Saturn Sky Forum banner

Will the Trifecta tune stop my ecu from pulling power?

10K views 61 replies 11 participants last post by  MattM 
#1 ·
I keep noticing this with the stock ECU settings. In second gear it really has a tendency to pull timing and boost, and I can really feel the power drop at the higher RPMS. It doesn't matter if I have traction or not. I'm just wondering if after I get the Trifecta tune my ECU will never again pull power due to loss of traction or it perceiving that power has increased "learn down."

I would bet that the majority of the 14.1-14.6 1/4 times that magazines and car owners see in stock form is the computer pulling timing and boost in the upper RPMS because it "thinks" the driver is too aggressive or more power is being made than should be. That's just my speculation anyway.

Back to the question though. Please let me know if after you got your Trifecta tune your ecu stopped pulling timing and boost just because it thought you were making too much power. I know it should still pull timing if it sees knock, but with anything else I wish it would quit pulling timing!!!!
 
#4 ·
This is why I like to scan to figure out what's going on with the car.

With the RL, I have far less experience with it's programming to be able to tell you what it's doing. If your car is stock, this isn't necessarily the learn down feature you hear about. The learn down feature is really in regards to performance modifications not netting more than stock power levels because of the ECM. If you're stock, something else is going on.

In my experience (which isn't vast by any means) there may be a number of things causing that pulling of power. You mention time of day...does it do it more when it's warmer out? This could be the ECM pulling power due to the temperature of the air charge to prevent KR OR it could be KR generated due to higher air charge temps that come in the warmer part of the day.
 
#6 ·
Either the Torque app with a Bluetooth OBDII Dongle or something like HP Tuners. The Torque app is usually the cheaper way if you just want to scan. The dongle is about $100 on eBay for one that gives enough resolution to record a scan and then the Pro software is about $5 I think. You can find cheaper dongles that will let you monitor the ECM values while you drive but they don't work as good for recording those values.
 
#7 ·
Is it worth trying this test out in competitive mode so see if a loss of power is still felt? This will help eliminate if the loss of power is deliberate. I dont seem to experience this condition as described, with or without Trifecta enabled.
 
#10 ·
Stock program has limits all over the place, so no doubt it is doing its job. Take a look at a factory torque curve some time - if it didn't limit torque low down the tires and possibly driveline would have a significantly shorter life. I never noted the tendency you did - and I suggest that you not word things to imply that you know the car "really has a tendency to pull timing and boost" when you have no evidence of that without instrumentation. You may have some issues that could be determined by having a shop with appropriate tools evaluate your car.

This is the stock dyno curve. It is not what actually happens in first (and, IIRC, second) gear, as the program has limits that smooth performance.



Suggest that you read these:

LNF Mod Primer - Pontiac Solstice Forum

GMPP Tune Learn Down?? - Pontiac Solstice Forum
 
#11 ·
Sorry WSpohn I didn't know I was offending anyone by saying what I see with my own eyes and feel. I literally see the boost in 2nd only get to around 13psi, and then I feel the car stop pulling after 5000 rpms or so. I then shift into 3rd and the car getting around 18psi all the way to redline and feel it continue to pull.

As I said, I'm getting a Bluetooth dongle and have the torque app and will know soon if it is because of knock or not. I don't know if they dealership put premium fuel in it, but I put 89 octane in on my first fill up. That might not be high enough octane, and causing detonation, but I won't know that. If it is not knocking then I will still have to figure out if it has to do with the ecu learning my driving habits or what's going on with it. I read in the thread that you linked were a few people were talking about the fact that they were steadily getting lower times at the track but there was no change in driving habits, and they suspected the ecu was just "learning" their track habits and slowly adding timing each run.

Is this something that the ecu does (i.e. learn driving habits)?
 
#13 · (Edited)
I don't know if they dealership put premium fuel in it, but I put 89 octane in on my first fill up. That might not be high enough octane, and causing detonation, but I won't know that.

Is this something that the ecu does (i.e. learn driving habits)?
You aren't offending anyone, but please explain how you know that the program is 'pulling timing". Isn't that an assumption on your part?
Actually, it may not be.

MattM, a brief run down on ECM "learning".

What most people talk about when they are talking about an ECM "learning" has to do with fuel trims. There are two types, Long Term Fuel Trims (LTFT) and Short Term Fuel Trims (STFT).

Whenever you reset your ECM (through a programmer, scanner, or unplugging the battery for about 30 minutes) you are resetting your fuel trims to 0. When a fuel trim, Long or Short term, is at 0, it is neither adding or subtracting fuel from the basic fueling table. This means, whatever values you have in your various tables that dictate how much fuel to add (fuel injector tables and Mass Air Flow, or MAF, tables mainly) are spot on and the trim has to do nothing.

Now, if your trim goes into a positive number, this means that whatever you have in your tables is off a little and the trim has to add more fuel to achieve the proper Air to Fuel Ratio (AFR) for those specific conditions...ei RPM, Throttle Position, MAF reading, fuel injector duty cycle, etc. If the number goes negative, then that means the trim is pulling fuel to achieve the proper AFR.

Like I mentioned, there is a Short Term Trim and a Long Term Trim. The STFT is what the ECM is doing RIGHT NOW to adjust fuel. If, for a certain condition the STFT is constantly adding or subtracting fuel, it will re-write the LTFT to the average STFT value for that condition.

So let's say for a certain condition your STFT averages -6. This means it usually is rich in this condition and thus the STFT is pulling about 6% of the commanded fuel to hit 14.7:1 AFR (because for this condition you're just cruising...makes things simpler). So since it's doing that, it writes -6 to the LTFT. Now your LTFT is at -6 and your STFT is at 0 for this condition. Why did the STFT change? Because the LTFT is already pulling the 6% the STFT averaged. Now, there may be times when the LTFT is at -6 and the STFT is at 1 or -1 too but it's average over time should be at 0 once the LTFT cell is learned for that condition.

What's important to know is that the values for the LTFTs and STFTs are cumulative. So if the LTFT is at -6 and the STFT is at -1 then you are pulling 7% fuel at that condition. If LTFT is -6 and STFT is 1 then you are pulling 5% fuel for that condition.

This is how the ECM "learns" not just your driving style but the little nuances of your engine, the environment you're driving in and how the engine is reacting to different conditions.

This is constantly happening and your LTFTs and STFTs are always updating themselves. As a tuner, one thing we do is tune the MAF table to get those LTFTs to learn to 0...or as close to 0 as possible. This means our ECM is loading the proper values to hit the correct AFR target in these conditions within +/-2%

To get a better idea of how all this works, check out these two images I made on closed and open loops in the ECM. This was for a Harley forum but the principle applies here too. This is just a very basic overview on what's going on with our ECMs:





HOWEVER, This learning has nothing to do with spark. This is only fuel. Our ECMs are designed to run our cars optimally on 91 octane or higher fuel. When you do this, you should see no knock and when the ECM sees no or only sporadic unpredictable knock (which is defined as false knock) it runs on a spark advance table called "Good Fuel Spark Table". This is your basic spark table and determines how much spark advance the ECU bases it's spark calculations on. Now, whenever you see knock, false or otherwise, the ECM will pull timing (reducing performance) to try and combat this knock. With False knock, because the knock is usually just there and gone, this timing comes back very quickly. With real knock though, the timing will come back more slowly as knock is reduced. The worse the real knock, the longer it takes the ECM to add it back in.

However, if you have knock consistently, the ECM will switch over to what it calls the "Bad Fuel Spark Table". This table has lower timing across the board. Remember, these Good Fuel and Bad Fuel spark tables are the base spark advance table your ECM uses before adding or subtracting timing for various conditions. Thus if you run on the Bad Fuel Spark table, your timing will always be less than if you're on the Good Fuel Spark table.

Running your car on anything other than 91+ octane will put you on the Bad Spark Table. This is why our engines are "recommended" to run on 91+. If you run on 91+, you get the Good Fuel Spark table and get more performance. If you run on less than 91+, the car isn't going to get damaged but it is going to run on the Bad Fuel Spark Table and the reduced performance that goes along with that.

So if he is running on 89 octane, it may be a safe assumption the ECM is "pulling timing" because it should be running on the Bad Fuel Spark table. This also means just scanning for knock may not tell us anything. If the car is running on the Bad Fuel Spark table and it's lower spark advance numbers to start with, the car may not be seeing any knock...and thus the Bad Fuel Spark Table is giving us the results is should be giving us, no knock when running on fuel with an octane lower than 91. We need to see the spark advance commanded to get an idea of what table his ECM is running on.

I will also add that running on 89 octane is the worst decision you can make. It's more expensive than 87 but gives you 0 performance benefits over 87. Run 87 or 91+...and with the turbo motors you really only want to run 91+.

Dropping your boost to 13 psi though is related to neither of these things. That sounds like your ECM is pulling boost for some reason...or you have some other issue that's causing a loss of boost pressure...though usually that's an all or nothing kind of thing if you're not getting some kind of engine code along with the lower performance.
 
#12 ·
You aren't offending anyone, but please explain how you know that the program is 'pulling timing". Isn't that an assumption on your part?
 
#16 ·
The factory tune is built to produce 260 ft pounds of torque and protect the motor mechanicals.

It adjusts (limits) power production parameters to achieve and not exceed 260 ft pounds of torque.

It also limits parameters to prevent over speed of the turbo. If it predicts that the turbo is going to over speed it will change parameters to prevent that condition.

It also has knock sensing and will change parameters to reduce / prevent damaging knock happening.

The boost level indicated is affected by air density. This is discussed at length in several threads on the Solstice forum by GM powertrain engineers. 18 pounds at 5000 feet is roughly equivalent to 13 pounds of boost at sea level. With an after market tune, it is typical to see boost levels in the low to mid 20s at all air density levels. My car went from 17-18 pounds indicated to 2 -24 pounds of boost with the GMPP tune at 5-6000 feet altitude in Colorado.

The confusing part of your report is that you are seeing a significant difference in one gear. My experience has been that each gear experiences the same boost level through out the gears when under acceleration. The oddity is seeing a significant drop in one gear. I don't know of any feature in the tune that would produce this result on a consistent basis. Basically the ECM does not know or care which gear its in so it should be providing the same boost for the same power demand at similar RPMs.
 
#17 ·
I guess I should have mentioned that I'm not new to tuning. Just with this particular platform and ecu. I built and installed a megasquirt standalone system in my supercharged probe and also tuned it.

The one thing I wish this system could do is just run off of the timing and fueling maps. There are so many different sensors inputting information that will cause the ecu to veer wildly off of the timing and fuel maps which can make it difficult to tune. I would much rather have a wideband and get on the dyno and update the tables accordingly.

With this ecu I just don't know what all could be affecting what.

Rob, I am with you on the gearing issue. That was why I was originally asking my question. The fact that it's in a certain gear should not dictate what level of boost I'm seeing, unless there is some function with the ecu that can read how quickly the RPMs are increasing and possibly see that as a problem then cut power back. It's the only thing I can think of.

In any event, I should have the Bluetooth dongle today or tomorrow. If I can get it set up correctly I'm hoping I'll be able to determine what is going on.
 
#18 ·
I guess I should have mentioned that I'm not new to tuning. Just with this particular platform and ecu. I built and installed a megasquirt standalone system in my supercharged probe and also tuned it.
That would have been good to know. LOL I never like to ASSuME so UNODIR (UNless Otherwise DIRected for those that don't read Marcinko...and if you don't, shame on you. I'm waiting for my referral check Demo Dick.) I default to the person having no familiarity with the subject.


I would much rather have a wideband and get on the dyno and update the tables accordingly.
Well, good news is you already have a wideband. The O2 Sensor in the LNF is a wideband sensor thus you should be able to get an AFR reading out of the Dongle setup. In Theory. I never have tried it with the LNF.

Also, keep in mind the LNF is a direct injected engine, not port injected. From what I've read, tuning a Direct Injected engine for fuel is VERY different than a port injected engine. Not sure how different or what the differences are but I wouldn't just trust your port injected knowledge from the Probe to tune fuel on this engine.
 
#21 ·
Trying to figure out that last column...is that KR? (Header is k2211a6)

Having throttle position would help too...RPM is jumping all over the place...2163 then next row is 1000 more then next row is 2500 more than that...Doesn't look like a great deal of consistency in the resolution of the scan to tell much. Seems your idle is around 2000 pm???? or are you cruising there? Either way, if that last row is KR, you have KR up the wazoo at 2068 RPM and it is running a bit lean there too (could be connected...pretty steady 15.3:1 AFR there).

You're right though, AFR looks a bit lean if you're at WOT in the 4-5K RPM range. I don't know what the stock tune on the LNF SHOULD run AFR wise but I know most tuners shoot for 10-11:1 AFRs...you're high 12s-13:1 it looks like. 12s is a bit lean for WOT I feel and 13 is really not where you want to be. Maybe the DI of the LNF is different though. My knowledge is with a Port Injected system.
 
#23 ·
Throttle position never registers 100% in our cars for some reason...it's like 85% but that's WOT. At least that's what I've seen in HPT. I had to hunt around a bit to find it too. LOL

If that last row is KR, keep in mind that if it sees that much KR constantly like it did at 2K rpm, it's going to switch to the bad fuel table and run on that. You were pulling 3% timing in some of those cells.

Seeing where that KR is will help a bit...but you really need someone who understands DI to look at it to be certain.
 
#24 ·
The last column is KR. It doesn't really bother me atm just because it is at cruise and a lot of vibration occurs there which could be causing the knock sensor to trigger. That said, at WOT I don't see any KR at all which is good.

I also checked the AFR information on these cars and I guess 12.7 to 13.2 is actually normal because of the variable valve timing adding additional air to the mix at the end of the combustion event. This seems to make the wideband read lean. I have to agree with this being correct because with only a 12.9 ratio at WOT you would have thought there would have been detonation, but there wasn't.

I'm going to run the fuel out of it and then put in 93 octane just to be sure.
 
#26 · (Edited)
GPS speed showing 180!!!!!! I hope to hell that's in KPH and not MPH!!!

What gas are you using now? I'm seeing a TON of KR in there...with all that KR, you have to be in the bad fuel table...no surprise it's pulling timing with that kind of KR regardless if it is false or real...and some of it looks like it could be real.

I see about 2 degrees KR in the 2,000 RPM range and 20-30% throttle range. Looks like high load low speed KR.

May have some burst KR too...rapid throttle increase and then one hit of KR. Though you did have one hit in there of 8% KR (!!!!) which gave you 0 timing. Stuff like this is not helping you.
 
#27 ·
I'm still running the 89 octane out of it.

I wish someone that knew redline datalogs would come in to let me know definitively what's wrong with it. From the boost I'm seeing it's definitely a stock tune so I hope that just the octane is causing the KR.
 
#28 ·
Here is a thread that discusses the proper afr in the lnf.

http://www.skyroadster.com/forums/f24/what-air-fuel-ratio-full-throttle-25488/

User Shabby posted this in the thread:

Here is the afr the ecm commands at certain rpm, if you're seeing 13.5 your probably at around 4500 rpm. Log one of your wide open throttle runs from 2000 rpm to redline and you'll see this on your dashhaw.

14.0 afr @ 1200-4000 rpm
13.6 afr @ 4520 rpm
12.9 afr @ 5000 rpm
12.8 afr @ 5320 rpm
 
#31 ·
From those posts looks like you AFR is right where it should be. Interesting it commands that lean of an AFR but as those articles state, this is an attribute of DI engines.

The KR is probably, most definitely, coming from the octane issue. If you're running 89, that's going to pull timing by putting you in the bad fuel timing table and then pulling more if you get KR.

If I were to guess, I'd say the KR is a combination of driving style and octane. You should have your RPMs in the 3000+ range when accelerating up hill or going hard on the throttle. considering your throttle says 67% is WOT, 24% is about 1/3 throttle. That throttle position and 2000 rpm is where you're seeing the most KR issues. Run a tank or two of 91+ in it, downshift a gear before getting on the throttle that hard, and I bet you don't see KR.

The KR you see in the upper RPM range I bet will disappear with the higher octane too. Now you may see a blip here or a blip there...and that is probably false KR...but you would want to see if you can find where that's coming from and address it. My highly modified Grand Prix could run with 0 KR and the only time you'd see any KR is hitting a massive bump in the road and jostling the car pretty good. Down the strip there was 0...not even .1%. Lucky for me most folks knew what areas of the engine bay were culprits for generating false KR.

So put some premium fuel in that bad boy, drive a tank or two of it through the engine to make sure the premium fuel hasn't had it's octane lowered by the remainder of the 89 you have in the tank when you first fill it with premium, then rescan and see what you get.
 
#32 ·
I'm going to put octane booster in it since I still have a half tank, and then I'm going to fill it with 93. I guess we will see then.

I was going to ask you if your blue tooth dongle provides datalogs with a lot more resolution than mine does? I don't think my dongle is good enough for quality datalogs.
 
This is an older thread, you may not receive a response, and could be reviving an old thread. Please consider creating a new thread.
Top