KManager V1.4.5.7 Beta

[VT-Doo]

Hi,

Just back from a quick run. These are the errors I got within a 5 minute period...

Access violation at address 0047B256 in module 'KManager.exe'. Read of address 08FE4DF8.

Access violation at address 0047B42C in module 'KManager.exe'. Read of address 08E1BD38.

Access violation at address 00480664 in module 'KManager.exe'. Read of address 0907D1C4.

Access violation at address 0047B2E3 in module 'KManager.exe'. Read of address 0919A1D0.

Access violation at address 0047B289 in module 'KManager.exe'. Read of address 08A06870.

Access violation at address 0047B23A in module 'KManager.exe'. Read of address 08287248.

Access violation at address 0047B2E3 in module 'KManager.exe'. Read of address 087F7510.

I actually got these in the reverse order I've written them.

Using Windows 7 Starter Edition 32-bit. English UK.

KManager V1.4.5



Cheers,

[VT-Doo]

I have another problem.
Last week I bought a new Netbook and proceeded to install KManager V1.4.5 BETA. Since then I've noticed my Short Term Fuel Trim is way out causing a Long Term Fuel Trim. This is usually at '0%'. The Long Term is 'positive' at idle, 'negative' at low load, and OK at a higher load. My map has ALWAYS had a Short Term of +/- 2%

I thought my O2 sensor was on it's way out (no CEL though) so changed the Closed Loop corrections to +/-0% so the ECU would not make any changes to my map until I can get another sensor.

Once I did this I noticed I was running an air/fuel ratio perfectly at 14.6-14.8:1. Surely this reading should be off if the sensor was indeed broken? :?

I have a copy of V1.4.4 here to try and hopefully V1.4.3 too but not too sure.

I will let you know what happens when I resort back to an older version of KManager.

[VT-Doo]

I have another problem.
Last week I bought a new Netbook and proceeded to install KManager V1.4.5 BETA. Since then I've noticed my Short Term Fuel Trim is way out causing a Long Term Fuel Trim. This is usually at '0%'. The Long Term is 'positive' at idle, 'negative' at low load, and OK at a higher load. My map has ALWAYS had a Short Term of +/- 2%

I reinstalled V1.4.1 and as I expected the fuel trim went back to normal. It would seem V1.4.5 isn't seeing the right signal from the O2 sensor or is processing it wrongly giving out the wrong short term trim.


Hope all this is meaning something and is helping...


Thanks,

[Hondata]

Ok, folks, there is a new version available (see the first post in this thread for the download link).

I have worked on the isolating the regional settings, as the exception problem only happens to non-US locales. Additionally I have added an exception handler which collects more information and gives you the option to email it to us for troubleshooting.

VT-Doo: All the exceptions point at something being unhappy with the format strings in the controls on the display window. If it happens again then try closing the display window to confirm the problem; then you could send me the display.ini file.

I've made no changes to the closed loop operation, but I suspect that this might be related to the exception problem. If possible send me the calibration from 1.4.1 and then the calibration after it has been loaded and saved in 1.4.5 (or 1.4.5.1).

[Hondata]

Thank you for the reports which were emailed. I believe I have fixed the access violation problem - see the first post in this thread for the download link.

[Hondata]

It seems that there still is a problem. I've updated KManager to give more information about this - sanding the reports has helped a lot.

[eXceed]

Thanks for the updates, I will get on the road and put some miles on :). I can confirm that with 1.4.5 beta I was having a similar fuel trim issue to VT-DOO.
I will try 1.4.5.3 after work today!

[VT-Doo]

Thanks for the updates, will get it uploaded later tonight :)

[VT-Doo]

All the problems are still there. At idle it went straight to -11% S.Trim and 4 access violations appeared. I could forward them on as I didn't have access to the internet in my car.

I immediately uploaded V1.4.1 and all was fine, S.Trim returned to 0%.

I can forward on both calibrations if you wish?

[doms22]

PRA Problemes
1.4.1 closed loop , s trim OK
1.4.5.1 - 1.4.5.3 always open loop !! no correction s.trim !

I is not driving in open loop . I can not use the new features Knock control! Never ??

[Hondata]

Calibrations? Datalogs?

[doms22]

Hello

[dfws2k]

Is there any chance of a future release where the wideband input from the ELD can be used for the lamda targets? I understand that currently the ELD input is only for display and datalogging purposes only, but for almost all S2000 users, the current implementation limits the functionality of Kpro.

I realize that this may be a small subset of your user base, but it would fully unlock the Kpro for everyone. Not sure how much recoding would be required, but I believe it is at least worth taking a look at and letting us know the feasibility of it.

x2

also, what is the harm in running a dynotuned kpro'd s2000 without a wideband? is one necessary for engine safety?

[Corey872]

- serial number of your K-Pro: 2227
- KManager version: 1.4.5.3
- the calibration: attached
- a datalog of a problem: attached
- what sort of vehicle you have: 2002 RSX Type S
- what sort of ECU you are using: Stock PRB ECU

Main issue: I do not seem to see an 'interpolation' between the load points of the new Target AFR scales.

As I understand from the help file, the table labeled 'Target Lambda ECT Low' is for a load point around 32kPa while the 'Target Lambda ECT High' is for a load point around 82kPa. (Seems like better names might be 'Target Lambda MAP Low' and 'Target Lambda MAP High' ?) At any rate, a number is interpolated between these low/high points and the lower/richer number between the low/high interpolated number and target number used. (my understanding - may be wrong?)

I've attached a datalog running the car fully warmed up, I started at low MAP and made a few steps up to high MAP, pausing a few seconds at each. The results I generated:



At T=0:33 it goes open loop and we see I need to tune up my fuel table for summer E85!

In this instance the low MAP entry is 15.7, the high MAP entry is 14.7 with a target of 15.7, so I would expect the AFCMD to taper down toward 14.7 @ 82kPa?

Other questions:

The 'target lambda' only goes down to 123F coolant temp. The 'Low' and 'High' tables go to -24F. In the <123F calculation, how is the target lambda / low-high interpolation comparison made?

Is there any way to add gear compensation to this? 1-3 and possibly 4th gear are acceleration, the lean AFR is not really desired here. 5th and definitely 6th gear are high speed cruise - this is the best spot to lean out a bit for increased mileage.

[Hondata]

V1.4.5.4 Changes:
• Moved multiplexer options to a new tab.
• Fixed lambda tracing always showing on cam angle 0 table.
• Allow boost control to run below 100 kPa.
• Fixed PRA closed loop.

Corey - have not had a chance to look at the high/low load tables yet. The interesting thing is that I have run at 15.5 and 16:1 AF at cruise and found that 14.7:1 was best for mileage.

PRA target lambda will not go leaner than 14.7:1, since it does not use a wideband. I believe this was causing the PRA not to run in closed loop, so it should be fixed.