Permanently Running Open Loop Cold Start AFR

[LSXtreme]

I had a quick question. My AFR is really close to what I am commanding all throughout my VE so I am happy with it. My only problem is on cold start up it is really lean until about 120*ish is when it falls right to what I am commanding. I know this is about the temp at which Closed Loop takes over so in order to add fuel at the colder temps without messing with my VE table where is this done at? Is it the Afterstart Enrichment Initial Adder VS ECT? Because even though I am commanding in OLFA a 12.8 -13.3 in 40* temps the AFR is up in the 15s on the wideband. Then the slowly meet at 14.12 commanded AFR at 160*.

Thanks

[GHuggins]

Fuel - Open Loop EQ ratio. Try multiplying out the areas where it's running leaner in the temp zones by 1.01 or 1.02. That's all it usually takes to make some pretty big differences. Otherwise is there a reason for a full time open loop tune? If it's just for the cam, you can do an open loop at idle to a 1200 to 1000rpm only tune. Actually not that hard to do. Otherwise your A/F is staying pretty close to desired, but you O2's are saying rich...

Hope this helps.

[LSXtreme]

Just to confirm, the OLFA multiplies against the VE table right? So yes I will just add more fuel in there then. That was easy! No wonder no one has been responding because I was having a brain fart.

The reason i am running Openloop is because I am sick of the narrowbands throwing off my fueling because of long tubes. I know There are some adjustments that can be made to the O2 readings but I need to get the VE dialed in anyway. Since I might want to run open header at the track and 02s at the end of the collector do not work well because they are too close to the outside air. As I tune the VE in openloop I keep making small changes to the O2s to try to get them to start oscillation at the commanded fuel I have set. I am pretty close because they start switching under certain airflow modes. Slowly but surely I will get the O2s to match and I will go back to closed loop.

[GHuggins]

Well as for O2's and long tubes with NO cats. I finally got to the bottom of what's going on with our ZL1 and it's being fixed as we speak. It was very obvious when I saw it. Talk about brain farts Anyway we are using Kooks, which like most longtubes they put the precat O2 right at the front of the collector pretty much to where they can only read off of two of the tubes. SO no cats = more flow through via scavenging and no back pressure = more exhaust gasses bypassing the O2. Aparently the little bit of exhaust "hang up" created by the cats being in place is just enough for the O2's to read somewhat like they should. We are currently in the process of relocating the precat O2's to just in front of the the cat housing, so they can correctly read the exhaust flow.

As for the O2 proportional and integral settings. They were pretty much just like what I thought and how all other proportional and integral settings are withen the ecm, not how they were attempted to be explained to me. In other words you "add to" the proportional then "subtract from" the integral settings. DO NOT add to the integral unless you want a dang near open loop tune. Increasing the integral "slows" the O2's ability to make fuel changes. To ME if you increase the distance from it's original location, would you not want it to make adjustments quicker!!! IN FACT, just to demonstrate this you could mulitply your integral table by 2. See if it is no longer making long term corrections. Again to me if the O2's are slowly going to under a 100mv, then you need the ecm to make corrections quicker. Atleast that's what I already proved by playing with these numbers to myself... NOW I know I don't fully understand all of this and I will get negative comments for it, but this showed the absolute best results for fuel control.

Sorry for rambling on. Just thought I would help you learn from my experiances playing with these tables.

SO, if your running gutted cats relocate your O2's, then dial in your tables for best results. Otherwise you can run the open loop at idle only setup or open loop all the time. It's up to you.

[LSXtreme]

This is good info thanks for rambling! I understand what you are saying I guess for me I understand what needs to happen but I was never sure what I needed to change to get there. I am glad you got yours figured out because I always get into trouble by changing to many tables at once so I am always chasing my own tail. Right now I have my stoich set to 14.12 and the wideband is reading 14.0-14.3 so its pretty darn close. My narrows stay hovering around 830mv but every now and then when the wideband dips to 13.5ish I get a narrow to dip to 600mv which means I am not far off for them to start switching. I think if I was commanding a 14.7 stoich that they might be switching correctly but I am tuning my fuel for the winter blend here right now. If I change the stoich afr would I have to change the switch points because stoich is not 450mv anymore? or is there a different way to compensate for that? Or I guess a better question would be that will the O2s even start switching if PCM isnt in closed loop? Because I am trying to kill 2 birds with one stone by dialing in my VE in openloop but watch the narrowbands respond to the VE. Then once close I'll switch back to closed loop. But right now in closed loop it will pull a ton of fuel out because it would think its rich from the narrows are reading.

But to recap on your solution was you made the O2s switch faster by adding to the Closed Loop Proportional Airflow and subtracting from the Closed Loop Intergrator Delay Airflow correct? I have no cats and the O2s are right in the collector. But I also have a cam with 235 intake duration that adds to the madness. Because I am sure the overlap inducing air into the exhaust at idle is going to throw the reading off.

Side note. I added more fuel to my OLFA and now at colder temps and its commanding a 10.0afr at 50* and the afr did come down to 12.0ish but they are still off...

[GHuggins]

I hate to throw more into the madness here, but did you change your injection timing. If not start raising your boundaries with that cam. Heck it even shows nice improvements with stock motors... This will both help your fuel consumption and fueling "lean" issues.

Think - but not sure as I don't remember perfectly - That I multiplied Proportional by 1.02 to 1.05? Then multiplied Integral by .25.... WAS able to get O2's to start switching just a touch, where before they just slowly dropped to below 100mv. Waiting on my O2 boss's to show up to 100% fix the issue, which is something you might really want to consider.

You can leave your stoich rich, BUT (depending on year model I beleive) You will need to pull out airflow from your MAF table to keep the ecm from taking away the fuel. EVEN THEN, it's going to run at a 14.6 to 14.7 air fuel ratio because of your O2's... You can counter tune for this - by increasing your O2's normal switch points to around 550 to 575mv's. However - this is only going to get you somewhere around a 14.5, 14.6 air fuel ratio. This is why "I" personally like to set my stoich values somewhere around where I know my O2's will be switching.

Hope this better straightens things out.

[LSXtreme]

No madness! I appreciate covering every aspect because its that one table that is missed that usually bits me. Ok I will try what you recommended first! I do have stock LS1 injectors in it right now to TRY and keep out more factors. But I am kinda at a crossroad because it is a truck 6.0 with an LS1 PCM. So do I start with truck closed loop O2 tables or LS1 tables. Because they are significantly different. (Truck-6.0, intake, heads) (LS1-PCM, O2s, injectors) I have been using the LS1 tables since I am using LS1 electronics but the electronics are calibrated for LS1 cubic inch, heads, intake, manifolds etc. I know I do this to myself, I should just start with a bone stock car with matching components then if I change a cam or headers the changes needed will be easier to diagnose. But whats the fun in that!

[GHuggins]

Hey you can't learn unless you experiment with things like that I would use the truck settings unless they're just completely different tables...

You can change your O2 settings to make it run richer, but it won't run much richer than 14.5. "I" would change stoich back to around 14.5, 14.6ish then tune. Start with injection timing first. Probably want to raise the boundary about 15 - 20%, then maybe even more... Then retune everything...

[LSXtreme]

Ok, I'll run the truck settings, Are you adding 15-20% in the Normal & Makeup tables within the Injection Timing or just adding to the Boundary?

[GHuggins]

I do boundary only. Typically you add to the boundary and it always retards the injection timing. The other two depends on build dates and everything else as to whether or not it retards the timing or not. Gets confusing, but usually if you follow the trend you can kinda tell which it is. Boundary is about the easiest and best way to retard the timing of the injectors.

[LSXtreme]

I am assuming that by retarding the injection timing your not spraying against the back of the valve? Is that correct?

[GHuggins]

Yes

[LSXtreme]

So I changed my stoich back to 14.68. Added about 18% to the Injection boundary which made the engine have a little cleaner free lope to it with the cam. AFR was still pretty close. I added about 10% to the Closed Loop Proportional
Airflow and subtracted about 10% from the Proportional Idle. I forgot to do the Integral but the O2s were really close to switching freely so I decided to enable Closed loop and give it a shot. First this is that the narrows started to oscillate at idle and was idling fine. As soon as I started to drive.... haha well I think I have something way way to sensitive! Check out the scan because the STFTs were oscillating just like the O2s. I think I am pretty close if I know how to fix the sensitivity.

Also, while I have your attention. I can't get it to enter into PE mode for some reason? I never had this problem before. Unless I zeroed out something wrong?



ASX_Camaro_SS_(CL)_v1_(Truck Fueling).hptSTFT Too Sensitive.hpl

[LSXtreme]

I fixed the PE. I had the MAP HYST at 60 instead of 6. Thats prolly why. Just need the STFT sensitivity figured out and I should be all set!

[GHuggins]

So I changed my stoich back to 14.68. Added about 18% to the Injection boundary which made the engine have a little cleaner free lope to it with the cam. AFR was still pretty close. I added about 10% to the Closed Loop Proportional
Airflow and subtracted about 10% from the Proportional Idle. I forgot to do the Integral but the O2s were really close to switching freely so I decided to enable Closed loop and give it a shot. First this is that the narrows started to oscillate at idle and was idling fine. As soon as I started to drive.... haha well I think I have something way way to sensitive! Check out the scan because the STFTs were oscillating just like the O2s. I think I am pretty close if I know how to fix the sensitivity.

Also, while I have your attention. I can't get it to enter into PE mode for some reason? I never had this problem before. Unless I zeroed out something wrong?



ASX_Camaro_SS_(CL)_v1_(Truck Fueling).hptSTFT Too Sensitive.hpl

Finally fired the computer up for the first time this weekend... I would recommend going back closer to stock "maybe increase 2 to 4 percent over stock" on your proportional Open/Closed Loop O2 settings and then maybe multiplying your O2 integral table by .8 or so at a time until it's just right for you. The 10 % increase is a lot for a proportional table. I typically won't change these beyone a 4 or 5 % max addition. The integral table "fine tunes" it I guess is the best way of putting it? Plus the integral settings seem to always play the biggest "best" effects on getting something dialed in just right - if that makes any sense.

I think it's kinda funny you mentioning the zeroed out something wrong. I had one about two maybe three weeks ago where I was driving and datalogging one while resting my arm on the laptop to keep it from bouncing around and somehow managed to click on the tune "minimized the datalogger screen" then set my entire idle airflow integral table to 1's. Without thinking about why my screens had changed, I made some very mi-nute changes to stoich, reflashed, then what's ya know - it wouldn't idle anymore without revving back and forth from 500 to 700ish rpms. Had to compare my previous tune to the one I just loaded before I figured out what had happened... Talk about a moment

Glad to hear things are coming along and you've about got your tune how you want it. Atleast driveability wise anyway right

[LSXtreme]

Finally fired the computer up for the first time this weekend... I would recommend going back closer to stock "maybe increase 2 to 4 percent over stock" on your proportional Open/Closed Loop O2 settings and then maybe multiplying your O2 integral table by .8 or so at a time until it's just right for you. The 10 % increase is a lot for a proportional table. I typically won't change these beyone a 4 or 5 % max addition. The integral table "fine tunes" it I guess is the best way of putting it? Plus the integral settings seem to always play the biggest "best" effects on getting something dialed in just right - if that makes any sense.

I think it's kinda funny you mentioning the zeroed out something wrong. I had one about two maybe three weeks ago where I was driving and datalogging one while resting my arm on the laptop to keep it from bouncing around and somehow managed to click on the tune "minimized the datalogger screen" then set my entire idle airflow integral table to 1's. Without thinking about why my screens had changed, I made some very mi-nute changes to stoich, reflashed, then what's ya know - it wouldn't idle anymore without revving back and forth from 500 to 700ish rpms. Had to compare my previous tune to the one I just loaded before I figured out what had happened... Talk about a moment

Glad to hear things are coming along and you've about got your tune how you want it. Atleast drive-ability wise anyway right

I tried a few things over the weekend that help a ton. I kept adding more and more to the Closed Loop Proportional Airflow and it was getting better, I found the 00 LS1 Stock tune was so much more than the truck settings I decided to go ahead and just start with the LS1 Closed Loop settings. And right off the bat it was much better. I ended up adding 20% on Proportional and removed 20% From the Idle Base and its is very close. I may pull those numbers back if you say that is way too much and adjust with the Integral tables if you think that is better. But it was so close that it stopped the huge fuel trim oscillation that followed the O2 switching. I just need to fine tune these tables a little more like you said.

What is the normal rich lean over under for O2s? My latest scan you can see the wideband reflect the oscillation of the O2s but at light throttle say 10mph- 40 in get a little RPM oscillation that follows the the AFR oscillation. That light throttle surge is the only thing I am down to at this point. The VE is dialed in. I fixed the PE (The throttle Cable was out of adjustment! DUH) runs great at highway. I do have a little Knock at low load but nothing on the scanner would show a change that would result in know. So I think it must be false knock. Or a have a bad sensor?

However back to my original question about Openloop cold fueling.(which is how this all got started haha) I still can not get the actual AFR to match the commanded until it warms up. Its very minor at this point because I just have to wait a minute to warm up or its kinda lean even though its commanding it to be richer.


Here is my latest Tune and Scan:

Modified LS1 O2 CL Settings.hplASX_Camaro_SS_(CL)_v2.hpt

[GHuggins]

Have you tried adjusting your open loop eq up anymore? If you have your O2's oscilating correctly, then I would assume like you said you just need to work the fine kinks out and dial it in a little closer.

Now next question which I should have asked up front. What all's been done to the vehicle?

[LSXtreme]

Have you tried adjusting your open loop eq up anymore? If you have your O2's oscilating correctly, then I would assume like you said you just need to work the fine kinks out and dial it in a little closer.

Now next question which I should have asked up front. What all's been done to the vehicle?

Ya I added more to my Openloop EQ that its commanding like 11.2 at 50* but the wideband is showing like 13.3 at 50*. And when its warmer (like 120*) and start it up its like 15.2afr until it warms up normal temp then the AFR matches the commanded. I will copy over the LS1 stock table and see if that helps as the LS1 table is even richer than what I got now.

So a quick run down on the Closed Loop Integrator Delay / Airflow Mode table. (I think thats why you were referring to I look at mostly changing) If I add more Delay will that slow the STFT's down a little so they are not adjusting the actual fueling so quick? If not, what will slow the fuel adjustment down. Would any of the O2 Error tables help?

Whats been done to the vehicle.
Its a .030 over LQ4, dished pistons for turbo (currently 8.65:1), TEA CNC ported 317's, Comp LSXr Turbo Cam 235/231 115, BBK LTs with ORY. I had a bunch of problems with the FAST LSXRT manifold I have the stock TBSS intake and the stock LS1 injectors on it currently until I get the tune right then I will try the intake and 60# injectors again. Wanted to simplify a couple things so I can see results when changing tables in the tune instead of fighting with aftermarket parts. Hence why it doesnt have the turbo on it yet. I want to get it dialed in as best as I can maybe throw a little spray on it for the spring then put the turbo back on in the fall. But trying to tune a turbo right off the bat with all the different bolts ons was a nightmare on my last project. Chased my tail too long and got tired haha.

[GHuggins]

How did the LS1 table work out for your cold start fueling issue?

Do believe adding to the delay tables slows them down whereas lowering them as I've previously suggested speeds them up. SO increasing them should help you out with this... If you have a chance "I haven't" take a look at a 2010 5.3l auto colorado tune in the repository. The only reason I say this is because the last one I tuned had the slowest switching fuel trims of anything I've just about ever seen... Thought it might give you an idea on which ways to go.

[LSXtreme]

Ok, so the LS1 table did work for the OL fueling. (Check)

I checked out the Colorado tune and those O2 proportioning tables are completely different than the LS1's I did notice there is a little more delay in the integral as I have continued to add a little more to mine each time to see if that helps. The biggest thing right now for me is trying to plot which airflow mode I am in. Once I get that figured out I think I can fig my light throttle surge from the fast trim response. Everything above 50mph is great. Its just the 1200-1800rpm range at low speeds is where it needs attention.

If you, or anyone can help me in logging airflow mode against VE instead of maf that would be great.