As most of you know our ecotecs are capeable of sequintal fuel injection up to a certain load/RPM no big deal right? well if you take a step back and look at our motors youll notice something.... out of place.
The majority of motors out there use what is called bank firing injection. this means that when the crank sensor senses that its bank of cylinders are up it fires. regardless if the one cyl, intake valve is closed (more for v6's and v8's). But with sequintal fuel injection the motor knows exactly which cylinder is on its suction stroke, and fires only one injector. However for this to work it needs valve reference wich is usually in the form of a cam signal. meaning it needs to know when the cam is about to open the intake valve
ok so whats the big deal?
We dont have cam sensors, but we have sequential fuel injection. so how does this work?
upon inquireing to my instructor at school he did some reasearch and found the awnser. The ecotec uses a waste spark ignition which means that it will pulse one coil pack for two spark plug. the reason this can work with out low voltage problems is because the non compressed side is on its exhaust stroke, and has litte resistance other the the intial jump of the plug gap. where as the other cyl has to be fired threw the compressed air and gas, and gets all the strength.
so what does this have to do with sequintial fuel injection? well someone somewhere came up with the idea that we can use a resistance check to determine which cyl is on its compression stroke. basicly put it sees which cyl has to work harder and knows which injector to fire next. thus elimanating one more moving part.
however, Its only good up to a certain RPM, making it still less usefull then a true cam sensor. oh and it makes hooking up a convential spark system without throwing a code, a real PAIN IN THE ASS,
"
boobs now with Riboflabin"
All they did was use one of these:
Click Me Nothing new there and can be done to any DIS ignition car.......been around for a long time.
it was news to me. just thought I would share
"
boobs now with Riboflabin"
umm yea its been well known that we have wasted spark.
My car may run 18s, but I can do your taxes in 10 seconds flat.
JBO lube - they would never have enough in stock and we'd never see RodimusPrime again
I posted about the OE ignition system maybe 2 years ago? It is a way cool system, and I don't believe MSD was marketing their product at the time. I actually posted a link to an article which describes the electronics within the module which monitor the spark voltage and current.
-->Slow
This technology looks to be owned by GM-Saturn and used by GM-Chevy/Pont. according to the patent information. Sorry I'd post some of the info but the program used to get the details and search through patents here at work wont let me copy anything out of the program. MSD was marketing their system longer than 2 years ago btw.
Joshua Dearman wrote:All they did was use one of these: Click Me Nothing new there and can be done to any DIS ignition car.......been around for a long time.
I still am confussed on how that works. I know when you do the DIS set up you need to get this also, but what if you go 1 step farther and do eather Individual coil on plug set up or the 2.2 ohv coils and use the plug wires? Do you still have to use number 1 plug wire for the signal?
^^^ WTF are you talking about? DIS (Direct Ignition System) is the same as COP ('individual' Coil On Plug). Why would you want to use the ohv 2.2 system on an Eco?
^DIS is not COP.....DIS is similar to "wasted spark COP", but not true sequential fire "COP". You can use this on the 2.2OHV since it is a DIS ignition system that uses 2 coils to fire 4 cylinders.
DIS stands for distributorless ignition system, NOT direct ignition system
and jcavi's discovery is old news, but still interesting stuff.
the eco in the Jbody (L61) has no cam sensor so there's no real way for the ecu to directly know which cylinder on TDC is firing a power stroke, and which one is on the exhaust stroke. it uses whats called "compression sense" whereas the resistance across the plugs is measured by the ignition cassette and based on that information the ecu can figure out which cylinder is on a compression stroke and which is on the exhaust stroke.
DIS however, is not new. waste-spark ignition has been around for a while, but the compression-sense of the L61 is pretty unique.
annoying and troublesome, but unique.
Thats weird. I wonder why they went that route. Sounds like a simple Cam Position Sensor would be easier. But I guess it cuts down on sensors/systems the car has.
PSN ID: Phatchance249
Quote:
annoying and troublesome, but unique.
you said it sister. that @!#$ makes after market ignition a whole barrel of fun
Shawn McKenzie wrote:Thats weird. I wonder why they went that route. Sounds like a simple Cam Position Sensor would be easier. But I guess it cuts down on sensors/systems the car has.
Money, did you really think there would be any other reason? its all about money. if GM spends say 20 bucks to make 15 million cam sensors, just so they can say SFI. they why not just use 20 cent resistors and get the same badge of honor.
but alas. its dead tech. all the best ecos now use cam sensors (lsj,lnf.......the other one.(2.4))
either way, It really is sweet, Im going to see what eles I can dig up on this and mabye try and figure out a way to aquire the signal to a useable source. I could play with it on the MS, and do some neat stuff.
I can see it now first CAV with displacment on demand.
"
boobs now with Riboflabin"
Jcavi wrote:
but alas. its dead tech. all the best ecos now use cam sensors (lsj,lnf.......the other one.(2.4))
either way, It really is sweet, Im going to see what eles I can dig up on this and mabye try and figure out a way to aquire the signal to a useable source. I could play with it on the MS, and do some neat stuff.
I can see it now first CAV with displacment on demand.
Uh, the LNF and 2.4L eco both have variable valve timing, hence the need for a cam position sensor. The saab 2.0L ecotec uses compression sense similar to the 2.2L but with coil on plug.
Don't mean to be rude but a 4 cyl with displacement on demand!?!
DaFlyinSkwirl (PJ) - APU wrote:DIS stands for distributorless ignition system, NOT direct ignition system
and jcavi's discovery is old news, but still interesting stuff.
the eco in the Jbody (L61) has no cam sensor so there's no real way for the ecu to directly know which cylinder on TDC is firing a power stroke, and which one is on the exhaust stroke. it uses whats called "compression sense" whereas the resistance across the plugs is measured by the ignition cassette and based on that information the ecu can figure out which cylinder is on a compression stroke and which is on the exhaust stroke.
DIS however, is not new. waste-spark ignition has been around for a while, but the compression-sense of the L61 is pretty unique.
annoying and troublesome, but unique.
Wait, I thought that's the point of the crank sensor.... to tell the ICM and the ECU the crank angle. Which as an end result would let it know what cylinder is where. I've never understood why the crank sensor is supposed to indicate what cylinder is where, but the cam sensor was necessary for SFI.
I was under the assumption that the point of SFI was to open the injector when the intake valve is open. This gets you a bit better fuel economy and such, but when you go putting your foot down you run into the problem of your injectors running at 80% duty cycle and the intake valve only being open 30% of the time which means that you have to spray fuel while the intake valve is closed. That's why SFI is only SFI under certain conditions.
jcavi - there's 3 kinds of port injection. Sequential, Bank, and Batch. I can set my ECU to run either Bank or Batch. Most older MPFI cars are running Batch fire, not Bank fire. Batch fire is all injectors at the same time (in my case all 6 at once). Bank fire would be sets of them (in my case this would be 3 and 3). I'm not exactly sure which is better when it comes down to fuel economy and power.
SHOoff wrote:DaFlyinSkwirl (PJ) - APU wrote:DIS stands for distributorless ignition system, NOT direct ignition system
and jcavi's discovery is old news, but still interesting stuff.
the eco in the Jbody (L61) has no cam sensor so there's no real way for the ecu to directly know which cylinder on TDC is firing a power stroke, and which one is on the exhaust stroke. it uses whats called "compression sense" whereas the resistance across the plugs is measured by the ignition cassette and based on that information the ecu can figure out which cylinder is on a compression stroke and which is on the exhaust stroke.
DIS however, is not new. waste-spark ignition has been around for a while, but the compression-sense of the L61 is pretty unique.
annoying and troublesome, but unique.
Wait, I thought that's the point of the crank sensor.... to tell the ICM and the ECU the crank angle. Which as an end result would let it know what cylinder is where. I've never understood why the crank sensor is supposed to indicate what cylinder is where, but the cam sensor was necessary for SFI.
because the crank position only tells the computer which 2 cylinders are up, and which 2 cylinders are down. It can't determine which cylinder is the firing cylinder with this information alone.
the cam sensor then tells the computer where the cam is, and therefore which exhaust valves are open. thereby with these two variables, it will know which cylinder needs to fire.
injection is another story altogether, dependent on which mode you're in.. single or batch fire. Most EFI systems use batch fire from a certain RPM and higher
SHOoff wrote: I was under the assumption that the point of SFI was to open the injector when the intake valve is open. This gets you a bit better fuel economy and such, but when you go putting your foot down you run into the problem of your injectors running at 80% duty cycle and the intake valve only being open 30% of the time which means that you have to spray fuel while the intake valve is closed.
That is a mistaken idea.
In cases where the pulsewidth is short enough to actually fit within the time the valve is open, it still isn't done.
Injectors are timed ot open when the valve is closed.
Spraying into an open valve actually cause far worse economy and drivability and emission levels. Probably because the fuel isn't fully attomized and mixed with the air before it is compressed.
Quote:
That's why SFI is only SFI under certain conditions.
Most OEM SFI is SFI all the time unless there is a failure or the commanded pulse width is too short to open consistantly.
Aftermarket, many cheaper standalones go into batch fire at higher rpm only because the processor is too slow to do the necessary calculations for all the individual cylinders at higher rpm. So it switches to batch and cuts the math load in half or more.
sig not found
Quote:
Most OEM SFI is SFI all the time unless there is a failure or the commanded pulse width is too short to open consistantly.
PW too short to open consistently with sequential is much to short to work with synchronous batch fire injection. PW must be decreased by 1/ (# injectors firing at any time). Have you seen OBDII pcm's capable of something similar to "quasi-asynchronous" mode in some OBDI ecm's? Injectors firing at 2x commanded PW 1/2 as many times? Still... seems like the math doesn't work out if switch to batch fire is part of the game. Love to hear more.
For others, the 7 notch crank sensor does not provide ecm / pcm with true knowledge about crank position. There is not enough resolution nor enough signals. PCM measures time between notches, uses number of cylinders, and calculates rpm. Then PCM calculates time required for crank to turn desired number of degrees before delivering trigger pulse to coil or ICM, causing spark to occur before expected next TDC. LSx engines, latest 3800s, and Northstar engines have much better crank position information.
-->Slow
protomec wrote:SHOoff wrote: I was under the assumption that the point of SFI was to open the injector when the intake valve is open. This gets you a bit better fuel economy and such, but when you go putting your foot down you run into the problem of your injectors running at 80% duty cycle and the intake valve only being open 30% of the time which means that you have to spray fuel while the intake valve is closed.
That is a mistaken idea.
In cases where the pulsewidth is short enough to actually fit within the time the valve is open, it still isn't done.
Injectors are timed ot open when the valve is closed.
Spraying into an open valve actually cause far worse economy and drivability and emission levels. Probably because the fuel isn't fully attomized and mixed with the air before it is compressed.
Quote:
That's why SFI is only SFI under certain conditions.
Most OEM SFI is SFI all the time unless there is a failure or the commanded pulse width is too short to open consistantly.
Aftermarket, many cheaper standalones go into batch fire at higher rpm only because the processor is too slow to do the necessary calculations for all the individual cylinders at higher rpm. So it switches to batch and cuts the math load in half or more.
thats why ares switch out.
I dont know the parameters but it switches to bank after a certain rpm.
jcavi - there's 3 kinds of port injection. Sequential, Bank, and Batch. I can set my ECU to run either Bank or Batch. Most older MPFI cars are running Batch fire, not Bank fire. Batch fire is all injectors at the same time (in my case all 6 at once). Bank fire would be sets of them (in my case this would be 3 and 3). I'm not exactly sure which is better when it comes down to fuel economy and power.
guarnteed bank batch fires every single piston event. thats an assload of uneventful fuel squirting.
I know its a calculated spray to be set to a perfect amount, but I cant imgane it being any where near as efficient as bank.
"
boobs now with Riboflabin"
slowolej wrote:Quote:
Most OEM SFI is SFI all the time unless there is a failure or the commanded pulse width is too short to open consistantly.
PW too short to open consistently with sequential is much to short to work with synchronous batch fire injection. PW must be decreased by 1/ (# injectors firing at any time).
I didn't intend to imply it switched to batch when pulsewidth was too short, but rather that it dropped sequential, synchronized injection.
sig not found
