I'd really like to see 19pph 305 injectors but if you have numbers for any of them I'd be interested (please list year/engine if you know what they came out of).

I just disabled EGO correction, set the timing static, disabled closed loop idle control... in my megasquirt and set it to 4 injector pulses and switched between simultaneous and alternating pulses (so one way I should have double the injection events with half the plusewidth) and tinkered with the dead time number till I had the same AFR both ways, or as close as I could get it.

I wasn't able to get as steady an idle with 4/simultaneous, but got it close, and was able to get it to get the correct response at higher and lower dead time settings, so I'm tempted to believe what I got, except the number seems out of range of what I expected- .980ms.

 

come on, no one at all???

 

have you looked at gearhead.com? I suspect they may be there.

 

Huh? That's an ATV site..., am I looking in the wrong place?

 

Not entirely sure what you are asking.
A GM TPI ECM works in double fire batch mode. That is, it sprays all 8 injectors at once every crankshaft revolution. Comparing to your MS terms, it is sprarying all 8 injectors every 4 distributor reference pulses (DRPs).

There is an open, on, close, off time.
open/close is usually around 1ms for each.
open is spray time, depends on engine buy usually 2 to 4ms
off time is the rest of the crankshaft revolution time (60/600rpm) = 100ms

What is "dead time"? Is this the open/close time? If so, a GM 305ci bin has a table of open/close times vs battery voltage.

 

I thought it was 4 all on one side 1357 every crank rotation meaning bank to bank?

Please clarify as I am going port fuel next week adnd this is good to know.

 

GM TPI '165 and 730' ECM spray all 8 injectors every crank rev which is also 4 DRPs.

An ECM like the '427 can do bank to bank firing. I run it using port fuel 8 saturated injectors, but it requires hardware and code changes to do it. This also allows a poor man's fuel trim so that you can trim 4 injectors relative to other 4. Think Holley Stealth ram and LT1 intake with front / rear distribution differences.

 

Thank you for clarifying.
I wil use EBL/PortMod soon.
this is off dynamicefi site:
"With the Port Mod the TBI ECM firing rate is:

•Port on a 6-cylinder, the injectors will fire once per engine revolution.
•Port on a 8-cylinder, the injectors will fire once per engine revolution."

 

You have me confused, "8 injectors every 4 distributor reference pulses" that implies 2 distributor refrence pulses per revolution. I don't think that's correct.

As far as I know under normal conditions our factory GM setups fire all the injectors at one time, once per revolution (MS calls this 2 times Simultaneous), except under certain enrichment modes where it will fire more often. I found that my MS setup runs best firing them what MS calls 4x Alternating, where it fires the injectors bank to bank 4x per power stroke.

What I'm asking about is injector dead time, a factor used in most modern injection setups (I don't think I've ever seen it in a GM ecm but it might be there, honestly, the GM stuff frustrates the hell out of me and I'm doing my best to forget all my forays into chip burning/tinkering for the factory setups). It's the time in ms that it takes the injector to open after it gets a signal to open. Essentially, a time where the ecm thinks that the injector is open but it isn't flowing yet (or at least not at full flow). You can test it empirically (take the injectors out of the manifold spray them and measure the flow for set time periods and when you graph them you'll get a line that does not cross at 0, that ends up being the dead time).

The reason its a big deal is that as the pulsewidth gets shorter that dead time where you're not getting fuel flow is a larger percentage of you open time, introducing error. IF it was just up to the VE tables to get the fuel delivery correct that would be irrelevant, but when you start introducing % based corrections like AE, PE, WUE... that error ends up greater at lower pulsewidths which often gives you an unsteady idle and other tuning weirdness that you can chase forever.

FWIW, MS and others also have a battery voltage correction for it, since the battery voltage getting to the injector drivers will affect how long that dead time is, so things like your fans kicking on could have a big affect on your idle, even if you compensate for the additional load on the alternator.

 

Here are the values from an '89 LB9, calibration APYS. Which would be 19#/hr Multec injectors. The left table is the offset versus battery voltage. The right table is the short PW injector offset. Both table values are added to the current injector PW, short PW first.

Note: these values are only good for the listed injectors, none other.

RBob.

 

huh, rbob, can you post pics of the graphs or export the numbers and send them to me? Thanks

If I try to guess where it would be at 13.8VDC and if I'm understanding that chart correctly it might be closer than I thought to the number that I came up with (which is for 13.8VDC with a correction factor for voltage up or down from there)

 

RBob.

 

dont some V8 ATV's use 19 lbs injectors?

gearhead-efi.com

 

Depends on what an "engine revolution" is defined as.
Most would define it as 2 crank revolutions on a V8.

Idle/low rpm typically degrades when the injectors are fired less often.

 

There are 4 DRPs per "crank rev" - 4 cyls have fired
There are 8 DRPs per "engine rev" - all 8 cyls have fired

Defining Power Stroke to be two crank revs.
MS defines req_fuel for 1 power stroke
MS defines alternating as firing each bank. Fire injecters twice per power stroke.......fire 4 injectors for a bank each crank rev.
MS defines 2x mode as firing injectors twice per power stroke.

If you configured 4x alternating:
Alternating is bank to bank
4x is 4 squirts per power stroke (2 squirts per crank rev)
Would look like this:
Crank degrees 0 -> Bank 1 inject
Crank degrees 180 -> Bank 2 inject
Crank degrees 360 -> Bank 1 inject
Crank degrees 540 -> Bank 2 inject

In your case: In one power stroke, 4 injectors were fired 4 times which is same "fuel squirted" as firing 8 injectors 2 times, which is same as firing 8 injectors per crank rev.
GM TPI Batch fire which is MS terms would be 2x/simultaneous. That is 8 injectors fire per crank rev, or 8 injectors fire twice per power stroke.

Overall, you are like GM Batch fire except for you do Bank to Bank firing. Which I recommend you make Front/Rear banks instead of Left/Right banks. What intake are you running?


First post estimated open/close time for a saturated injector at 1ms.
GM ECM does battery and "short pulsewidth" correction tables. Sounds like MS does the same.


Yes, that is why the correction tables are there.

Yes, that is why the correction tables are there.
GM also did spark correction in some ECMs based on idle variation.
MS is a stripped down version of a GM ECM in terms of both hardware, software, and error functionality.