Hi everybody, hope someone can:
We are currently reaching the end of our map scaling with our current boost settings 1.4bar on the boost gauge.
MAP sensor range: 416
MAP sensor offset: 5
MAP range unit: 2 (default)
Table kpa unit: 2 (high boost)
At the current offset atmospheric pressure is correct 102 hPa for where we are. As soon as I increase the range to 4 I cannot get the offset corrected even at its maximum negative value.
Any ideas? ???
It is bit unclear what you reached? Does your tables is construed just up to 140KPa? Then you can freely edit these bins!
The tables go to 360 kPa. The dash panel gauge redlines at 240kPa. We are want wanting to boost 200kPa +101.3kPa (barometric) so we should be around 301kPa. The problem is the vehicle is breaking down at the 240kPa mark. Hope I am making sense. Or am I missing something?
Are you mean gauge in Vemstune? It is not related to your engine cut.
There are several settings in configuration what could affect it e.g. overboost fuel cut.
Range of each gauge in VT is editable. But I hope you do not tune full boost by watching on gauges???
No tuning on gauges. I am using Megatune 2.25 and boost is manually controlled. Just checked and my boost controller is enabled. I will disable and check this. I have run logs and will post these where the breakdown occurs.
Don't post just the log, post the configuration file as well. Since you're running MegaTune, the log does not contain the configuration like it does with VemsTune.
Put the log and configuration file into zip archive and upload somewhere (real easy to send to a compressed archive using features built into the Windows explorer interface, just mark the two files, right click, send to...).
http://uploading.com/files/9e2mm335/datalog201011271148.xls/
http://uploading.com/files/32758cea/vemsv3.ini/
I hope this helps. Not so good with this stuff. Let me know if you need any additional info.
By configuration file, I meant MSQ file. The vemsv3.ini is what everyone has, it comes with MegaTune.
Judging by your datalog I'm surprised you haven't blown up the engine yet. I won't tell you how to tune, this is just my first reaction.
Your lambda target and actual lambda is very, very far away from each other. If you don't tune to make the actual reading match target, then your VE table will look very weird and the tuning won't make sense - they are both part of the fuel calculation.
I would advice you to make a complete upgrade to VemsTune and 1.1.x to get more control, but it might be confusing along the way (learn a new interface). A nice start would be to post your actual MSQ file and tell us which firmware version you're running.
If you're running an Audi S4 (an impression I got from another thread) then there are good known configuration files for the newer firmwares and VemsTune, and lots of other people here run these engines. Are you running a PnP ECU ?
I am running 1.1.27. When I tune I also use my PLX wideband as a secondry monitor and I'm not getting readings more than 13.5:1 on the AFR at light engine loads and about 12:1 under full acceleration. I was looking at those decrepencies last night and doing some research on getting the ve table right to help the ego side of things. This is the first time I am tuning with a closed loop system so I will take that into account now when I go out with it again.
I don't mind learning a new system if need be.....
There should be no difference in the readings between the two wideband systems unless one of them is badly calibrated, or if there is an exhaust leak at some point between the two.
Make sure you're using the calibration values for the wideband controller that comes with the ECU.
Free tips :
- Always datalog.
- The VE table is essentially very flat on these engines, don't assume you need lots more at higher pressures and rpm.
The new VemsTune that is under comm model testing is more stable and useful than ever before, but you might want to hold back if you don't like the word "testing". It outshines MegaTune in every respect so it's all good.
Test release
http://www.vems.hu/download/VemsTune/NIGHTLY/
For regular release:
http://www.vems.hu/download/VemsTune/
(only download and use the latest dated version, check back every week and ask on this forum if you experience a bug or wonder how stuff works)
Quote from: mattias on November 29, 2010, 01:44:13 PMIt outshines MegaTune in every respect so it's all good.
Outshine. But not in all categories. MT is faster - especially notable on older PCs below 1Ghz/1Gb. And much easer/faster/efficient to use in conjunction with MLV at table tuning stage and for datalog reviewing.
Sorry for offtopic!
Since you started it..
Given that you can't use late firmwares with MegaTune you are forced to use VemsTune, the only downside to you seems to be MLV support. Personally I do just fine with the built-in log viewer and VE tune by statistics, all which probably can be improved to suit everyones way of working, including yours.
CPU consumption has mostly to do with graphics acceleration, on the same IBM laptops from 2003 with Pentium 4 1.5 GHz processor, only different graphics chipset, the difference can be staggering. Make sure you have the correct driver and chipset. This mainly affects machines 6-7 years or older. Finding newer hardware than that is not difficult, people throw this stuff away.
For me built in logviewer seems unflexible/slow/unhandy. slow even on newer PCs (T61). And due to this I use older firmware for maping with MT. and after that upgrade to newer f/w. Vemstune is superior for config edditing, catching channel conflicts or similar mistakes, calibrating sensors and many other things as triggerlogs and firmware upgrading, importing separate things from other configs...
But it is clearly appreciable that developer guys did not used MLV in daily basis. Otherwise they has not to invent bicycle-alike things anew.
===
Nevertheless built in logviewer might be considered as above average or even good comparing to many other ECU GUIs. But we used Phil Tobin's stuff before. And he put the benchmark quite high.
Do you recommend I go for the latest firmware?
Quote from: GintsK on November 29, 2010, 05:59:04 PM
For me built in logviewer seems unflexible/slow/unhandy. slow even on newer PCs (T61).
Strange, I've never had an issue with VT on my T60 running W7.
It was always snappy.
I suspect my wideband config is not right. If I set the fueling at idle that it barely wants to idle due to the richness (10.4:1 on PLX) I only get a lambda of 0.85..... and my PLX is 600mm+- from the turbo and my Vems wideband is 200mm from the turbo. I'm just looking for my config file sent when I got the system.
Send a mail to info @ vems.hu and ask for the calibration numbers for your ECU (give them the serial nr).
Found them
> Board_version=v3.3
>
> Serial_nr= 2066
>
> pump-= 3.96V
>
> wbo2_pump_pw_zero=0x64
>
> wbo2_nernstdc_target=0x87
>
> pump+= +2.3mV/200 Ohm
>
> AREF=4.9V
>
> (C103=1nF, no D100)
>
Quote from: Dylan@AM Racing on December 02, 2010, 09:27:30 AM
Found them
> Board_version=v3.3
>
> Serial_nr= 2066
>
> pump-= 3.96V
>
> wbo2_pump_pw_zero=0x64
>
> wbo2_nernstdc_target=0x87
>
> pump+= +2.3mV/200 Ohm
>
> AREF=4.9V
>
> (C103=1nF, no D100)
>
It means pump zero PW =100
and nernst DC target = 135
all rest values - default! Check that!
except Sensor free air calibration. You have to perform it.