Results 11 to 20 of 96
Thread: adventures in turbo tuning
02-02-2011, 08:09 PM #11
OK, time to get back in to this. I filled up my car yesterday and realized that it was the first time I put gas in it since the day after Thanksgiving. And no, I'm not one of those 500 mile per tank guys. Or even a 200 mile per tank guy, come to think of it. just been travelling a lot lately for fun (good) and work (not as good).
The last few posts I made were all about what AFR was "best" for boost (12) and non-boost (14.6). Now comes the fun part. How do you do that?
The answer is as simple as saying "Piggyback ECU." but the execution isn't as simple as wiring it in and driving off (or even using a tune from another car).
Modern engines are universally controlled by an Engine Control Unit (ECU). The ECU has many functions, but here were going to limit the discussion to controlling the AFR. For a normally aspirated engine (ala smart) the ECU is programmed to keep the AFR at 14.6. Were going to talk more about this, but for now, let it suffice to say that the ECU does a really good job at keeping the AFR at 14.6.
All other things being equal, if a turbo was simply added to the engine, the ECU would keep the AFR at 14.6 in boost, which isn't the "best" AFR for boost. But all other things are not equal. The stock ECU is going to fail to keep the AFR at 14.7 for at least two reasons. The stock injectors can't flow enough fuel and the stock manifold pressure (MAP) sensor is going to max out at 1 atmosphere, so the engine really won't "know" how much air is flowing into the engine. A lot of air will be under reported by the MAP sensor, and even if it was, the stock fuel injectors can't keep up.
Any credible turbo kit will come with bigger injectors so, that problem is (hopefully) solved.
Any credible turbo kit will also (likely) come with a piggyback controller that will have a new MAP sensor that can detect the increased pressures properly. There are other ways of solving this problem, but this thread is about my smart, which has a piggyback controller, so that all I'm going to talk about.
Oh, if it was as simple as wiring in the piggyback controller....
Here we go! The piggyback controller serves three functions. Here, I'm specifically referring to the Split Second FTC -1, which is what I have and what SFR sells with their kits:
1- Has a 2 bar MAP sensor to "over ride" the smart's stock 1 bar MAP sensor.
2- It has a fuel trim map that can increase or reduce the smart's pulsewidth to the fuel injectors
3- It has an enriching circuit that detects when the car is in boost so that the proper AFR can be achieved in boost and non-boost.
I might be missing some obtuse fourth function -- any of you gear heads keep me honest, but I'm thinking those are three main ones.
Now I'm going to give a teaser here. Items two and three above are separate and independent functions. I'll be discussing that in future posts, but I'm asserting here that the elusive 12 AFR in boost phase CANNOT be achieved by adjusting the fuel trim map.
OK, so let's wrap up today's post.
- The stock ECU for a normally aspirated car tries really hard to keep the AFR at 14.6 and it is really successful at it. (We'll get in to that more in a later post)
- Add a piggyback controller to provide the desired 12 AFR in boost.
- Adjust said piggyback controller to work in concert with the stock ECU, Ah, now there's the rub.
Keep it tuned here for further epistles.
Last edited by geosynch; 02-02-2011 at 08:26 PM.
02-03-2011, 07:34 PM #12
Yes indeed, and just important... Ignition timing control under boost....
02-04-2011, 06:33 PM #13
Ah, yes, thank you o-sensei. I'll get to spark advanced after AFR... I have less to say about spark advance, maybe because it is more of an artform?
Off to do a bit of tuning now...
02-05-2011, 02:29 PM #14
Wasn't refering to advance timing but the general rule of thumb of retard timing applied to offset boost.
Overall just filling in the blank for the fourth function of a piggyback ECU you mentioned previously.
What will be the next chapter in, "the adventures of turbo tuning" ?
Last edited by BARNEY; 02-05-2011 at 02:55 PM.
02-05-2011, 05:42 PM #15
ADvance, retard... Whatever. The retard (me) has his brain stuck in advance. Sorry, but I think in terms of advancing a negative number, which is what normal people call retard.
Spent the day (re) tuning my vacuum region. My LTT have been in the -20% range since I got new plugs in last fall and this is really the first chance I've had to tweak it. But, I'm getting ahead of myself.
My aim with this thread is to lay a foundation of what needs to be done and why. Reason is, I think I'll get a few raised eyebrows when I get to the nuts & bolts of the fuel map. So, I figure it's best to lay a foundation I can stand on when the time comes that I get darts thrown at me.
Next chapter, I think, is going to be a bit of a tangent on control systems. I have to think a bit more about it.
Signing out for now.
02-05-2011, 08:42 PM #16
Keep up the good work. If I get stumped I'll ask for extra handouts.....
02-08-2011, 06:52 PM #17
Max boost for no tuning
If Brett is reading this, I have an answer for his question he posted on SCOA that is different than the other replies.
First, I forget who suggested the O2 sensor, but (s)he is dead on. That should be the first step of any FI project. To be specific, an aftermarket wideband O2 sensor that is completely outside the ECU's control loop.
Second, the max boost you can add without tuning is "not much." Let's me be more specific.
Without any tuning, your ECU is going attempt to maintain a 14.6 AFR. As long as the MAP sensor is not saturating, the ECU will be successful in maintaining the AFR at 14.6.
As soon as the MAP sensor saturates, the mass flow rate of air going into the engine will be under reported. The ECU will calculate the fuel for a 14.6 AFR but that will be correspondingly low. Then the ECU's feedback loop will determine the engine is running lean and it will attempt to compensate by adding fuel.
As you increase pressure, the difference between the ECU's initial guess and the error signal in the feedback path from the ECU's O2 sensor will grow. This will lead to hesitation in power delivery and thrown engine codes.
I think the MAP sensor saturates at 15.1 psi, but the output of the sensor will read 5 V. I may go for drive around the block and see what the stock ECU is reading on my scangauge while I'm in boost. That is the "saturation" I'm referring to. The difference between saturation (assume 15.1 until someone verifies otherwise) and your local atmospheric pressure (14.7 if you are at sea level) is what I would recommend for max boost without tuning.
If you go higher, the engine will run "weird" in boost and you'll start throwing engine codes. If you go higher still, your injectors won't be able to keep up. You'll see that in the aftermarket O2 sensor you installed in step one.
02-08-2011, 07:44 PM #18
OK, I'm ready to pontificate further.
Before we get into the nuts & bolts of tuning, it is important to understand what a control system is and how it works. This is because the ECU is in fact controlling the AFR (among other things, but we're limiting the discussion to AFR).
Back in the day, engines had no ECU's. Everything was "open loop." To fully understand what this means, please permit an analogy. I think it is illustrative.
Once upon a time, before Jet was born, man lived in a cave. In the winter he heated his cave with fire. If he got cold, he threw another log on to it. This is open loop.
Even as late as the turn of the 20th century, most homes were still heated "open loop." A home may have had a wood (or coal) burning stove. When it was cold, someone added fuel. They may have even added different fuel in different amounts depending on whether they wanting quick heat for cooking, or long-lasting heat for keeping the house warm all night, but it was all still guesswork and if it got too hot in the house, there wasn't much you could do (except open a door or window).
Now when I was a kid, my parent's house had a simple thermostat. If someone was cold, all they had to do was bump up the thermostat a little bit, and the furnace would blow hot air until the temperature reached the current setting. You guessed it. Closed loop.
It was a pretty big house, but there was still only one centrally located thermostat for the entire house. Some rooms got cold, even when the room with the thermostat was warm. In fact, I slept in the basement. I remember waking up and seeing ice in the toilet by my room more than once. But I digress.
In my home, the environment is controlled in zones. The loft in the attic is one zone, three are two zones on the main floor, and another zone in the basement. I can set the temperature in any of these zones independently, and the temperature settings can be different depending on whether or not that zone is occupied and/or time of day.
The point is, it is closed-loop, but much more sophisticated than the environmental controls in the house I grew up in.
Further, the furnace in the house I grew up in and two settings: on & off. When the temperature was lower than the thermostat setting, the furnace kicked on. When the temperature was reached, the furnace kicked off. In the trade, this is referred to as "bang-bang controller."
The furnace in my house doesn't operate that way at all. Depending on how large the error signal is, the furnace will generate more BTU's. The error signal is the difference between the actual temperature and the setting on the thermostat.
Why is any of this relevant to tuning the smart (or any late-model FI project)? Because the ECU is in control. Fail to understand that, and you'll fail to achieve the desired results.
All cars were "open loop" until the mid 70's. When the first "lambda sensor" was invented in the mid-70s car manufacturers quickly saw the benefit for ever increasing pollution control laws and market pressure to increase efficiency.
Through the 80's and 90's, the control systems were fairly simple. Not too unlike the simple furnace I described above. But within the last few years, the control systems get increasingly more sophisticated. Tuning rules that were learned in the 90's and the early part of this decade no longer apply.
For earlier cars, the AFR was controlled only at the lower end of the RPM and/or load range. When the engine was under load, the ECU went "open loop" because the O2 sensors simply couldn't deal with the increased flow, nor could they work outside a narrow band around the stoichiometric ratio.
Tuners capitalized on that. When the ECU was open loop, they were free to add as much fuel as they wanted via bigger injectors, higher pressures, and/or by adjusting the injectors pulsewidths.
That's no longer true. Now most engines can, and do, stay closed loop essentially always. On the smart, the only time I have observed open loop is during fuel cut (deceleration) and at wide open throttle at high RPM and high load. For my case, I'm at 10 psi boost by 3K RPM and an engine load of 50% or less. I affect my AFR waaaay before WOT and redline.
That means you have to tune the car when it is closed loop. Trying to achieve an AFR by only using any of the "old ways" (bigger injectors, higher pressures, and/or by adjusting the injectors pulsewidths) simply isn't going to work because the ECU will detect the added fuel and compensate.
And not only that, when the error signal between what the engine using actually using and what the ECU expects (its initial guess*) is large, the car will simply not deliver power smoothly, and it will start throwing codes.
And that, my friends, should get a few people sitting up straight in their chairs.
Next up, will start getting into the nuts & bolts of tuning a SS FTC-1.
*This will become important in later discussions. Stay tuned
Last edited by geosynch; 02-08-2011 at 09:20 PM. Reason: still finding them grammar errors...
02-08-2011, 08:01 PM #19
This is scary... this is starting to make sense to me. ME! (if you only knew).
Just wanted to say thanks.GetSmrt
02-09-2011, 09:43 AM #20
I can happily say that my project plan starts with putting in a wideband O2 sensor first, the bung should be welded in place in a few more days. You thread is such a good read, thank you.
By SMART MADNESS in forum SMART MADNESS Product SpotlightReplies: 6Last Post: 04-30-2015, 11:41 AM
By MADNESS Jason in forum SMART MADNESS Related News and AnnouncementsReplies: 0Last Post: 12-04-2014, 02:51 PM
By NukleoN in forum smart General DiscussionReplies: 3Last Post: 10-30-2011, 03:20 PM
By DonTaylor in forum smart car Performance ModificationsReplies: 7Last Post: 03-03-2011, 11:25 PM
By SMART MADNESS in forum smart car GalleryReplies: 0Last Post: 02-11-2011, 02:14 PM