Thanks to the guys who gave me some pointers regarding a suitable turbo upgrade for high-altitude use. I've done a lot of research in the various technical articles here, and I have one simple set of questions (which probably will lead to a very convoluted set of answers). So, here goes.
How does the ECM adjust for altitude? I assume that it shortens the pulses for the injectors. Does it also adjust the timing? What sensors does it use (i.e. MAF, O2, MAP, etc.) And, how long does it take for the ECM to know the altitude has changed - or does it do it on the fly?
The MAF measures the airflow by mass, hence its name. Proper airflow measurement should be the primary factor which allows the ECM to properly meter fuel and spark adjustments. This shouldn't take any significant amount of time. I'll leave the hows and whys to those who are active students of the details.
Note that the MAP sensor on the TRs only controls the in-dash boost guage, and does not connect to the ECM.
The MAP sensor is responsible for altitude change. Inside is a little pressure sensor, I have seen Motorola's version. It measures pressure seen at the intake manifold (the vacuum line hooked to it), which is in turn outputed as a voltage from the sensor to the ECM. I used one to build a barometer in my senior design class in college. Hope this helps.
On the GN's, the MAP only drives the boost gauge as pointed out by Scott, above
(Copied from a post from Bruce to the list and put here to maintain some continuity to this thread)
Yes, the stock GNs don't use the MAP sensor as an input. (Jay Carter has a chip design that uses the Turbolink CTS to MAP conversion cable). There is also something else in the works to violate the all GNs do not use the MAP statement.
Anyway:
Other cars do run a baro correction using a similiar if not the same sensor (note: the Carb'd engines used a true Baro Sensor).
Anyway the fuel correction reads just exactly would you might think was right, ie high baro pressure, less fuel and lower pressure more fuel. Now how the heck does that work you say?. Well while the engine may consume a bit more air there is also a change in exhuast back pressure. Yep, you read it right. If you feel this is in error please feel free to look at the commented code for the Mask ID 8D. The correction factors are small, but it is interesting.
Bruce
87 Black Car, runs kinda better then stock
OK, guys. I've got a bit better than half the answer to my question. So, let me restate and poke a bit further.
Using the MAF to establish the total amount of air going in makes sense, as this is density-independent. I also agree that there would be a change in back pressure. Now for the specifics:
How does the ECM use this data to adjust the timing and mixture?
How does the ECM detect this change in back pressure?
Your insights are greatly appreciated.
Part 1 - Closed-Loop Fueling and Spark
(This is kinda long, so you may want to grab a cup of coffee)...
Here is my understanding of the whole fueling and spark timing process in the ECM. I invite those with more experience to correct me where applicable. All of the below statements apply to a "warmed-up" engine.
While you are in closed-loop (i.e. part-throttle or idle), the ECM uses the corrected MAF reading to estimate the amount of air coming into the engine. When I say "corrected", I mean that the ECM uses the reading from the MAT sensor to adjust the reading from the MAF sensor. There is an entire table in the code for this purpose. The ECM then calculates the fuel injector pulse width by multiplying the following numbers together:
Injector PW = [MAF(g/sec)]*[Injector Constant (sec/gram)]* [Commanded F/A in closed loop (0.068)]*[BLM/128]*[INT/128] / [RPM]*[Some conversions to make the answer be in msec's]
The BLM and INT values are determined based on the long-term and short-term readings of the O2 sensor. Notice that exhaust back pressure and MAP do not appear in the above equation. Those two things would appear indirectly in the MAF reading.
As far as closed-loop spark timing, it is determined from engine speed and engine load (LV8). There is a table to correct the spark timing for coolant temperature, which is the only correction table I could find with non-zero values. There IS a table in the code for spark advance correction versus MAT, but it is completely zero'd-out in the stock code. Again, MAP and back pressure are not directly accounted for.
One other note on closed-loop spark timing: at certain speeds and loads, the spark timing is advanced for the so-called "highway spark mode". In the stock code, highway spark is enabled. This results in an extra 4 degrees of spark timing for most cruising conditions.
Yet another note on closed-loop spark timing: There is a table for retarding spark when the TCC is locked. However, it is zero'd out in all codes that I have seen (including stock).
(see my next post for Part 2 - WOT Fueling and Spark)
Now for WOT:
As far as I can tell, all of the above statements on part-throttle spark timing apply to WOT spark timing. The only exception would be if knock is detected, in which case the ECM retards timing based on how many and how fast the knock sensor is pinging.
As far as fueling, the equation is essentially the same, but a few additional things are added:
Injector PW = [MAF(g/sec)]*[Injector Constant (sec/gram)] * [BLM/128] * [INT/128] / [RPM] * [PE F/A vs coolant temp] * [PE correction vs. RPM] * [PE correction vs. TPS] * [PE correction vs. time in PE] * [Some conversions to make the answer be in msec's]
In the WOT case, there is a table for the WOT F/A ratio versus the coolant temperature. In the stock chip, this value is 0.068 for all temperatures that we normally run at (it's a lot richer when the engine is cold). There are also three correction factors ("trims") that are used to adjust the F/A ratio. The first correction factor is the trim for RPM. In the stock table, this trim is zero'd-out for RPM's above 4000. For modified cars, this table MUST be altered above 4000RPM to allow extra fuel delivery when the MAF is pegged at higher RPM's. The second correction factor is for TPS position. Note that this table goes to 19% added fuel at TPS values above 75% (a little less than 4 volts), at least in the stock chip. Thus, the fuel delivery based on TPS does NOT care if your WOT TPS is 4.1 volts or 4.9 volts, contrary to some opinions. The final correction factor is the one least talked about. There is a timer in the ECM that times how long you have been at WOT. If you have been at WOT for over 10 seconds, the ECM starts increasing the fuel delivery. By the time you have been at WOT for 25.6 seconds, the ECM is adding 25% more fuel than when you started. This code is in there to protect you against knock and detonation as your pistons, chambers, etc. heat-up during WOT runs. So far, I have not seen any chips that alter this table. So, at the end of most people's quarter-mile pass, they're getting more fuel than when they started.
Note that there are no direct adjustments for MAP or exhaust back pressure. They would be indirectly accounted-for in the MAF reading.
I guess that's it for my dissertation on fueling and spark. Again, they may be a few things I don't know, so I invite corrections or additions to my information.
Mike Mshar
1986 GN
11.97@115mph
Stock Engine, I/C, and Trans, TA-49, Red Stripes, Lots of Tuning!
An open note to Bruce:
Please forgive the petulant tone of this post. Trying to weed through what you are trying to teach me (and others) is getting really confusing, and therefore frustrating. The reason I started this thread here (rather than simply taking it to the list) was to have all of the responses in one place. That way, anything that's murky could be referred immediately back to the previous post. Besides, this thread will (hopefully) lend itself to a coherent technical article submission.
PLEASE post back to this board! You obviously have a lot to add to this thread, and I don't know why you wouldn't want your thoughts here with everybody else's, standing alone on their own merits (and not interwoven with somebody else's ideas).
Please remember that you are the expert, and very conversant on this subject. However, not all of us share your knowledge level. Please know that I am trying to understand what you're saying, but, to this point, you've gone over my head! At least, if everything is gathered in one place, I can ask coherent (if somewhat simplistic) questions.
As an alternative, if this forum is not to your liking, we can take this to private e-mail and exchange questions and answers. Then I can put all the information together and do the article I referred to earlier.