatest of detail.
Please add anything else you know, or speculate.

Thank you for your time and responses so far.

Zombie

iTrader: (2)

macsperformance, i'm not sure if i'm understanding what you are saying. Are you saying that once the injector is open at 100% duty cycle it never closes, or that it closes after the commanded pulse width? Thats what it sounds like to me. It does close usually though, at least on every ECU i've ever scoped which is a few GM's and some Mazda's, maybe not all the way but it attempts to. The problem gets worse as the fuel pressure increases.

High impedence injectors are slower to react vs low impedence injectors, so that will make a difference as well.

The delay in opening is known as offset. Thankfully GM ECU's have great tuning software and grant access to this table which makes tuning different injectors much easier.

I can play around a bit and see where a low imp. injector doesn't want to close since I'm currently developing a fuel injection system controller. I don't have the ability to supply pressure or fluid to the injector though. I can test them electrically though.

joecar

Hmmm...

If you captured a scope screen, can you please post it (showing ms/div)...?

6000RPM => 100rev/s => 50cycles/sec => 20ms/cycle

100%DC is 20ms at 6000RPM (edit) ;

>100%DC would be into the next cycle, so what happens to the injector opening during the next cycle...?
(>100%DC would just look like straight horizontal line.)
each injector is fired once per cycle, right...?

an injector opens on each cycle, so if PCM is commanding >100%DC, when does PCM close injector (draw a diagram)...?
and when does it open it again for the next cycle...?

Bink

Great read!

Can someone answer joecar's questions?

white2001s10

iTrader: (4)

and what about my question?

macsperformance

iTrader: (2)

sorry friend I missed your post...


I do not know could be a multitude of things that are showing you those numbers.



Is your dry kit sprayed before or after the MAF?





Chris Macellaro

gameover

by definition injector duty cycle is InjPW / (time for 2 engine revs)

you physically can't inject more fuel than the time it takes for 2 engine revolutions. The PCM may command more but physically you can't get it.

I've never seen a car that doesn't lean out once the injector duty cycle hits 100%, you see it on FI cars with too smaller injectors all the time.

Chris...

white2001s10

iTrader: (4)

That's strange because it seems like that's all I do see. I see them get richer.

Have you tested the injector wires while they are supposedly running at 100% DC?

macsperformance

iTrader: (2)

Maybe it would help all of us if you could post a data log of this event taking place.




Is your nitrous injected before or after the MAF?



thanks
Chris Macellaro

white2001s10

iTrader: (4)

Dry system before the MAF.
Sure it would help but it takes more than one log and in the end you have to take my word for it that the 3 logs go together. There's the LM1 wideband, the EFI-live scanner, the vericom2000 g-meter, and witnesses.

The fuel was obviously there to put up 127 MPH with more fuel to spare. What brings up a question is that there was definately stock injectors doing the job.

In another case a camaro put up less MPH but similar E.T. It was showing 85% DC running NA, over 100% with a 100-shot, and 150% after jetting (dry) up... also with stock fuel system/injectors.

Jetting up with the nitrous well past 100% DC produced much more fuel flow, even with the extra nitrous going in and making the predicted amount of extra power.

This wasn't a case of "it feels faster" or "I think it's richer", but reliable observations shared by more than one set of people. The only evidence we have of supposedly being over 100% DC is from EFI-live. If the software is reporting on only 360* of rotation, then that would explain the numbers to an extent, but everything I've seen so far suggests that it reports for 720*.

I only have the LM1 and g-meter data in my possession, but I could bring at least 2 witnesses.
It would be more helpful if we concentrated on why/how it happens rather than if it happens.

A friend of mine is working on getting another LS1 car to use for testing and a complete set of logs could be put together. It looks like there will be no answer until then, and maybe not even then.

Zombie

iTrader: (2)

Here is a thread from turbobuicks.com that has some great info pertaining to injector flow vs duty cycle.

http://www.turbobuicks.com/forums/sh...threadid=20689

All the injectors he tested increased past their rated flow numbers after 90% duty cycle. I find that suprising.

I also find it strange that your fueling is increasing when you are so far over 100% duty cycle.

I ran my car to 120% duty cycle when I had 42lb injectors and was making about 500 rwhp. I never ran out of fuel which I though was very strange as well.

joecar

Pictures...?

white2001s10

iTrader: (4)

who? and of what?

joecar

Picture of scope screen.

gameover

here's a better way, log the injector pulse width using the scanner. The GM PCM doesn't actually return injector duty cycle as raw data. Then you can verify the duty cycle equation.

It's also easy to work out roughly what your injector duty cycle should be at WOT by taking the Dynamic Cyclinder Air PID (g/cyl) and divide by your commanded AFR, then divide again by the injector flow rate (in g/sec) at 0 Vacuum.

Chris...

joecar

Anyone...?