ECMlink AFRatioEst leaner than wideband with large spikes at cruise

[OwenRonn]

My AFRatioEst is way off from my wideband and it shows around 18afr at idle while the wibeband shows around 14. the AFRatioEst also has seemingly random large spikes at cruise. I don't think it is related to the MAP sensor because I just installed a new AEM 3.5 bar sensor.

Is it something related to my VE table? I have more logs if needed to narrow down the problem.

 

[We're on Boost]

The thing that sticks out to me in the log is the Combined Fuel Trim which is way too much negative. It's like 27% negative. So closed loop is pulling a lot of fuel out. That means the ecu thinks it would be way too rich without the 27% correction downward.
That is from 16.9 sec to the end. Before 24.7 sec the wideband is still sleeping, so the wideband numbers you see there are fake, but after 24.7 sec the wideband is awake and giving you numbers that are probably true.

Anyway, looking for why it would be too rich without the fuel trims:
Your VE in there is only ~59% which should be ok.
Your Global Fuel is actually a couple % leaner than mine with the same injectors, so I don't think that is too rich.
Your injBatteryAdj numbers at 12 and 14 volts look just like the numbers on the FIC test sheet that I've seen for those injectors, so that's probably ok. But then you are adding deadtime to those numbers with Global Deadtime of 300 microseconds.
So I'd try changing that 300 number to 0. Shouldn't need anything there. The InjBatteryAdj numbers should be enough all by themselves.
I'd try that and see if it helps.
Also, put CrankingFuelAdj in your logs.
And, there is only idle in this log, no cruise.

 

[OwenRonn]

I just got a log where I tried idling on 0 deadtime and I have a log where I drove it home from where I was working on the car. Do you think it could be a boost/exhaust leak.

 

[OwenRonn]

I just got a log where I tried idling on 0 deadtime and I have a log where I drove it home from where I was working on the car.

This is the cruise log i had it locked in open loop because it seemed to be running better and i needed to get home

 

[We're on Boost]

The log from post #3:
Ok, now the combined fuel trim is adding fuel (16.8%) instead of subtracting, and LC2 has swapped places with AFRatioEst. And the FrontO2 is not even cycling. And the 16.8% added by the fuel trim is not even doing enough. So it looks like we overshot by going from 300 to 0.

You should try setting Global Deadtime to about 180.

With an 850 rpm target idle, it will probably be pretty sensitive to deadtime. If it turns out to be too sensitive, then raising the target idle to like 1000 rpm should make it easier. My car with these same injectors has a target idle of 1100 rpm. Part of that is cams though. I don't see anything about cams on your car. Are they stock cams?

I don't understand why the MAP number came up in a hump at 15.7 secs, and then back down. Throttle didn't move at all. That could be a leak somewhere. But then it went away? Don't know.

Your cranking fuel adjust is working - that's good.

 

[We're on Boost]

The log from post #4 (locked in open loop)(with Global Deadtime at 300 again):
At 933 seconds in 4th gear with load factor around 1, your LC2 says AFR is about 11.7 which is pretty rich there. Your VE table has numbers that are quite a bit high in that area. So that one I would put down to the VE table.

Otherwise in the last log your LC2 AFRs look good (in the low 14's usually) whenever you are cruising along in 2nd or 3rd with light throttle, and they look way too rich (~11) whenever you are idling. The too rich at idle would come down to the Global Deadtime again.

I'd say try setting Global Deadtime to about 180, turn off the lock in open loop, then do another drive. If the idle seems sketchy, turn up your target idle to 1000 rpm but still with about 180 Global Deadtime. The idea would be to run a whole driving log like this with fuel trims working. Need closed loop to get fuel trims. With a log like that, you can run a SD VE Adjust (Combined Fuel Trims) to get a better idea how far off the VE map is.

 

[We're on Boost]

Is your VE map supposed to be good for the Evo 3 intake manifold?
If there is no particular reason for it, you'll probably start off better with the "stock" Speed Density map that is in the "Stock V3 ECU Config data files". If or when you want to try that, here's the file for stock 2g:

 

[OwenRonn]

I set deadtime to 180 and ran another log today and both idle and cruise seem to be much closer to stoich i am still seeing some funny behavior from wb and afr with lean spikes so maybe that could be a leak. I am wondering if I should drop in the stock ve map or tried to adjust the one i have with sd ve adjust. Also i am running stock cams.

 

[We're on Boost]

Looks like the 180 global deadtime really got it right for idle condtions. As near as I can tell the 850 idle speed is ok for it in this log. Both during warmup and at the very end of the log. BTW it is really good to show some idle at the end of the log like you are doing, because everything is stabilized warm then. It's also good to see the cranking, starting, and warmup like you have been doing.

I looked at the AFRatioEst spikes and all the spikes I see are immediately after you lift entirely off the throttle. Those are ok. The InjOn time doesn't even go to 0 in those places. It is just settling back suddenly to about what it would be at idle. I don't think there is anything wrong there. When you are cruising or at steady idle, I don't see spikes. You will see InjOn time going to flat 0 when you lift off from higher rpm, and that is ok too.

In the cruise (light accel actually) parts of the log, the Combined fuel trims are very high, around 15% when throttle is around 15 to 20%. Light accel. At 26% throttle accel (like at 608 seconds) the CombinedFT is less but still positive, +5%. So it needs more fuel, but the VE map already has numbers that look kind of high, so I think you might need to do that with the Global Fuel number. Instead of -72.7% you could try the number I use, which is -69.9%. (same injectors). ECMtuning says that the FIC 1650cc HiZ injectors are rated different so they work more like a 1500 or 1550cc in the calculator, and that's why there is this discrepency.

This log is pretty good for running SD VE Adj (Combined Fuel Trim) because there is a lot of light accel that stays in closed loop. The one at 608 stays in closed loop all the way. At the end of it the Combined Fuel Trim gets all the way down to 3.5%, and sure enough, the SD VE Adj suggests no change to that cell. But above that and to the left on the map it suggests large increases in a lot of the cells that are nearby. Farther above and left (near idle) the changes suggested are very small or none. Closed loop won't go much farther than what you did here at 608 secs. But when you go farther in your accels, getting into open loop, we can compare the LC2 WB to your table that's called OpenLoopMaxOct. That table looks not too bad but maybe it will need some changes too.

Anyway since the changes suggested by SD VE Adj tapered down to nothing at the high end of your closed loop accel, maybe the rest of the VE table that is lower and to the right could turn out to be ok? I really don't know.

I have to do some other stuff for a while but I can look at this more tonight. The VE map, I don't know, where did you get this one from? Was there some particular theory about it?

Here's a shot of the suggestions SD VE Adj (Combined Fuel Trim) made for your VE map. This is easy to run from a right click and it won't change anything if you just want to run it to look when you aren't connected. Notice the cell that I highlighted says 96.0 and that is no change from what you already have in your table. That is at 608.4 sec in your log. But several cells nearby to the left and above have large changes suggested.
If you change your Global Fuel number, of course then all of these SD VE Adjust suggestions would be different, the VE numbers would be lower.

You must be logged in to view this image or video.

 

[OwenRonn]

My VE table was left on by the previous owner so I cant tell you exactly why it is set up the way it is, I noticed my table looks different from others tables but I didn't want to make a lot of changes not knowing why it is set up the way that it is. I have made adjustments to the table in my ownership but nothing major. The log attached is after i changed the global fuel to 69.9.

 

[Tony1094]

My VE table was left on by the previous owner so I cant tell you exactly why it is set up the way it is, I noticed my table looks different from others tables but I didn't want to make a lot of changes not knowing why it is set up the way that it is. I have made adjustments to the table in my ownership but nothing major. The log attached is after i changed the global fuel to 69.9.

I'm still learning ECMLink myself, but it looks like your deadtime can probably go back to 0 or at least lower now since you increased global fuel. Personally, I would idle to operating temp, reset your long term fuel trims, and with it running slowly lower your deadtime until CombinedFT is as close to 0% as you can get it.

 

[We're on Boost]

with it running slowly lower your deadtime until CombinedFT is as close to 0% as you can get it.

Yes, Owen what he is talking about is, look at the checkbox in the lower right part of this shot that says "Save changes to ECU as they are made". If you turn that on, you can change the global deadtime while the car is running. You can watch the live log to see when the combined trim gets around 0 and also the AFRatioEst and LC2 wideband numbers should be close like they were in the previous log. Just at warm idle. You don't see this checkbox unless you are connected. When you are done, you don't need to push a Save button because it is already saved.

You must be logged in to view this image or video.

 

[We're on Boost]

In the log from post #10:
The combined fuel trims are still between +5% and +10% on the light accels, even though global fuel is more now. So it wants more fuel in those areas. You could do that in the VE map using SD VE Adj (Combined Fuel Trims). I think you could even fix the cell that is right at idle that way. VE Adj says to put 53 in that one (1000 rpm at 5.5 psia). If that fixed it you wouldn't need to change deadtime. That might be cheating a little bit. Not sure. But if you change deadtime, it will change everything a little bit, so you'd have to run another log like this to get new suggested VE numbers. I suppose changing deadtime is the most correct way of doing it.

In the heavier accels, the best one is 27% throttle at 179 secs, the combined trims get very close to 0, and the WB and the AFRatioEst are right on top of each other. So like in the previous log, the VE table that you have is starting to look pretty good as the LoadFactor goes up. The end of that 27% throttle is at 179.8 sec and the LoadFactor there is 1.20 with 4 psi of boost, combined FT is only 1.8%.

So now because of the way it's converging at higher loads, I don't think you should switch to the stock 2g VE map. You would either have to increase global fuel by quite a bit, or else you would have to bring the stock VE numbers right back up, because those stock numbers are so low. Look at this shot of the stock 2g VE (upper table) and the VE Adjust numbers suggested (lower table) from this log.
The current VE map is much safer as you start doing open loop pulls, which you could actually do right now even as it is. Up to a point. Like going 70 - 80% throttle up to 5000 rpm in 2nd gear. Watching the CEL which should flash on if there is any knock over 2 degrees of retard.
Anyway the current VE map is safer because it is richer. It will be safer to start with a map that is richer (higher numbers) than to start with a map that is too lean. That cell that is highlighted in the lower map, that is Tracking the log at 179.802 seconds. It's got 94.5 in it which is unchanged from what you had already. But look at the suggested numbers that are above that compared to the stock 2g numbers. Pretty big differences.

I noticed also that already you are going into ignition timing numbers that are really low, like 8 degrees. That whole lower left area of your Timing Max Oct table is all light blue, very low numbers, which is how some people like to start out because it is safe. So your timing is also very safe as is for starting to do light pulls to look at open loop AFRs.

You must be logged in to view this image or video.

 

[OwenRonn]

In the log from post #10:
The combined fuel trims are still between +5% and +10% on the light accels, even though global fuel is more now. So it wants more fuel in those areas. You could do that in the VE map using SD VE Adj (Combined Fuel Trims). I think you could even fix the cell that is right at idle that way. VE Adj says to put 53 in that one (1000 rpm at 5.5 psia). If that fixed it you wouldn't need to change deadtime. That might be cheating a little bit. Not sure. But if you change deadtime, it will change everything a little bit, so you'd have to run another log like this to get new suggested VE numbers. I suppose changing deadtime is the most correct way of doing it.

In the heavier accels, the best one is 27% throttle at 179 secs, the combined trims get very close to 0, and the WB and the AFRatioEst are right on top of each other. So like in the previous log, the VE table that you have is starting to look pretty good as the LoadFactor goes up. The end of that 27% throttle is at 179.8 sec and the LoadFactor there is 1.20 with 4 psi of boost, combined FT is only 1.8%.

So now because of the way it's converging at higher loads, I don't think you should switch to the stock 2g VE map. You would either have to increase global fuel by quite a bit, or else you would have to bring the stock VE numbers right back up, because those stock numbers are so low. Look at this shot of the stock 2g VE (upper table) and the VE Adjust numbers suggested (lower table) from this log.
The current VE map is much safer as you start doing open loop pulls, which you could actually do right now even as it is. Up to a point. Like going 70 - 80% throttle up to 5000 rpm in 2nd gear. Watching the CEL which should flash on if there is any knock over 2 degrees of retard.
Anyway the current VE map is safer because it is richer. It will be safer to start with a map that is richer (higher numbers) than to start with a map that is too lean. That cell that is highlighted in the lower map, that is Tracking the log at 179.802 seconds. It's got 94.5 in it which is unchanged from what you had already. But look at the suggested numbers that are above that compared to the stock 2g numbers. Pretty big differences.

I noticed also that already you are going into ignition timing numbers that are really low, like 8 degrees. That whole lower left area of your Timing Max Oct table is all light blue, very low numbers, which is how some people like to start out because it is safe. So your timing is also very safe as is for starting to do light pulls to look at open loop AFRs.

You must be logged in to view this image or video.

Im going to adjust the deadtime then continue to adjust the ve table building off the table i have. I really appreciate the help.

 

[We're on Boost]

In order to be settled on a certain number for the Global Fuel, you have to see how well it works out at full throttle in a pull to 5000 - 6000 rpm. Because if it's way off there, you would have to change the Global Fuel number, and then you'd have to change the whole VE map all over again.
By "way off" I mean that your maximum VE should be a number pretty close to 100, and the AFR you get in that region should be the AFR that you want (and that works!).
Maximum VE usually occurs somewhere in the 4500 to 5500 rpm range.
You'd probably want your AFR at high power to be about 10.5 if you are running 93 octane gasoline.
You could have it a little richer than 10.5 and it would still be ok, or you could have it a little leaner (like 11) if you aren't having a problem with knock.
The reason you'd want your max VE to come out to be about 100 is because that is what is realistic.

Hopefully your boost control setup and boost gauge are working ok. When you start doing pulls you should be keeping the boost below about 20 psi just so you can see if there is knock or anything else going off.
As boost goes up, the spark plug gap needs to be less and less. Usually people wind up with a plug gap of 0.020" for any boost of 20 psi or more.

 

[OwenRonn]

In order to be settled on a certain number for the Global Fuel, you have to see how well it works out at full throttle in a pull to 5000 - 6000 rpm. Because if it's way off there, you would have to change the Global Fuel number, and then you'd have to change the whole VE map all over again.
By "way off" I mean that your maximum VE should be a number pretty close to 100, and the AFR you get in that region should be the AFR that you want (and that works!).
Maximum VE usually occurs somewhere in the 4500 to 5500 rpm range.
You'd probably want your AFR at high power to be about 10.5 if you are running 93 octane gasoline.
You could have it a little richer than 10.5 and it would still be ok, or you could have it a little leaner (like 11) if you aren't having a problem with knock.
The reason you'd want your max VE to come out to be about 100 is because that is what is realistic.

Hopefully your boost control setup and boost gauge are working ok. When you start doing pulls you should be keeping the boost below about 20 psi just so you can see if there is knock or anything else going off.
As boost goes up, the spark plug gap needs to be less and less. Usually people wind up with a plug gap of 0.020" for any boost of 20 psi or more.

I will check my boost control setup before doing any pulls it works as far as I know but its hard to judge the amount of boost im running with the manual control I have it turned down pretty low right now.

 

[We're on Boost]

I will check my boost control setup before doing any pulls it works as far as I know but its hard to judge the amount of boost im running with the manual control I have it turned down pretty low right now.

Right. But yeah you can go with that.
If you want a better boost control setup in the future, you can use the Boost (WGS) tab in ECMlink along with the Ingersoll-Rand boost control solenoid that ECMtuning recommends as being most compatible with the driver chip in our ecus.