Lowest Load value doesn't seem low enough

[ReadyXB]

I was reviewing a recent datalog session from TunerPro RT v5 and had a question about the Load value (0-255) which is used by the ECM for the Fuel Map lookup. For reference, this is the DDFI-3 "Load Rear" live parameter at offset 27.

Throughout a good 20 minute drive, and then idling afterwards, the Load never went below 17 (as nicely calculated by the Tunerpro "History Tables" as "Running Minimum"). Does that seem right?

My BUEOD Fuel Map Load Axis in the EEPROM defaults to "00, 06, 10, 15, 20, 25, 30..." (in decimal), and so on.
So if my Load never goes below 17, are the first FOUR entire rows in my Fuel Maps not being used? What am I missing?

In what cases would my Load value ever be 00 while the engine is running? Is that even possible?
I'm just wondering why the Buell engineers made a Load=0 row if it is never used, which I'm sure is not the case. So that makes me think something is wrong with my values.

Also, I noticed that newer ECMs such as BUEZD and later (which also have 16 bytes for the fuel map load axis) start their Load at a higher value:
For example:
BUEZD: 10, 15, 18, 20, ...
BUE1D: 10, 15, 18, 20, ...
BUE2D: 06, 10, 15, 18, ...
Note that I am only looking at XB12 data.

Any ideas? Just trying to understand. Knowledge is power, right?

 

[ich]

bueod ist scheisse und buggy.
upgraden auf bue2d.

 

[Negative]

Le sigh ....

Translation:

bueod is **** and buggy.
upgrade to bue2d.

Have a nice day ...

 

[s0dhi]

So, does that mean one should get the ECM upgraded to the latest firmware before doing any tuning?

BTW, I've never found a "finished" piece of software - ask any developer.

 

[ReadyXB]

bueod ist scheisse und buggy.
upgraden auf bue2d.

If true, that actually makes me feel a little better.

So, does that mean one should get the ECM upgraded to the latest firmware before doing any tuning?

BTW, I've never found a "finished" piece of software - ask any developer.

These are just my observations. Sample size of one. I would really really like to know what other people find. Just need some people to look at their live data "Load" values for various conditions, but mainly for idle at cold and fully warmed up states. What is the lowest value observed?

I agree - virtually all complex software (and hardware) has bugs. Was there a dealer firmware or EEPROM update for BUEOD? I need to check into that! There's enough user experience on this forum that someone else probably already knows the answer. Does H-D publish a Change History log for each ECM software update? I doubt they would air their dirty laundry, but you never know.

In any case, I wouldn't want anyone to take action based on my findings until they are confirmed. I could be just doing something dumb here. So this is mostly an academic exercise at the moment.

 

[Negative]

Explain to me these "Load values" you speak of please ... if possible ...

 

[ReadyXB]

I'll explain as best I know.
Computer-controller engines use two main tables to determine the appropriate amount of a) fuel to deliver, and b)spark advance to use, for a given operating condition. In basic terms, those operating conditions are a combination of RPM and engine load at a given instant in time. RPM is self-explanatory (well, IAC and idle does muddy the waters, but that's a different story). In absolute basic terms, Load is determined by throttle position (TPS, which Buells have), but a more accurate representation of what the engine sees as load can be found by evaluating throttle position in tangent with manifold pressure (MAP) or airflow (MAF), which is why most engines have a MAP/MAF sensor in addition to the TP sensor.

When looking at those fuel maps and timing tables in ECMSpy or Tunerpro, you are looking at a big 2D matrix of RPM to engine load. With 2 dimensions, there is an X and Y axis. Where points on each axis intersect is a value. For fuel maps, the value is a number for how much fuel is needed (in technical terms, it gets translated by the ECM into how many milliseconds to open the fuel injectors). For timing maps, the value is the spark advance in degrees. FYI, this is conceptualized as a 3D map in tuning software - what they're doing is making the "fuel" or "timing" value the 3rd dimension and adding colors.

While the engine is running, the ECM constantly performs this table lookup process for every cycle of the engine (ie, 2x crank turns). As RPM and/or Load increase, the amount of fuel and spark advance needed generally increases. You will see that reflected in the Fuel and timing maps.

Having said all that, the fuel and timing maps are just base/foundational values. Once grabbed, the ECM makes additions/subtractions to the base values depending on many factors, such as engine temperature, O2 sensor evaluation, air temp, etc. Which is what makes tuning so much fun If desired, you can actually turn all those do-dads off and force the ECM to only use the base values (ie, true "open loop") ).

Using your favorite Buell tuning software, you can see the runtime, dynamic load value being used by the ECM as you operate the bike. On Buells, the value is called "Load" and is an 8-bit value, so it ranges from 0 - 255. As mentioned above, that Load value is part of a lookup into the Fuel and Timing tables. My post meant that I had reviewed the historic Load value for a datalog session and I never saw Load go below 17, including when I was idling fully warmed up. I assumed I was doing something wrong or had a problem, so I wanted other people to report what THEIR load value is for different conditions.

Let me know if that doesn't make sense!

 

[Negative]

But these bikes have neither a MAP nor a MAF ... If you mean "load value" as in what cell the X/Y axis is pulling from on the timing/fuel maps I could see that as a viable value ... is this where the ECM determines it's load value? ...

 

[ReadyXB]

Load is an input number (y-axis, if you like) for looking up values in the fuel/timing tables at a given RPM. Our ECM almost certainly determines that Load value from the TPS alone. It's the only sensor we have that can be correlated to engine load at a given RPM.

 

[ich]

Our ECM almost certainly determines that Load value from the TPS alone.

stimmt nicht für ddfi-3.

 

[Negative]

Hmm ... well that doesn't make much sense to me ... as I imagined you'd need a reference signal for demand on the engine (TPS), a measurement of the supply of air for fueling (MAP/MAF), and a feedback signal (O2 sensor) to really determine the load of the engine in my mind ....

However, I am VERY much not a DDFI 3 specialist ... But I'm definitely interested in digging further into how DDFI 3 does it's magic ...

 

[ReadyXB]

stimmt nicht für ddfi-3.

ich, I would love to learn more about this. What is the ECM's additional -input- for determining load on DDFI-3?

 

[ReadyXB]

stimmt nicht für ddfi-3.

Aah, is it the MAP sensor on the 1125? I could easily see that as an input for determining Load, on the 1125. I keep forgetting that not everyone has an XB like me So i'll take a stab: on XB's, Load as an input for the Fuel tables is probably derived from TPS alone (my former statement, which is corroborated by the Tuning Guide V2) . On 1125's, Load is probably derived from TPS and/or MAP.
Does that sound right?

 

[ich]

nö. raten ist immer scheisse. messen ist besser.

 

[ReadyXB]

ich, unfortunately I don't understand German, and a google translation of your last response didn't make sense to me.

 

[ReadyXB]

ich, please comment. I said that our ECM almost certainly determines the Load value from the TPS alone. You replied "Not true for DDFI-3".

Why is it not true for DDFI-3?

 

[Negative]

Well ... the XB's lack the MAP sensor the 1125's have ... so you've got nothing to give inputs to the ECM except for the TPS itself ... these are still alpha/N systems with O2 correction ... DDFI 3 is just the next evolution of it ...

I could (with a stretch of the imagination) see the ECM taking into account the speed at which the TPS value changes, along with engine speed, to make up some kind of "load value" ... Because, if you crack the throttle open in a very quick motion and the engine revs start climbing to match ... it's under a load ... mayhaps that might be the source for this value for the ECM ...

Hmmmm says I ...

 

[ReadyXB]

Alpha-N by definition is TPS and RPM only. I have found no evidence that any generations of DDFI, by default, use any other method, including on the 1125.
Take a look at this diagram:

TPS and RPM and the ONLY values ever used as Fuel and Timing table lookups. I trust Nick Pashley as a credible source, but of course he could be wrong in some areas.

Even in the 1125 Service Manual, it clearly states that Load = TPS.
Section 6.1:
"- Engine load measurement via throttle position sensing"

And in section 6.3 "Throttle Position (TP) Sensor:
"The output of the sensor is a voltage, dependent on the position of the throttle plate, and used by the ECM to determine ignition timing and fuel required at any given RPM and engine load"

There is an EEPROM table called "MAP to Load Table MAP Axis", but I don't see any evidence of it being used. I checked BUEOD (feature not even available), BUEYD, BUEZD, BUE1D (XB9), and BUE2D EEPROMs and the feature "TPS and MAP Load Setup: Enable TPS/MAP load feature" is disabled, and the "MAP to Load Table MAP Axis" tables are empty. Additionally, when I suggested MAP as a factor for Load calculation on DDFI-3, ich said "nö. raten ist immer scheisse. messen ist besser" which I cannot decipher (though I do know what "no." means, in German). I have stayed up past 2am the last several nights trying to figure this out, all based on ich's statement that I am wrong for DDFI-3. Without more info, I don't see how MAP factors into this. I was just trying to give ich the benefit of the doubt. I would test this but I don't have an ECM that supports this "Enable TPS/MAP load" feature.

Our ECM does account for the increased fuel needs during acceleration! BUT, I would bet a cookie that fuel addition percentage is made AFTER the values are grabbed from the Fuel Maps . Unfortunately, where the ECM makes the Accel Enrichment is not documented in any text or diagram that I can find. My theory could be tested by making a throttle stop, say midways between throttle grip closed and open. Compare the max load value between the following cases while datalogging:
a) throttle very slowly opened to the stop, and
b) throttle quickly opened to the stop
For case b), if the max observed load value is greater than case a), then the ECM's algorithm for adding Accel Enrichment fuel is to temporarily increase the Load value. I just don't see that happening. Every other fuel adjustment is made after the TPS/RPM lookup.

 

[ReadyXB]

Fun stuff!

 

[EricZ]

While the DDFI-3 has MAP sensor tables, neither the XB or 1125 models ever utilized a MAP sensor from the factory. The 1125 models have a Barometric pressure sensor (same sensor but samples pressure before the throttle plates) but not a MAP sensor. I am slowly working on a DDFI-3 setup for my XB that will utilize the MAP sensor and TPS to determine load. It is quite a project, though.