Guys,

Did anyone compare the starting tune (VE) and the last tune after quite some logs. The reason I am asking is that the more logs and tunes I make the closer I get to the VE's of the stock tune. In some spots the VE values are even getting lower then stock. So everything is getting leaner. I started with a tune from Jamie, that shoudl have put me close to the ballpark. But what is going on I have no clue.
Looking forward to hear some ideas from the guru's

 

Good point, as I don't think there is such an option. I'll send this suggestion to DJ.

 

ICLICK -- i tired it yesterday... if i change units from SAE to metric it seems to display in metric.. i will test it out and let u know if works ok.. but... weather here still sucks for riding.. would u believe snow is in forecast ...

 

Do you have a stock build? Would be good to see this info in your signature. What do you have your AFR set to in the area of concern?

 

I just got an email from DJ and he said to do that. Let me know if it works or not.

 

I agree with Lonewolf that your problem is in the timing tables. Your tables show very low advance in the problem area and the Knock Retard is pulling out even more. I suspect you have the same problem that I have with false-knock events.

Mine would show knock at ~55-60kpa @ ~2250-2500rpm only on the front cylinder, but not every time--perhaps once every 4-5 transitions. It would rarely be more than 2° but this would cause PV Tune to retard a bit for each tuning session, and after multiple sessions the timing in that area got lower and lower. No matter how low the advance was set, the false knock events would still happen.

I caught it before it started to cause performance issues and never got down as low as yours is at this time. I no longer let PV Tune adjust timing and if I make any changes I do so manually and never in that area. This is not a fault with PV Tune, IMO, as it is responding legitimately to knock events reported by the ECM. I asked Jamie about this at the time and he said false knock events do happen on some bikes, and you can only work around them. The knock retard at the ECM is slight, usually no more than 2° for only a few seconds, and I don't even notice the change.

If your original tune for the 255 cams was from FM or DJ I would put that area of the Spark Advance tables back to those values and avoid any changes PV Tune wants to make in the future there. For quite some time I've simply deselected the Spark Tables from being applied back to the tune in WinPV. OTOH I do check them in PV Tune and if I see anything that might need some adjustment outside that false-knock area I will make any changes manually.

At some point I'm hoping DJ allows us to select the areas of the VE and spark tables that we allow PV Tune to manipulate, as at this point it makes tuning more difficult. I have this false-knock issue plus another one that causes 2750-3500rpm at 0% to tune leaner and leaner with each passing tuning session. I must avoid that area too.

 

I am really starting to miss trust the narrow band sensors. I am seeing areas that seem stable over a few data logs then all the sudden after a few more logs areas really start to lean out the bike is running good I have no serious spark knock. I am wondering how much farther I want to let the VEs go south. I think the short story here is I dont trust all the data feed back now I have to figure out was is good data and waht is not.

 

I think Sam2010 is talking about the same thing. If you run a couple of logs, after flashing a new tune. It seems the VE's lean themselves out.

 

I honestly have not experienced this - yet. My new VEs will change a bit each collection run, but not consistent downward. I'll be curious as to the reason for this if uncovered. Might be good to detail the process that is being followed in case something in that is contributing to the problem.

 

My previous understanding, and not specific to the PV, was that once spark knock was detected, it may show up in a cell that it did not originate in. When log tuner makes a recommendation, do you also adjust the spark in the cells around that one? I try to manually smooth out the cells around the one that log tuner catches a spark knock in by viewing the graph table. Not saying this is working 100%, but the more I collect, the fewer event's I'm seeing. I'll have to wait until it warms up some to be sure.

Also, I tried a set of SE plugs as I had them in my tool box. After installing them, there was spark knock events showing up all over my map that I never expected. After replacing with stock plugs, things started to look more normal. I think this possibly speaks to the sensitivity of the ion sensing system. Not that other plugs would cause more pinging, but what the system detects seems to change dramatically.