LBSZ28BLOWN

I'm running the Bosh 27# injectors rated @ 36psi ( I have my Fuel pressure set at 50psi) in my 92 Z28
7730 ecm.
My question is at what pws do the Injectors max out, is it
12.8 pws @ 5500 rpm or what.
I'm seeing 13.8 pws
Thanks

funstick

at that rpm upto 15msec or so should be fine. that just about 85% duty cycle.im not gonna do the math.

Tomcat

The maths for duty cycle is
BPW*RPM/1200 for sequential
or BPW*RPM/600 for batch fire

so at 6000rpm on a batch fire 165 you have max of 10ms of injection time at 5500 you have about 10.9ms


so at 12.8ms you are way past it - although I havent played with the 730 seriously so it may not be batch fire , if its not batch then you are only at 64%

funstick

unless hes using the doulbe fire method i wouldn't worry about it. doesnt matter if its bacth or sqential or bank to bank one fire per revolotion is still gonna keep the injector open every 8 ref pulses for 12. millsec. weather it be one at a time or all 8 at once.

Grumpy

You're kidding aren't you?.
This doesn't make any sense.

Firing strategy has everything to do with injector on times.
Injector PW is only ignored by those who haven't figured out what it means.

Running injectors static, is just a fool's game. When you actually build an ecm bench and hear and see what happens when an injector APPROACHES static, you'll never do it again.

funstick

sure it does 8 injector fire once every other stroke for a total of 12.1 mill sec.

1 injector fire every 1/4 stroke for 12.1 mill sec. for a toat of 12.1 mill sec.

4 injectors fire every 1 stroke for get this 12.1 mill sec.

so how does that not make sense ? i can undersatdn what your saying about not exceding or having to high a PW ie say 17millsec at 600rpm. but i dont think hes pushing it to hard. also depend on injector style. common grumpy not every car runs at 6000 rpm all day long.then i would worry about inj duty cycle. but the occasional jaunt into the 85-90 duty cycle range is nothing to worry about. now sustained dutys cycles for sure will tear it up. like the daytona 500 yeah running a injector at that duty cycle would most certianly screw things up. and maybe i dont need an ecm bench i have acess to a simutec when i want it.

LBSZ28BLOWN

So from what I'm understanding I should probably raise my fuel pressure and retune my VE tables, and try to get the injectors pws down to 12.1ms @ WOT in P/E mode.

Grumpy

Define what the heck your referencing stroke as meaning.
All the engines under discussion here are 4 strokes and it takes 2 revolutions for all 4 to occur.
If you have some new secret meaning that makes sense to you, well fine, but for the rest of the world you're not making sense.

17 msec of PW at idle?.
That would be the most insane error in injector sizing ever.
85-90 %, please note the percent symbol so as to make the figures meaningful. So who said anything about 85-90% DC as being a bad thing?.

You simulator doesn't show what happens as an injector approaches going static. So you last comment is also meaningless.

Daytona 500 and EFI?.
I haven't seen many F Bodies in any Daytona 500 lately.

funstick

actually bruce the simutech can show current cuosumption and injector firing wave form. also i was making an exapmle with the dayotna 500.

as for injector firing.

2 strokes for all 8 cylinders of a v8 to fire.

fire every injector once every 2 strokes batch
fire the injector 4 at a time every stroke bank to bank
fire 1 injector every 1/4 stroke sequential.

how does this not make sense ?

also i was saying that 12 mill second should be fine at 5500rpm. in a v8 running batch fire. unless the ecm runs in double fire mode halfing the injector pulse width and firing once a stroke.then it would be border line.

Kaiser

I think I see the problem. I believe you misunderstand the way the ECM here works. (Or else I have been reading my Diacom data wrong for a lot of years.)

A 730 ECM fires every injector once every engine revolution. I think you are calling this 'double fire'.

Thus at 5500 rpm there is only 10.9 ms available for each injector 'on' pulse. A 12 ms pulse width would not be 'border line' - it would be impossible.

(BTW, a typical idle pw for this ECM is more like 2-3 ms.)

Is your scanning software telling you something different?

HTH

funstick

im not gonna assume that every ecm is the same. i mostly work with bank to bank fire ecms. its just the way it seems to go. the 730 i beilive is batch fire so basically it fires a pulse every revolution. but that would mean that each cylinder would get a say 12mill PW every revolution. thats actually get this per combustion cycle 24mill of injector pulse wdith. thats alot of fuel. im pretty sure its halfing it. and then spitting it out on a per cylinder firing event based sort of basis.then again i cant read the source code and i dont have a good hack so its hard to say if im wrong or right.
and grumpy this is where simutech comes in handy. i can look at the correlation bewtween the ref pulse and injector firing and even derive the pw as well. its a nice tool. maybe not good at manipulating the input side of things but the best at capturing the output. imagine having a lab scope hooked to every ouput of the ecm at the same time ??

Kaiser

That's very interesting, but the original poster asked about a 730. I expect that is the one he wanted info about.

As previously stated, you cannot do a 12 ms event in 10.9 ms. Multiplying it by 2 doesn't really help - you also can't do a 24 ms event in 21.8 ms.

HTH

funstick

i was using 12mill sec as an example. that why ive stuck with 12mill sec.

LBSZ28BLOWN

Well I'm up to 55psi and it gose to 65psi under boost.
So I retunned the VE tables & the P/E table.
Now at 5500rpm I get 11.2 pms an no Knock retard at all My O2 is .830 to 840 MV the whole run.
Next stop Norwoods Dyno for some real tuning

Grumpy

It takes 32 strokes for 8 cylinders to fire, see it's a 4 stroke engine. So each clyinder takes 4 events for it to fire.

It takes two engine REVOLTUIONS for all 8 cylinders to fire.

And to not get too confusing this is 3rd Gen v8 specific.

anesthes

iTrader: (12)

Dont mean to bring this back form the dead, but a stupid question here.

I figured at:

5000 rpm, you get 12.05 ms of pulse time
5500 rpm, 10.9
6000 rpm 10.0

To fire each injector. Now I'm seeing stuff like 4.65 @ 4400, 3.54 at 5000 rpm (not under boost!) at like a 134blm. I'm trying to
determine if I maxed out the injector at those rpms, or if I'm not even close.. 3.54 @ 5000 seems like 1/3 duty. Right?

This is 24lbs sv0s at 41ish psi..

I'm trying to understand whats going on here.

doc

Assuming higher engine load at 5000rpm vs 4400rpm from your last post, I would think that the PW at 5000rpm would be longer than at 4400rpm, but you show the reverse, so then I figure that at 5000rpm you must have been under less engine load.

Your PWs are OK. At 5000rpm, you reach 100% duty at 12msec, and at 4000rpm you reach 100% duty cycle at 15msec, and at 4400rpm, rou reach 100% duty cycle at 13.63msec. You are well short of those PWs.

Looking at the above posts from back when, it is obvious that batch fire EFI systems are poor compared to sequential fire systems like that in my '99 SS. I have 20msec at 6000rpm to reach 100% duty cycle on my '99 LS1 car!!!!! And only 10msec on my '87!!!!!!!!!!!!! This is the main reason why fuel injectors for our 3rd gen cars must be sized much larger.

Grumpy

SEFI is one shot for 2 engine revolutions.
Batch in one shot in 1 revolution.
6 to one half dozen of another.

The key is that while DCs are interesting the key is actually making sure the injector has enough time to completely shut off before firing again. So while 90% is getting critical at 5K at 6K you right at the limit time wise for the injector to still be functioning accurately.

FWIW, I have better and better luck with large injectors and short pulse widths. And with the bigger injectors running in batch fire mode. The advantage of firing the injector twice as often helps with atomization.

funstick

this is totally dependant on the number of winding and dissapation time involved.

im gonna have to go back and recalc these engine speeds. 10msec seems a bit short for 6000rpm. you can have a cumbustion event occuring every 10msec but a complete revolution ?? i dunno. im not questioning show me the math on this one.

JPrevost

RPM stands for revolutions per minute.
It takes 2 revolutions before an intake "filling" stroke. It also takes 2 revolutions/rotations of the crank before a power stroke, exhaust, and compression stroke, hence the name 4 stroke.
So 6000rpm will have 3000 intake strokes a minute (a stroke remember is only half of a revolution). With that being said, 6000rpm is 100 revolutions per second. 100 revolutions per second gives you .01 seconds per revolution which is 10ms. That gives you 10ms to fire the injector if the injector is firing once every revolution. It's got 20ms if it's fired once every intake stroke (2 revolutions). So yeah, 20ms is the time allocated for the injector to open, close completely, and then do it all over again. Going static would mean the injector has no time to close and just stays open but "burps" when it's given the signal to close and reopen but never going closed. It's just getting abused at that point. If I'm wrong somewhere don't be shy to correct.
Just thought I'd add that at 6000rpm, 10ms would account for an intake and compression stroke, then another 10ms goes by and the combustion and exhaust stroke have taken place. So if the injector fired right when the intake valve closed it's got 20ms before the valve opens again. That's just one way to think of it.
Doc has it right. Sequential has more time to fire the injectors than thirdgen batch fire. Batch fires the injectors all at once (duh) once a revolution so it fires twice a revolution while sequential fires once every intake valve opening (once every other revolution). It's easy to get lost in the math. I had a hard time with piston and crank accelerations. Doing it by hand was a big waste of time when I had maple 7 that could do all the math for me in a matter of minutes . Computers are my crutch

anesthes

iTrader: (12)

Right on!

Well looking more at my log, at higher rpms, and under boost my PW was kinda low. I think the fact that fuel pressure raises with boost because of the FMU is why.. If I ripped out the fmu, and rescalled my ve tables to read 2-baro, I think my injectors wouldn't be able to deal with it at a static fuel pressure.

And even if they could at the high pressure the FMU goes to, I don't think I could get the idle pw short enough.. Hrmmm..

-- Joe

Grumpy

I run 55 PPH injectors on my lil v6.
1.7 msec PWs at idle in batch fire mode.
Actually that 1.7 is including a 7th injector that is 36 PPH.

Single batch fire at 6K RPM has 10 ms for each injector event.

funstick

ok but heres my question. is it firing asynchrnos or quasi synchronos. if you fire the injector for 10 msec every 8 ref pulse yes youll be out of fuel really quick. if you fire the injectors every 4ref pulses for 10 msec youll have plenty of fuel.

8 ref pulses occure every 2 strokes. 4 ref pulses happen every stroke.

if you fire the injectors for 10 msec every ref pulse youll wind up with alot of fuel.

doc

Grumpy,

I like what you said in your last two posts. I have decided that I need my 24#/hr SVO injectors not in my '87 but move them to my '99 where I am 100% duty at 6000rpm, that is 20msec of PW at 6000rpm.

What you said about large injectors and short PWs for batch fire cars hit home with me. I think that when I move the 24s over to my '99, I should install 30#/hr SVOs in my '87. I have heard that injectors dont do well at less than 1.5msec (at idle stuff). So I am concerned about that one point. I will do a scan at idle on the curent 24s to check the PW at idle.

funstick, This is not meant to be a critisicm, and I still want to get together soon with you, but I dont understand your posts. I am sorry for saying that, but you have had many great posts on other topics, you are a very knowledgable person, so I cant figure out where you are coming from on this topic. Anyway, be cool. Dont be mad at me now.

funstick

ok heres a thought for me to share.

various injector firing schmes will create different fuel ouputs states.

asycnhronos will create a injector event for each ref pulse.
quasi synchronos will create and injector event every 8 ref pulses
quasi asycnchronos should in thoery create and injector event every 4 ref pulses.

youve gotta ask you self why does this matter ??

ok stop think about the injector timing relavtive to the intake valve. think about it only in terms or Fuel needed and time to deliver it. with 8 24pph injectors you can at 100% dc come up with 192 pph. this isnt a safe amount of injector duty cycle so lets cut back to say 85% at a bsfc of 0.5 per hp per hr.

we now have a safe delivery of 163pph. so whats the output of the engine. ??

the real question is which injector firing schme buys time to deliver fuel. you would think that youve only got 10msec to fire the injector but your wrong.

you have 20 msec. if you fire the ecm every 4 ref pulses. even thought each revolution take 10 msce @6000rpm only half the cylinders are firing. so if you inject 10 msec of fuel every 4 ref pulses you end up with 20 msec every 2 complete strokes.

its not that complicated. it just depeneds on your firing strategy.

what im saying is if you put 10msec worth of fuel on a closed valve then you have 10 msec when it opens to shove fuel in.

a73camaro

Something to keep in mind, all injectors have a transitional time to consider. It is a couple of milliseconds. The open/close and vise versa is not instantaneous.

I did a test with a saturation injector at 6000 RPM or 10 ms period. I discovered the injector peaks its flow at 7.5 ms. That is where one should not exceed 80% of the period of an engine revolution.

On the other end of the spectrum, I discovered that one should not go lower than 10% of the period of one revolution, but that was at 6000 RPM.

a73camaro

Here is the graph, the injector was pulsed for about 5 seconds.

funstick

to put the 85% duty cycle itno light on my last post youd have 17msec for every 2 revolutions to feed fuel. thats plenty of time.,

JAY87GTA

Ok, I read all the way through this. The way I understand it is with the 730 ECM I have 10ms at 6K to get my fuel in the way it is stock. Am I correct? If so, how or can I do like funstick is suggesting and get it to fire for the 20ms?

funstick

use quasisynchronos fueling mode above xxxxrpm.

doc

OK funstick, I know now you are bull$hitting us.

Jay, at 6000rpm the crank makes one rev in 10msec, just work out the math. Since all injectors are fired (batch mode) once per one rev of the crank, you are at 100% duty cycle if the PW is 10msec. There is no way around it. Its in the math.

funstick

hmm but there 20 msec between combustion events and i did outline how to get the longer PW at the high rpm. no body payed attention.

and doc when you wanna meet up ?

Grim Reaper

Combustion occurs every second rev, therefore 20ms/2revs in quasi-mode = 10ms/rev.

No difference inthe amount of fuel delivered.

anesthes

iTrader: (12)

This is exactly why my 24pph sv0 injectors work great with my FMU, but if I run computer controlled boost fueling, with a static fuel pressure its gonna be iffy .

FMU right now boosts the fuel to the mac the pump will do under boost.. I'd think that even if I set my WOT pressure that high, i'd run pig rich at idle cuz the pw couldn't get low enough, and if I mad idle happy I bet wot under boost would be lean.

Guess I need to do some more math on a spreadsheet to figure out what the right injectors are for my combo if I rip out the FMU..



-- Joe

Grumpy

To date all I've seen on GM's is synch and asynch (again 3rd Gen specific here, as is what's mandated by charter).

JAY87GTA

Ok, that's what I thought, but then we get these guys that are light years ahead of me talking about all these quasi stuff. Just wanted to make sure I had made the proper ASSumption. Sometimes that bites me.

Grumpy

I haven't seen anthing really nail it as a couple msecs.. Maybe 1 but that would be a push, most of the corrections Ive seen stop at about .8 msec. Just a FWIW.

And 80% DC is when fuel rail pulsations are not an issue. Figure in recovery time, and large injectors, and you want to run just about as big of injector that you can that will have a decent idea. At least that's what I've found in my car, and I've had 4 different sizes in testing this idea. Plus I've added various sized 7th injectors. If you ever have the opportunity to run the same ecm in SEFI vs Batch the ratio isn't 2:1 time wise for PWs. This clearly shows the differences in PW as function of opening and closing delays.

Grumpy

There is only synch and asynch.
Synch means in synch with the ignition pulses, and asynch meaning they fire every 12.5 msecs. That's all there is, in this world. If it's really difficult to follow it just might be BS. Asking guestions will dispell fact from fiction.

Grumpy

Which ecm in the 3rd Gens gives you every other crank revolution injector firing (other then the TTA)?.

Grim Reaper

That is the lowest PW (.8) I generally see with my car just before DFCO occurs. The only reading I get less than .8 is then 0.

Grumpy

Look at you DC.
Calculate you fuel flow at the elevated pressure.
Figure out your target HP.

Remember a street non N/A application and your not gonna get .5 BSFC.

Then toss in a little cushion to actually get some room for AE.

Then look at where your at PW wise at idle now.

And figure on maybe having to do injectors twice. Sometimes just getting some used ones, is cost effective for testing.

Grumpy

Peak and Holds can drop down to sligtly over 1 msec.
Some saturated ones, will approach that.

Depending on what I'm doing I run 1.7-2.2 with 55 PPHs, at 850 ish RPM idle.

Grumpy

Actual snippet from data log.

Run Mph Rpm KPaTps CtsF MatF Spkf PW WB IAC PE AE PN
15:26 0 900 42 0 182 136 15.1 1.53 0.0 19 N N Y
15:26 0 875 42 0 182 136 15.1 1.53 0.0 19 N N Y
15:26 0 900 42 0 182 136 13.7 1.53 0.0 19 N N Y
15:26 0 875 42 0 182 136 15.1 1.53 0.0 19 N N Y
15:26 0 900 42 0 182 136 13.7 1.53 0.0 19 N N Y
15:26 0 900 42 0 182 136 13.7 1.53 0.0 19 N N Y

60 PPH effective injector rate.
ie 6x55s and a 36PPH on a 231.
Batch fire mode.

funstick

your still missing the point glen. its 20 msec to fuel each cylinder. the fuel can sit on the valve and it may help vaporazation.

this is true. but a simple code change can fix that problem. youd idealy want to fire every 4-8 ref pulses at high rpm. thats called bying time. theres more recovery and disspation time. as i stated even with a 20 msec window youd want a max of 17 msec. plenty of time for the injector to close open and dissapate current. check that out with your scope.

my replys have been based on the fact that every one has been overlooking the simple fact that you have 20 msec to fuel each cylinder. regardless of batch or sequential. the problem with the current injector sizing models is that its bascially treating the engien as if it were a 2stroke.

ive heard one of the new haltech units is capable of a stable .6msec p/h with luca disc injectors. i have no evidence to back it up but it does seem feasable.


glad you brought that up. it will fire every 12.5 msec. but itll also fire once every 8 ref pulses. depends on which time frame is shorteer. the point of quasi synchronos at idle is to help create fewer but large injector firing events to help stabalize the injectors.

at 6000rpm however there not going to be a 12.5 msec delay. that only obeserved at idle.

Grim Reaper

Just to make sure we are talking about the same thing, we are talking about stock bins for TPI F-bodies right?

It takes 2 revs for every combustion cycle to occur for each cylinder on a 4 stroke engine. If the injectors are open for 20 ms over the two revs need to complete each cycle, it will result in the same amount of fuel being delivered as a reported Batch Fire PW of 10ms (which is ms/1rev).

20ms per 2 revs = 10 ms per 1 rev. They give identical fuel.

JPrevost

Is this assuming the batchfire injectors are going static? Because an injector firing for 20ms vs one that fires twice for 10ms isn't going to flow the same amount of fuel. The one that's open longest will flow more. I'm sure you knew this but what was implied confused me. I guess if the injectors go static and never close then yes, they will flow identical rates.

Grim Reaper

It actually doesn't matter. But yes the 100% DC definitely proves that you receive no more fuel from "quasi-mode"; which was the initial point of using "quasi-mode" in the post.

funstick

not true at all. you not taking into account the flow loses from the pintle seating and unseating. if i had to subtract 1msec on each end of the injector pulse from 10 msec i wind up with 8 msec of useful injector firing.

if i have 20 msec of injecto cycle and i subtarct 1 msec from each end i wind up with 18 msec of usefule work. but these are working with 100% dc numbers here. you gte 2.25 times as much time to spray fuel. check it out.

SATURN5

I bit of error here.. firing ONCE for 20 msec vs firing TWICE for 10 msec each.

sub one from each end..

20 msec PW = 18 msec spray
2 x 10 msec = 16 msec spray.. not 8.

HighHopes85

You sure about this?

Say an injector is commanded for a 10 ms pulse width.

For an example, we'll give it a 0.5ms rise and fall time.

It rises at time 0 and is fully open at 0.5ms.

It is commanded closed at 10ms and fully closes by 10.5ms.

That's the way I see things. It has an effective pulsewidth of 10 ms, even though it too 1 full ms rise and fall. It doesn't know to close early, as what you mentioned suggests. It can't close until it gets told to.