Can't get my car to idle leaner then 11.5:1 AFR????

[nitrorocket]

I have an ls1 with a custm O.S. and a 2002 reflash( 2000 pcm) Problem seems to be that I cannot get the idle range AFR to go leaner then about 11.5:1. I keep auto tuning and everytime I do the paste and multipy to lower the VE on the overly rich cells, it multiply's them by .80. The numbers in my VE chart are getting really low, around 30. It eventually starts to bog real bad with throttle tip in. The leaner I make it(even though it does'nt get any leaner) it make the surging idle problem I have even worse. I run a 96 lb injectors. Have any ideas? Should I enable the idle STFT?? What would this do exactly?

[caver]

Did you scale the injectors?

[SS2win]

mine did something similiar. turned out the o2 was not calibrated properly. once i carefully recalibrated it the BEN's settled at 0.98-1.01.

[nitrorocket]

I have the injectors rescaled. The A/F ratio is perfect from about 3/8 throttle and up. It runs at a nice 13.5-14.7:1 AFR. the problem is below that. Part throttle and idle will not lean out.

Could it be an adjustment I need to make with pulse width or injector dwell? It's almost like below a certain point they will not aloow any less fuel to flow. I am running low imp 96 lb/hr injectors with a Acceleronics driver. I run 45 psi constant at the regulator.

My actual AFR PID shows everything is all good except at low throttle the map will drastically drop from 13-14 afr right to around 11-12?

Any ideas?? The car is tuned almost perfect except for this area. Not a big deal if all I want to do is go fast, but I do street driving mostly.

[nitrorocket]

My twin turbo ls1 runs 96 lb injectors with an acceleronics injector driver. The problem is rich idle. On my data log AFR map, anything below 1200 rpm between 50 and 75 kpa is anout 11-11.5:1 AFR. the next cells around this area are 13-14:1 afr. Why would just the area drop off so instantly from running correct to dead rich?

I have 96 lb injectors.
I changed my MIN pulse width setting to .75 ms. Idle is about 1.3-1.4 ms in this "rich" area.
I have lowered the ve map so far to try to lower idle AFR, that the VEnumbers ar e now into the teens(about 60% leaner then stock) Lowering VE seems to do absolutly nothing?

[bink]

{B3607} Open Loop Lean Limit (Idle) - Take a look at this value if you are Open Loop SD.
FWIW.

Cheers,
joel

[EFIGUY]

Guys,

I'm pretty sure the problem is in the injector's dead time. The relatively large injectors you are using require more time (even with the higher current) to accurately control fueling. The battery voltage comps and injector dead comps, and rise time comps that the ECU does is centered around the stock saturated injectors.

So, even though the CS453 injector drivers used in the Acceleronics units can handle the current draw of the 96'ers, you still end up with totally inaccurate settings for controlling those large injectors at small opening times.


I built my own P&H drivers and was considering marketing them, until I saw the Acceleronics thing, but nevertheless this was one problem we ran into then as well....

Now I'm thinking that I could use an on-board microprocessor to intercept the injection pulse width command from the ECU and modify the dead time and battery comps based on the type of injectors used before sending the signal to the driver. It could have a nice little GUI to let the user pick the injector type from a menu and alter the program! Thanks for the idea!



I'll bet our buddies Ross and Paul can find the location in the PCM for the table that controls the injector comps and give us something more to play around with???????

[GMPX]

I guess this is one limitation working with a factory computer.
There is a resonable amount of tables for the injectors.
There has been a few guys out here running low imp injectors via a driver box that got it all sorted with what is currently in EFILive, but, they are a business so we should not expect them to share too much info.

Ben, I know you can picture in your head injector flow figures, we do have a table that is a limiter for the minimum injector time, most factory tunes this is set to about 1.28mS, could we do the maths with the injectors and fuel pressure to see if that is low enough?. Would altering this help?. Or is the real problem the P&H drivers?

Cheers,
Ross

[foff667]

this could possibly help http://www.ls1tech.com/forums/showthread.php?t=306316&highlight=injectors

[nitrorocket]

I have the prgram setu to the custom os and auto tune tutorial's.
My (B3607) value is 15.395 in the column.

It is wierd how I have the minimum injector pulse time adjusted to .75 ms, but my logging shows they are only going down to 1.3-1.4??? I run at 45 psi fuel pressure.

There has to be something to I can change to make the pcm give less fuel at idle!?!?

[Ira]

It's quite possible that there is a minimum amount of time the injectors are allowed to open. There is no reason to do it, but there is also no reason not to do it.

Ira

[SS2win]

Do you have access to an ocilliscope? it may help to monitor the pulses while you play with the tables to get it right.

[EFIGUY]

Ben, I know you can picture in your head injector flow figures, we do have a table that is a limiter for the minimum injector time, most factory tunes this is set to about 1.28mS, could we do the maths with the injectors and fuel pressure to see if that is low enough?. Would altering this help?. Or is the real problem the P&H drivers?

Cheers,
Ross

The real problem isn't the drivers, its the injectors, and the way we control them.

Basically what is happening is that the 96 lb'ers are like a big heavy truck, and the stock ones are like a VW beetle. The engine in the ECU that drives the beetle (stock PCM) was programmed to work well with the VW's, but now that we put that engine in the truck(big injectors), we can't get going from a stop very easily...then once we get going, we are using the brakes from the beetle to try and stop again...

In essence, we lose control of the actual movement of the pintle inside the injector. Most stand alone ECU's compensate for this by regulating the current flow to the injector at opening and closing time. They do this by using a PWM strategy on the trigger side of the driver.

With the factory PCM, we have more of an analog type signal to the injector driver....on when its time to turn on, and off when its time to turn off. This can normally be compensated for well enough by simply increasing the PW demand over the theoretical amount required for the flow you want by the amount of "lag" or dead time the injector has.

Example: you need 2.5 mSec of flow out of the injector, but it has .8 mSec dead time, so the command PW equals = 3.3 mSec

The real problem though in controlling the injector is how well the oscillation of voltage (created when you open the circuit) is dealt with.

Most ECU's have a simple zener diode (typically about 5-10 watts and 35-40 V) to avalanche this "kickback" voltage to ground.

Again though because we can control the current through the circuit with the aftermarket ECU's you can custom tailor the control strategy for each individual type or part number of injector to get the control you want.

When you just take the signal from a saturated driver, and apply it to the signal side of a P&H driver, you do gain the ability to sink the current to the injector without letting all the smoke out of your PCM, but you don't necessarliy gain any more accurate control over the injector itself.

I am sure there are tables somewhere inside the PCM to deal with this, but they probably are not obvious because the PCM was designed only for one kind of injector, so there wasn't much need to make it adjustable.

Now that I think about it, we might be able to play around with tables:
B4006 and B4005 to get a useable solution????

Basically if anyone could do some testing and find out if changing these settings helps, hurts or makes no change, that would be great!

-ben

[joecar]

Most ECU's have a simple zener diode (typically about 5-10 watts and 35-40 V) to avalanche this "kickback" voltage to ground.

"Freewheeling diode": soaks up the injector coil's inductive back emf when the driver circuit opens to protect the driver transistor from popping.

Cheers
Joe

[nitrorocket]

There is a injector current table. I am going to log this table and then lower the current and see what happens. There has to be a way to control this injector. My afr at idle is also a little erratic. When I figure out out. I will write something saying how I did it.
Geez, does'nt everyone run 96 lb injectors?? I feel like I am in completely unknown territory!

[Delco]

There is a injector current table. I am going to log this table and then lower the current and see what happens. There has to be a way to control this injector. My afr at idle is also a little erratic. When I figure out out. I will write something saying how I did it.
Geez, does'nt everyone run 96 lb injectors?? I feel like I am in completely unknown territory!

Yes you are in unkown territory without a map and blind :lol:

I have quite a few cars running the big injectors and the same box , but they are combo's that can make use of the big injectors down low , on a stock cube motor you ar always going to have problems controlling the injectors at very low flow rates.

I have got around it with tuning but I would not say its 100% to my satisfaction but thats why I get paid to spend countless hours tuning monsters.