06 Solstice 2.4 with Werks turbo - fuel adjustment questions

[raygun]

I've just installed a Werks Stage 1 turbo kit on my '06 Solstice. The relevant (I think) modifications are:

- Upgraded fuel injectors to 40lb/hr
- Precision 5558 (e) Turbo
- GM/Delco 2 bar MAP sensor

Everything else is stock. I'm not sure how to approach tuning fuel delivery. Any advice or good resources? I'm new to HP Tuners and am not exactly sure where to start. (I'm an engineer. Surely I can figure this out. [famous last words])

(I saw the injector flow spreadsheet elsewhere on the forum, but the only 4-cylinder listed is from a Cobalt SS, which I am guessing is the 2.0T LNF rather than my 2.4 LE5.)

My understanding (vague though it is) is that I can assume that the MAP sensor will "just work," and that adjusting the fuel map is a simple matter of reading what's there, dividing the new rate by the old one (40lb / 24lb if the stock injectors are 24's), and multiplying each of the fuel map values by that value. So, if 40 / 24 = 1.67, I'd just multiply each of the values in the fuel map by 1.67.

It's very possible I'm way off base. Suggestions are very much appreciated.

[mkriebs]

Did you buy it from Dave? There should have been a base map included, which would have these changes already applied. The injector spreadsheet you found is likely from the LSJ Cobalts, which is the 2.0 supercharged motor. It should work for this application. Post it up if you don't mind.

It's been a while since I have tuned a port injected motor on HPTuners, but I don't think you're quite on the right track...

[raygun]

I bought the kit second-hand. The previous owner bought it, tried installing it, and got stuck. (Fortunately, the only evidence of this was red RTV where a gasket should have been, and some scratches on the new exhaust manifold and cat where he apparently tried and failed to get everything to fit in the car.)

Dave has been incredibly helpful re:installation advice, providing replacement instructions, etc. But, since I just paid $500 for HP Tuners, and a lot for the car and the turbo kit, and I like to make things, I'd love to get familiar with the tuning process. The car is not my daily driver, so I have some time to spend figuring things out.

My understanding is that injectors are basically solenoids that can run from 0% (engine off) to 100% (engine trying like crazy to compensate for forced induction or a bad O2 sensor stuck closed loop) duty cycle. (PWM - they're either open or closed, probably with some adjustment for the solenoid latency.)

If all of that is correct - and I'm not sure that it is! - then, given the linear relationship between the old and new injector sizes - for instance, on a 24lb/hr injector, a 50% duty cycle will deliver about 1.4 grams/sec, while a 40lb/hr injector at the same duty cycle will deliver roughly 2.5 grams/sec.

ALSO NOTE that this is a 2.4 NA, not a 2.0T (GXP). I don't believe it's direct injected. (Though I may have misunderstood your comment about port injection. I'm new. And willing to accept that I'm way off track and trying to learn!)

Now, the problem is down to figuring out what the computer's expectations are (in terms of fuel delivery vs duty cycle), how they are expressed in the tune, and adjusting those.

Of course, one of my favorite quotes: In theory, there is no difference between theory and practice. In practice, there is.

[TurboCobalt]

Disclaimer first: I do not have an LE5, i am not a "tuner" so my knowledge is very limited.
That being said i see a few holes in your logic.

1) MAP sensors are not "plug and play" you need to change the tune for the particular MAP sensors you use, I think at the minimum there's a scalar and an offset, so you need to find out those 2 new values for your new MAP sensor.
2) Fuel injectors are not linear. In my little engine management book it talks about "opening delay" and "closing delay", opening delay vs 12V source differences, etc. sure once they open you could call them linear, but each injector has different opening delay, i'm not sure how your controller "models" these behaviors, but it's more than 1 scalar in any controller i read about.
3) A lot of tables in ECMs are 3d table, one axis being RPM, and one axis being some measurement of load, be it MAP, the confusing "air load" in LNFs, and i noticed most GM controllers use grams/cylinder. Your stock tune probably only has values in those tables up to 14.7 psi absolute pressure (0 boost, 0 vacuum, WOT). now that you will be making boost, you will need to rescale just about every table to also have values in the boost region, up to whatever boost you plan on using. otherwise the engine will act the same at 0psi boost, as at 10psi boost, and that seems somewhat unsafe.

I feel that going from NA to turbo, will include changing most major tables, not to mention actually tuning Volumetric Efficiency and spark advance. You mention that you thought you only have to scale fuel injectors? did you already figure out everything else out or were you not aware of the rest?

my apologies if i misunderstood your knowledge of the system, you just left out A LOT of questions, im hoping that's because you already had everything else figured out.

4) Edit: you seem to have taken on a significant tuning project, i notice most people recommend having a wideband O2 for this kind of work. I bring this up because i know the LE5 does not have a factory wideband O2 like the LNF.

[raygun]

Thanks for the detailed response.

No, I haven't figured out the rest. I'm just more clueless than I had thought. (But, it's a great opportunity to learn. Fortunately, I don't need this car to get to work and back, and it was cheap.)

It turns out that the turbo kit I bought second-hand (amazingly never actually run due to some issues the previous owner ran in to) came with a second ECM, which is apparently already tuned for this setup. (I wasn't aware of that before.) Seems like a good place to start will be to read the existing tune on that computer and compare it to stock, to see what has been changed and in what way.

So, in a *very* simplistic view of the world, we have, given the four-cycle (suck/squeeze/boom/belch), a few things to consider:

- Actual pressure in the intake manifold
- Actual fuel delivery. Injectors are mechanical devices and subject to a certain margin of error (which I guess explains the difference in cost between various injector options), which is where closed-loop feedback comes in.
- timing advance/retard. If running a little lean, advancing timing (which, to me, means sparking a little sooner), should help prevent pre-detonation, as the spark should go off before the mixture gets compressed enough to explode on its own. (A friend calls that "going Diesel". I'm not exactly clear on why that is.)
- Fuel delivery curve. From what I've read, when under heavy acceleration or WOT, the computer goes open-loop since the sensor latency makes closed-loop adjustments impractical.

Yeah - that's a lot of stuff, and I'm sure I'm missing a lot.

Please don't apologize for correcting me. If you're willing, explain it to me "like I'm a six-year old." I'm entirely new to this area and, aside from some time spent rebuilding engines and designing hardware (NOT for automotive), I'm completely clueless.

Thanks again for your time.

[raygun]

After doing a lot more reading and comparing the "tuned" file against the stock ECM, there are a *lot* of differences. Wish I had a test car that I could play that I wouldn't mind blowing up.

If anyone has any "Tuning for the Absolute Moron" links, I'd be grateful if you would share.

[TurboCobalt]

raygun,
I think you're saying a few things wrong, or at least i have a different understanding than me:
Your understanding of spark timing and it's relationship with air-fuel ratio seems incorrect. You want to set spark to Maximum Break Torque (MBT), for every different load and rpm variation. MBT is generally a few degrees Before Top Dead Center (BTDC). In reality gasoline can pre-detonate, self ignite, due to the high pressure, high temperatures in the compustion chambers; aka knock. knock is bad, mmmkay? So in reality you can't always set spark advance to MBT, because the engine might knock before you get to such an aggressive spark advance, so you have to back off to a point where it doesn't knock. Notice this has little to do with air-fuel ratio.
A slighly rich AFR is slighly better torque than stoich, that's why engines go slightly rich at WOT. more importantly, a rich mixture doesn't burn as hot, so the unspent fuel in the combustion chamber acts as a cooling effect on the chamber and exhaust components, i think engines run rich under high load in order to protect the engine and turbo and cat, not for torque. Cooling the chamber will also reduce knock.

This is way too complicated to try to learn in a forum thread, i'm guessing that's why others have stopped posting.
You claim you're an engineer, just out of curiosity, what kind of engineer?
As a fellow engineer this is probably the best advice i can give you, and what i did. Do not listen to me about engine control, go back to the basics and get yourself a book, or a few. I went through the 2 Greg Banish books, "engine management" and "fuel injection systems" and those will give you an idea of what's going on, they're 20 bucks each, and you'll get way more out of those than the money you payed for your turbo kit. That should put it in perspective the amount of work needed to get that solstice running.

I don't like trying to learn such critical things such as "engine control" from online "links" because as it turns out, not everything on the internet is true, but you'd be surprised how far some google searches will get you. Try "engine control theory", "speed density", "MAF control?", "engine management", "engine volumetric efficiency", "wideband o2 sensor", i dont know, im making this stuff up.

Ultimately you have 2 choices: A) use this second controller you have, and have a blind trust in the tuner that made it. With the very little experience i've had with engine calibrations, i'd never do that. B) get some books and understand every change that was made to that calibration file, and have some educated opinion.

Once you get some knowledge and come back here with some more detailed questions these guys are very helpful (or at least they were for me).
I'm hoping if i said anything wrong, someone will come in and correct me.

[raygun]

Thanks for the information, and for taking the time to write such a detailed reply.

The tuned ECM I have was done by Werks for this car and this configuration, so I'm reasonably confident it won't blow anything up. I'll most likely have the car dyno tuned.

(Just for the record, I realize I'm effectively a moron in this area! Thus the questions and requests for resource references rather than blindly diving in and changing numbers. I'm not offended by assertions that I have no idea what I'm doing.)

I'm an EE and software guy (some embedded systems, mostly PC and server stuff), so this is *way* outside of my area of expertise. I'm trying to learn as much as possible without destroying the car. In a perfect world, I'd have a simulator that I could use to evaluate parameter changes without having to mess with the real hardware, but for something this complex, that seems like a lot to ask for.

I'll pick up a couple of books, read/process and come back with more intelligent and informed questions.