Good questions MarkEmark,
MarkEmark wrote: So is the ECU really that "static" as far as controlling timing goes? In other words, it doesn't deviate much (if) at all from the timing maps programmed into it, regardless of any other factors, such as the difference in air density between a 20 degree night and a 100 degree day, holding "load" constant (e.g., WOT, 4000 rpm, 4th gear)? It'd seem to me that it'd be able to run quite a bit more advance safely on a 20 deg. night compared to a 100 degree day.
The timing will deviate from the map value only due to engine temperature (you hit the nail on the head here). There is a seperate set of values that will dictate an addition or subtraction of timing advance due to temp. when the throttle is open, and one for when the throttle is closed (2 seperate tables, maybe more). Full understanding of the ecus exact usage of these tables is yet to be determined (see some of the later posts at eccs.hybridka.com).
Enthalpy/JWT are either unaware of these values or do not care to adjust them. There are certain applications that can benefit from a modification to this temperature based timing adjustment, most of which falling under "performance" turbo applications when running high boost. An unintended change of 1-3 degrees can certainly be an issue, depending on when the car was tuned (temperature) and what not.
Knock is a little different I, I will explain below.
MarkEmark wrote:The ECU changes timing based on load....what does, say, a JWT KA-T ECU use to measure load? Throttle positions sensor? MAFS voltage? RPM? Gear? I know it doesn't use boost because they don't ask you how much you plan on running....
"Load" as commonly referenced when discussing a Nissan ecu is actually a base factor of injector pulsewidth (TP, Theoretical Pulsewidth). The ecu has calculated that this pulsewidth, theoretically, would provide a 14.7 (stoich) AF for that particular RPM and load point on the map (at a "zeroed-out" setting). RPM is static, TP is factor of a main global fuel trim setting (K number), airflow percentage, rpm, temperature, and throttle position acceleration (simplified). This theoretical pulsewidth will then pass through the interpolated map value (which trims to a different final A/F).
So, the TP has more purpose than being the "Load" axis of the fuel and ignition table, but it works quite well for this task. TP reacts similarly to a pressure (map sensor) in nature, and the two can almost be directly correlated in linear fashion as far as map axis is concerned. The higher the boost the higher the TP (load) value, so it really isnt that difficult. The only problem is that any time changes are made to the global trim settings > TP to provide a different injector pulsewidth (maf change, injector change, more boost) The TP axis scales must be adjusted so that they lookup the proper cells on the map with the (generally smaller) calculated TP.
**EDIT**^this paragraph is generalized to make it easier to understand, see below regarding different turbo chargers with same boost level.
This is really simplified and condensed, you might want to do some more reading on nissans TP (theoretical pulsewidth). We have actual nissan patent information to reference for this calculation but it is more than I could fit in here.
MarkEmark wrote:Does the ECU at all change timing due to overly rich or overly lean conditions? (Of course, the stock ECU with a stock KA engine would not need to deal with anything but minor, insubstantial fluctuations from what is considered the ideal A/F)
Because rich conditions don't need as much timing advance because the mixture burns faster, does the ECU retard timing automatically if there's a really rich mixture; by the same token, will it retard timing if the conditions are overly lean?
How does the ECU function in conjunction with the knock sensor? I know that it takes readings from the knock sensor and automatically figures out what is too much knock, and if this value is reached, the ECU retards timing...but again, in WHAT manner does it do so if too much knock is detected? Is there some absolute value that once reached, the ECU quickly and abruptly retards timing, or would it gradually and progressively retard timing as the knock got more intense? From your description, it seems like it does not do the latter, but rather the former...
The ecu does have some "feedback control" using the factory narrowband o2 sensor. This is used after a certain engine temperature and only in a certain (low load, cruise, idle) portion of the fuel table. It is unknown how much this closed loop function can actually alter the fuel maps/fueling, but it is hardly anything substantial.
It can be removed if desired, or even set to encompass more of the fuel map. The targetted A/F for these closed loop areas is also adjustable.
The Knock function is almost similar to the feedback control (closed loop) function, except that it does not actually modify the value from the timing table. After a certain amount of times that the ecu has passed its knock listening threshold, the ecu will actually revert to a completely different fuel and timing table altogether. The areas in the timing table that are prone to knock, are similar to the closed loop fuel areas (specialized offset), and both the main and the "knock" timing table will have these areas to tell the ecu to listen for knock. In order for the ecu to switch back to the main fuel/ignition table from the knock table, it must meet a certain amount of no-knock threshold samples within this knock prone area.
MarkEmark wrote:Also, from your description it seems like the ECU can adjust the A/F ratio based on boost pressure (or vacuum)...If this is the case, why does JWT not care how much boost you're running when they program your ECU? For example, my ECU is programmed for 50 lb/hr fuel injectors, cobra MAFS, but they didn't care what boost I was running. So, from the standpoint of a company who doesn't want their customer's engines to blow up, do they assume that the max amount of boost and power will be run for a KA-T with 50 lb fuel injectors, e.g., 15-17 psi and 350 bhp? If the timing maps are made such that timing is retarded as load increases, wouldn't JWT pretty much be guessing about the load at some RPMS? For example, they could be retarding the timing too little at 3000 rpm if someone's running a small turbo that hits a full 16 psi @3000 rpm; similarly, they could be retarding the timing too much at 3000 rpm if someone is running a larger turbo that doesn't hit full boost till 4000 rpm...
It is generally accepted that timing advance should be dropped in accordance to boost (psi), this makes it easy to discuss and understand. The nissan ecu uses calculated load similarly to other maf based systems. Believe it or not, calculated load via airflow calculation is actually much more accurate when determining the engines fuel and timing needs than a pressure reading. The calculated load will take into account the actual flow entering the engine (turbocharger based, in this case) rather than the boost psi it is providing, and thus two different turbos running the same boost level would actually place them differently on the tables as they should be. Certainly there are limits to to how much boost/airflow the ignition/timing maps can handle, I assume JWT looks at your general needs based your your basic specs, and provides something that should cover your needs. It is quite simple to adjust the settings for additional capability while at the same time suiting the customers possible lower-level requirement.