Thank you, Steve.

I am referring to the AFV, not AFR directly. As in ±5%, ±30%, etc. The total range of adapting the fuel values in terms of fuel delivery modification (in percent) derived from the feedback loop.

Fair enough.

Not really asking for your opinions of them, or instructions for developing maps for them, per se. (By way of reminder/information, they will alter the fuel in rpm/tps cells either plus or minus, I forget how many percent, as well as advance/retard the timing in those same-gridded cells, and somehow they can raise the rev-limiter to 6200 or so, but it puzzles me how because the only devices it actually gets between from the ECU is the injectors. The PC-III gets parked between the ECU and the harness using OEM connectors. As I said, the only wires the ECU loses direct connection with are those to the injectors. ).

Anyway, what I had wanted to hear your opinion on were the statements (1) regarding use of the O2 eliminator (a 1 M-ohm resistor in place of the sensor) and it effectively locking in the present AFV. And (2) the necessity of having to have the same AFV on the target bike as the value it was on the development bike. I will try to explain my conclusions and why I have reached them.

My first assumption is that the dummy load on the O2 circuit will lead the Delphi to believe all is well with the current fueling calculations, therefore it will not modify the AFVs from where they were at the time the swap was made (off-line).

If that indeed happens as a result, and on the bike for which the map was developed on the dyno the AFV was, say, 115% at the time, and on the bike on which a PC was installed with that same map (having the same intake/exhaust components now installed) the AFV was currently, say 95%, then it would result in 20% less fuel being delivered across-the-board on the new bike than was being delivered on the development bike. Conversely, the fueling could wind up being a fixed percentage greater on the second bike if the AFVs between the two were reversed in relative values. Does that make sense? Is it indeed a sound premise? Remember, the PC-III merely increases or decreases the intercepted injector pulse-widths by fixed percentages as directed in its fueling table. (I am sure added latency is an issue in this regard, but will leave that discussion for another time, if ever)

The reason I ask your opinion on all this is that in practice for me it sure seems to be the case. Remember, I have not data-logged or otherwise used any other instrumentation in arriving at this conclusion... I had been running just slip-on mufflers for some time before obtaining a pre-mapped PC-III and replacement airbox assembly. The installed PC-III map was developed on a dyno using Dynojet software using the exact same aftermarket components (well, same model parts, certainly there are some production tolerances involved). When I first installed the rest of the gear, the bike ran well rich. It would not even pull past 90 in 5th gear. It was not blackening the plugs or anything, but it was obviously not very ideal.

After some study into this AFV stuff (which I mostly found pertaining to Buell gear, initially) I reverted all the running equipment back to stock configuration and rode the bike for about 100 miles on a Saturday morning jaunt under many and varied operating conditions (during which it ran progressively better than at the start of the trip). When I returned home I put all the aftermarket stuff back into place at once and it was very much better than before. Certainly much, much closer. Better, cleaner power. Better mileage. Would pull nicely in 5th up into the triple digits and was still heading upward when I would let off. Plugs looked (look) good (not just a casual inspection, either, I mean really good), etc.

I have also performed some other tests at other times and in other ways to verify some suspicions, and all of this has led me to formulate my hypothesis, which is overviewed in that first post I referred you to. It has been some time since I have done any type of field-experimentation, though. For the past 7 or 8 thousand miles I have just been riding. I am not interested in developing any software or hardware to interface with any of this stuff. For me, it is merely academic. A stone I have turned over and am not satisfied fully-enough yet to just lay it down and move onto something else.

Am I close enough to be able to say what I say about it? Have you ever considered this type of scenario? If you are not comfortable discussing this openly like this I can understand. Let me know either way, won't you?

Again, Steve, I thank you for your time.

Glen

 

First off the Buell uses different ECM's from different suppliers so there is nothing that goes across the board. People are trying to make more out of this then there really is. The ECM has a fixed amount of learn cells. Those learn cells control the fuel adaptive value when the engine is in there range. These cells can and have been changed in size and location so nothing is going to be the same on all calibrations or work the same across the board. The values are applied anytime the motor is in that range no matter if it's closed or open loop. You need to deal with one bike and engine combination with one calibration. If you try to take it across the board and say they all work this way you’re going to be wrong in a lot of cases. I have not read your hypothesis nor do I have the time too. The adaptive values are applied to the VE part of the calculation for fuel but there are lots of other things that can and do affect the overall AFR. I'm not one to say go do this or go do that unless I know what the results are going to be and any time you start mixing parts and calibrations the results are going to vary.

 

That is fair enough of an answer to a lot of cold-cocked type of questions, I guess. But considering they could be boiled down (as I feel I have done) to 2 or 3 specific queries, I am at a loss to favorably treat your response. Here I had thought I have finally encountered someone who can tell me about this mysterious AFV in a definitive way. Am I wrong, or are you just plain unwilling for some reason?

I have a hard time understanding why you will not at least take a few minutes to read the post I linked you to and give me some feedback, good or bad, public or private. I would like to think that if some stranger came to me with such a specific question so thoroughly backed-up with specific possibilities to answer yes or no to, that I would be willing to at least devote, at the most, most of an hour to respond in kind. I gave you more than fair warning about how much time it would take to critically consider it. I have to truly and reasonably ask you if there is something you might think you have to hide, or whether there is something you feel might jeopardize your position in the community if you were to speak openly with me, or in private, about the subject of the AFV, and what it entails in the real-world for everyone who relies upon it for the well-being of their investment.

A lot of people you could ask about me might tell you that in everyday contact I am an a-hole, but a great many more would no doubt tell you that I am a really nice guy who truly likes to be of service to his fellow mankind.

I do not feel that I have asked you to spend an undue amount of time to critically consider and respond to what I have said. Yet somehow I get the feeling that you do not want to help me, somebody who has taken the time to clearly articulate a specific question they have, and one which you are all-evidence-aside able to answer definitively, to help come to a reasonably certain understanding regarding their specific question. Why is that, I wonder?? I cannot relate in any way, shape, or form to your reluctance to discuss this with me, even in light of my clear and open, multiple-times, request to take it private if necessary.

I apologize for taking as much of your precious time as you have already spent on me.

 

Boy Steve you sure make friends well....lol
&nbsp\\;
Glens....We have Steve hooked up and pressured to finish writing maps so the new TTS V-Tuner can be released in the next two weeks, when I see him on here answering questions I call him and ask if all the maps are all done....lol.....this is why he doesn't have the time right now,&nbsp\\;not that he's hiding anything from you.
&nbsp\\;
Now get back to work Steve....lol

 

I'm not trying to hack the ecm. My interest is in computer simulation. This is really just a big physics experiment to me. If I were tuning then things like latency and frequency wouldn't much matter. Just let it settle into a steady state. My interest is more the dynamics. There time matters a great deal. You have the ideal gas law, but what's ideal about it is that it's static conditions. That's hardly a running engine. Within a dynamic system resistance, current and potential matters. So that led me to the ve table.
&nbsp\\;
I bought the SEST mainly for it's logging capabilities. I'm a bit disappointed. Stand alone at 2hz for 30 minutes is pain in the rump. Fine for tuning, not for what I'm doing. Race Technology has a DL1 logger that records at 100hz onto an sd card. It connects to a CAN bus for interfacing to a ECM. So one question is can I get a cable to connect it. Apparently not. As near as I can tell the bike uses a J1850 bus. I don't know if that's PWM or VPW nor whether it's 10.4k or 41.6k baud, but it defintely isn't CAN.
&nbsp\\;
So I started wondering what's my alternatives. Is there anything in the automotive world that can connect given the appropriate cable? I figure 4hz would be adequate which the SEST can go connected to a laptop, but I don't really want to throw my laptop in a saddlebag. Generally heat and water don't go well with computers. Chances are if I find anything it won't be cheap. So I want to do what I can with what I have. So I'm trying to understand the limitations of what I have. I also want to understand what I have so I can search for alternatives.

 

LilBudyWizer
&nbsp\\;
This compamy may be worth talking to.
&nbsp\\;
http://www.scantool.net/products/pro...products_id=21
&nbsp\\;
On the fuel side of the equation the data buss speed issue is just not a huge issue on most things that you are looking for in tuning an EFI system.&nbsp\\; You&nbsp\\;just&nbsp\\;have to look at plenty of data.&nbsp\\; Patterns are patterns.&nbsp\\; The AE changes fast enough to at least be a challange but with a few check calcs on the side, it is fairly easy to find the data that needs to be tossed out as a CAN bus data speed issue.
&nbsp\\;
AW

 

LilBudyWizer

It's 10.4k VPW buss. The bigger problem is the overhead that is needed to run on the buss which just kills the speed.

glens

"The ECM has a fixed amount of learn cells. Those learn cells control the fuel adaptive value when the engine is in there range. These cells can and have been changed in size and location so nothing is going to be the same on all calibrations or work the same across the board. The values are applied anytime the motor is in that range no matter if it's closed or open loop."

"The adaptive values are applied to the VE part of the calculation for fuel "

You asked about adaptive fuel values and I gave you the above answer, that's how it works. I do not have the time to go read everything so I just let you know how it works and figure you were smart enough to take it from there. Sorry if it was short but that's what I currently have time for.

 

Steve,

I really want to discuss this with you. Perhaps when you get more time.

I do not (at this point in time, anyway, though no doubt that will happen eventually) really care about all the tiny little details or all of the possibilities. What I am interested in right now is really just the big picture, with things that can safely be talked about in general terms.

 

One question I have is about closed loop operation. Is 14.6 in the AFR table simply necessary or is it sufficient for closed loop operations? Is that the only condition that need be met or are their additional conditions such as RPM and MAP pressure? The reason I ask is the closed loop bias table is a subset of rpms and map pressures. So I'm wondering if that's because that's the only ranges where closed loop is possible?

 

14.6 is a switch...the ECM sees 14.6 and and looks for the info coming from the 02 sensors. If anything else other that 14.6 is in the AFR table it is out of closed loop.
The CLB table MAP and RPM grid change from calibration to calibration, these are the learning cells for that specific calibration...(this is how I understand it)