Tuning: afrBins2 table correct?

[mriulvr]

I have made an ECM Bluetooth connector and begun tuning my 04 XB12S using MegalogViewer. She has a Hawk can, K&N air filter, breather reroute, NGK plugs, and 04XB12S race fuel map. When I use MegalogViewer and look at the afrBins2 I have the following, which is the same as the afrBins1:

So, a couple of questions.
1. Does this look right and should the front and rear have the same afrBins?
2. From what I have been reading, the high RPM and high TPS cells look lower than what they should be. Should the 14.7 cells extend further into the higher RPM and TPS cells or is this what the afrBins2 for a race map should be?

Just want to make sure that as I tune I am analyzing my datalogs against the correct maps.

Just for background, after checking timing, TPS, and error codes my AFV was consistently running at 110% so I multiplied the fuel cells by 1.10, reset the AFV to 100%, and then took it for a ride and let it learn. Since then I have done two data logging runs of about 30 minutes and ran MagalogViewer VE. Each time I copied the rear map, pasted it to the front map, then added 2 to each cell in the front map. Runs smoother but idling has degraded a bit. I still have the stock NB O2 sensor and am considering getting a WB O2 sensor and a WB controller. Wondering if it is worth the time, energy, and resources to add an O2 sensor to the front for tuning purposes or if just converting the rear to a WB will suffice. How much of a difference will there be in terms of tuning with and without the front WB O2 sensor?

I know, several questions and topics here. I know there are many of you who know what you are doing and provide expert advice, which I greatly appreciate and have learned so much from. Sometimes it is hard to distinguish the wheat from the chafe on the forum. I have read so much conflicting information about converting TPS10bit to TPS8Bit, y-axis setting and VE settings in MLV, tuning, WB O2 sensors, etc. I am figuring out who to listen to, who to LISTEN to, and who to ignore among those who are active on here. I, for one, greatly appreciate those who know there stuff sharing their wisdom with the rest of us. [up] Love my Buell. Wanted one since a friend of mine in Colorado bought one in 1992. It was love at first sight. I want her give me many more years and miles of smiles and eventually add an 1190SX. Might be a present for myself when I make tenure.

 

[buellxb9rs]

Use vebins2 and afrbins2. Add an 02 bung up front for datalogging but unless you upgrade your electrical system and ecm you can't leave one in there and the stock ecm be able to use it. Yes to the wideband. It will make tubing much easier, faster, and give a much better tune. It's not absolutely necessary for tuning but it helps majorly. Also you can leave the wb in after tuning or you can swap back to the narrowband once the bike is tuned.

 

[theMelvster6]

I'm putting two extra o2 bungs in my header, one for the rear and one for the front right where it's easy to get to. That way I can leave the narrowband in place, but still tune with a wideband and be able to swap between front and rear easily. You can feed wide band voltage straight into the ecm through the NB connection, turn off the o2 sensor and standard corrections as normal and ecmspy will record the wideband data. Always tune the closed loop to 14.7 if you want to keep standard corrections once it's tuned. When you get the tune good, hook the NB sensor back up and you're good to go. An INNOVATE LC-1 WB kit is only like $150 on eBay....a little older but they are still one of the best.

 

[theMelvster6]

Good Info Megalog Tuning and Innovate LC-1 install also read the ecmspy tuning manual...the second version has a lot more info. Helps get a good grasp on how the ecm works and using ecmspy.

 

[mriulvr]

I've used ECMSPY for a couple of years and MegalogViewer for a month. Read everything I could find before using each. The thing I was wondering about was the AFV values in the afrBins2 file and if they looked right for a 2004 XV12S race map. From what I have read they looked lower in the upper RPM and TPS ranges. Just want to make sure that the afrBins2 file I am using to tune is correct.

The idling issue has improved with the last two tunes. One weird thing that happened was that VE analysis was giving me really high values in some of the cells, too high. They converted to 0's and 32's in the resulting rule map. I used the smoothing function of Megalog Viewer to get them more appropriate and less of a big change from adjacent cells. Haven't ridden with this map yet as it has been 102 degrees here all day. Definitely going to look into the Innovative LC-1 WB O2 and run on both front and rear cylinders. The tuning info for the front map seems like such a guess and makes a lot of assumptions about what will work.

 

[theMelvster6]

The webpage I posted about the LC-1 goes over exactly how to setup the software to tune with a wide band. In MLV you can adjust the AFRbins as you please, VEanalyzer just uses it as a reference to try to adjust fuel values to those AFRs. You could make your WOT AFRs 13.4 instead of 13.0. Make sure to adjust the whole AFR map so the transition is smooth between cells. I would keep the closed loop at 14.7. That gives you most efficiency at cruising throttle positions. It's just difficult to tune the open loop effectively with a narrowband, it'll always be somewhat of a guessing game. A lot of people do the +2 method for the front cylinder which gets you close but not exactly on. Also...you could change your cell change to hard in MLV, it requires more data to cause a change to the fuel cells.

 

[gbalias]

the benefit of using a WB is the ability to skew the ECMs interpretation of 14,7

iirc i have my bike running at around 14.2 at cruise and even leaner in the decel (16:1, etc)

runs smooth, cool, and i get about 45-47mpg. this is possible only by setting up the output of the WB to a certain voltage afr range.

im in houston where its hot as balls and 45115423% humidity.

it takes time to get it right and even more time to understand how the ecm treats the corrections, but you will probably want to adjust the analysis table based on mods, etc. i generally go through almost all data points within mlv and compare them to my afr expectations on my own prior to accepting the new table for the bike. i start the log where the engine temp is above 160 or so, then press the right arrow on the keyboard to determine the map cell location and note the afr. then you have to look to see if you were steady, accelerating, or on decel to see whethtr or not the desired afr is reasonable as well as the recorded afr.

anywho, i could go on and on about it. just do a few iterations and call it good. lol.

 

[theMelvster6]

^^^Well if you are going to listen to anyone about tuning it gbalias or xoptimizedrsx. A user with screen name by ICH might post on your tuning thread...he posts in German. He's an *******!

 

[ich]

aber er weiss wenigstens wovon er spricht. das macht den unterschied.

 

[mriulvr]

Ich habe tatsächlich genossen Lesen Sie Ihre Antworten auf einige der Tuning-Beiträge. Viel gelernt und es war unterhaltsam. [up]

 

[mriulvr]

Update. Changed the afrBins to match the values in the ECM tuning guide. I've also been thinking about the front and rear maps, since I don't yet have the ability to catalog the front. I've read several places to copy the tuned rear map to the front or copy the tuned rear to the front and add 2 or 3 to each cell. Doesn't seem to make any sense since the stock and race maps are not set up this way. I have been playing with the race map in excel and found the factor each rear cell has to be multiplied by to get the value in the corresponding cell on the front map. Would it make sense to take the tuned rear map and apply these factors to get a tuned front map?

 

[ReadyXB]

You will likely get varying advice on this. My opinion is, that would be better than blindly adding a fixed value to each cell, though of course still not ideal. You should be able to discern whether your driveability got better or worse as you ride it. Just don't change too many things at once.

Changed the afrBins to match the values in the ECM tuning guide.

Are you referring to this screenshot: "Figure 37 AFR targets used for project."?
If so, you may already know this, but that Default AFR map doesn't match any of the race or stock values I have seen for the Calibration Mode (Closed Loop Learn) region upper boundary. Starting at 5000/60 and higher loads, that AFR map declines from 14.7. But the default eeprom CLL upper boundary extends to 5000/90.

Unless you plan to abandon your O2 sensor after tuning, you want your afrbins to have 14.7 in the areas that encompass the Closed Loop Learn, and Closed Loop (CL) Idle regions in your eeprom image. Having smoothing on the borders of the CL regions is ok, but inside CLL, it needs to be 14.7.

Now if you're like me and don't spend much time cruising at that RPM, don't worry about it and leave it as-is. But I suppose some do which is why it's part of the AFV calculation. I'm mainly sharing this for education purposes since you seem to be in the thick of this and enjoying it!

 

[gbalias]

I have been playing with the race map in excel and found the factor each rear cell has to be multiplied by to get the value in the corresponding cell on the front map. Would it make sense to take the tuned rear map and apply these factors to get a tuned front map?

i created spreadsheets many years ago that automated this process. i believe Xopti made some minor changes and still has them on his site.

you plug in your map and it calculates the front for you. while this is not the BEST way to do it, its a safe assumption to use a ratio multiplier to determine a decent front map. I have put many miles on both methods and it doesnt feel really all that different.

im in the process of retuning now (although its not too frequent i mess with it) since i installed the lightened bottom end. there is quite a dramatic difference in the tuning now, but im getting close.

 

[mriulvr]

gbalias, I feel you about the heat but luckily the humidity isn't as bad up here in Fort Worth. I think I downloaded them but didn't realize they were for generating fuel maps for the front. Will be interesting to see how they compare to the one I did.

ReadyXB, I used figure 19 on page 31 of the V.2 tuning guide to adjust the afrBins.

I'm not on my computer with MLV or I would post the actual afrBin I am using but it matches the values in figure 19 for now. I think I am getting close with the tune as VE analyzer is changing fewer cells and the changes are becoming smaller. Once I get that done I am going to compare the front fuel map I came up with, the add 2 or 3 map, and one from the spreadsheet gbalias mentioned to see what differences I notice.

This has been very interesting. The race map for my 04XB12S was off by a good bit. This has been good practice for when I get the LC-1 ordered and installed so I can get a better tune on the rear and actually datalog and tune the front. Great stuff gents.

 

[ich]

Oh, ehrwürdiger Meister, bitte sagen Sie mir: Wie wollen Sie mit einem Standard-O2-Sensor im geschlossenen Regelkreis eine andere Mischung als die stöchiometrische einstellen? Bitte erleuchte mich, ich will nicht dumm sterben!

 

[mriulvr]

Ohne als herablassend, können die afrbins in MLV eingestellt werden und dann als Teil des Kraftstoff Karte gespeichert. Ich habe nicht behauptet, dass mit dem Standard-O2-Sensor, der die Datenaufzeichnung der Lage wäre, für alles andere als stöchiometrische einzustellen. Die afrBins, die mit dem Rennen der Karte waren, und der Standard afr in MLV nicht stöchiometrischen im geschlossenen Regelkreis, so dass ich sie auf 14,7 und in der Lage sein, die Vorteile der vollen Palette von sowohl Kraftstoff Karten nehmen, sobald ich ein breites Band Sensor. Haben Sie nicht genug als Kind umarmt? Ich mache viel gelesen und bin ziemlich früh in der Lernkurve, aber haben eine starke Mathematik Hintergrund, so viel davon macht einen Sinn. Wenn ich tatsächlich den Anspruch sind Sie schlug dann vielleicht würden Sie etwas angehen müssen, das ist eine kleine Klasse würde schmackhafter zu machen.[smirk]

 

[user_deleted]

how the hell is this ich bastard still allowed access to our site?

 

[ich]

Ohne als herablassend, können die afrbins in MLV eingestellt werden und dann als Teil des Kraftstoff Karte gespeichert.

Und? Was hilft das weiter?

Die afrBins, die mit dem Rennen der Karte waren, und der Standard afr in MLV nicht stöchiometrischen im geschlossenen Regelkreis, so dass ich sie auf 14,7 und in der Lage sein, die Vorteile der vollen Palette von sowohl Kraftstoff Karten nehmen, sobald ich ein breites Band Sensor.

??? Was soll das bedeuten? Dass man mit einem Breitbandsensor und im Open Loop jeden Wert einstellen kann? Das ist nicht neu. Warum dann die Frage nach der Closed Loop Region? In der IST das Gemisch IMMER stöchiometrisch. Unabhängig davon, welche Zahl in der map eingetragen ist.

Ich mache viel gelesen und bin ziemlich früh in der Lernkurve, aber haben eine starke Mathematik Hintergrund

Ah, das erklärt vieles. Ein Mathematiker also ... Wie fängt ein Mathematiker eine Herde Kühe? Er wickelt sich in Stacheldraht und definiert sich als "draussen". So viel dazu.

 

[ich]

how the hell is this ich bastard still allowed access to our site?

Warum ist es in diesem Forum erlaubt, so viel Scheisse zu schreiben? Von Leuten, die nicht den Hauch einer Ahnung von dem haben, worüber sie schreiben?

 

[theMelvster6]

See I told you...he's infamous!! Lol