Myself and Emil have been to VEMS HQ the past week and helped bring VemsTune forward to a more useable state. Work includes some useful changes to 1.1.6x firmware like separate cranking and afterstart tables, improvements in injector staging and ALS and maybe most importantly the in-program "Press F1 for help" documentation actually covers most dialogs (to be forever continued I guess).
I think we are weeks from having a test release that is pretty much better than any combination of MegaTune and old firmwares that you've seen. It would be nice to hear from people later on if they have ANY reason to hang on to older firmwares and MegaTune, there shouldn't be any.
With guided information to upgrade to a stable .6X firmware from various older versions then I don´t see a reason to stick to anything old.
Knowing you two guys actually contributed information to documentation is very nice.
First: thank you for posting here!! :)
It is very welcome changes in firmwares! Many small and no so small changes since 1.1.27 bring VEMS level higher. Now it is possible smoothly run both: race and every day cars!
Last ones I liked - support of dizzy coil, fast acceleration. And now separated cranking/afterstart! And so many things work so well!
What depends Vemstune - there is not everything so smooth.
I use 1.1.53 with Megatune. Here is main reasons:
I am not able to adopt VT hotkeys like on MT. (coarse tuning: Ctrl+Shift-up/down)
Some 3D rotating problems. Especialy for spark map. Horizontal axis rotating is limited for some reason.
In VT arrow directions is coded against 3D window not graph itself (just checked big M... directions are linked to graph). It sits in my fingers: right arrow=more rpms; up arrow=higher load...
How to equal all selected cells let's say to 24?
But main reason is limited Mega Log Viewer support. It is very reasonable tuning tool: fast, simple, reliable...
Msq export still do not work.
Live datalog do not work. (Ctrl-T in MLV)
I did not find key to mark some event (e.g. knocking) during logging and later find this mark in MLV graphs. Converting to csv is slow. VT itself do not offer any reasonable .vemslog file analyzing. What in reality is reason to playback file and watching on 10 gauges? IMO it is historical thing from old DOS tuning programs. Now visual representation of huge amount of logged data in desired shape is the key. Let's say to plot knock value map in 3D based on average or max values. Or Ego correction depending on MAT. I know - it is so hard to create good software tool. But why then drop compatibility with MLV?
I know many dyno operators prefer manual tuning. But on VEMS VE tuning is just mechanical work - VE table really represent Volumetric Efficiency. Let computer do it automatically! Manual tune is necessity for lambda table, spark table and many constants and curves. Even on dyno I allow MLV to do the job - there are other important things for attention. Close to perfect Ego correction allow it!
Then output folders is some kind of mystery. VT saves files without asking where. And then it is hard to find where it is: documents, vems documents, C: , ProgramFiles...
I hope my chart isn't too depressing! And some of these things are solvable with ease.
EDIT: one more thing - MT in 3D graph by hitting Z gives top view of graph. Very usable when set table bins.
Thanks!
Gints
To all involved - thank you for your work!
I cannot stress enough how a quality software tuning/datalogging interface will help improve sales.
I have been holding back on using/offering many of the newest functions, because I cannot bring myself to use the VEMSTune program. I have had trouble with the lack of documentation, the strange nature of some of the functions (could not open a configuration file until a certain version?), the lack of simple things such as a default gauge layout, etc.
Regarding the Firnware - I have trouble understanding why every new version brings some new features, but also new problems, with existing functions that worked well before?
I would strongly prefer that the VEMSTune be made compatible with MLV (MegaLogViewer), in my opinion, this is one of the best aspects of the VEMS/MT ecu's.
Please keep us updated here of your progress.
I'd like to help if you are accepting volunteers. I can provide input based off of my experience with different ecu brands. Mattias, I will send you a PM about this.
Thanks,
Kevin
yep it´s all great if there just were some info on how to make the features work, meaning there is a total lack of info on how to set the new parameters,
I would like to try if the subaru trigger can work in some of the newer firmware, as the cranking trouble that was with it in the older firmware, but it´s not possible to set the reftooth table, as its seems that noone know what to fill the table with,
so i would suggest that there will be made a trigger dropdown menu, like there is now, but it needs to be the with more cars,
/Skassa
Aha so your the famous duo i've heard about!
I've been told about some the features to do with cranking and it sound very good
This will be a long post, but there is a lot to be said.
About upgrading from older firmwares. I'll get around to try this myself and maybe work around it or at least give you a guide. There are many new tables and some individual settings have been removed and some added. Luckily there are decent defaults for the new settings and tables - by new years I think there will be defaults for most stuff like general cranking tables for gasoline and E85/ethanol. The tricky part if you're upgrading from 1.0.x firmware is normally the trigger tooth table - it can be worked out with better documentation or defaults if the engine configuration is a common one.
Typical trigger setups can be contributed and made into default settings in a dropdown, there are already some like the BMW 60-2. Not sure about the Subaru trigger, but if anything it should work a lot better now as 1.1.x is superior especially for the trigger versatility with the most recent versions as of the past summer. I'll make sure to bring that up, I'll get news from HQ tomorrow.
Too many strange and undocumented settings is something I've wanted to remove, while that's not nearly resolved we got some things removed and those that are left are better documented. Some stuff is just plain obvious, things that normal users never really touch - like wideband PID settings, only the calibration is necessary in practice. I know that configuration handling has been greatly improved, some old 1.0.x are not supported in a good way but it warns about all mis-matches when you upgrade !
The in-program documentation will help immensly with some useability issues, now that there at least is not a total lack of it. The main two undocumented features are for the Idle PID controller and ALS, the first is Marcells job and the second is Fero (Hungarian installer). I was promised this would be on the agenda so I hope to see some results of that very soon and not when it's within minutes of an official release...
Without a doubt some stuff can still be clarified more, at least we've put up the main framework.
About useability..
How VT handles files might have been strange in the early stages, you now set a working directory. For each new box that connects with VT a new unique folder named by you is where all config and log files will be dumped, for anyone that wonders how? The serial number of the box is used along with a name of your choice.
Seeing the important data in one screen is important, me and Emil worked on a decent default gauge layout with 3D and 2D table views of the main stuff, the moving target and actual lambda gauge with a graph below is a real neat feature that makes datalogging less of a necessity in many situations. There is always a need to custom-make things that are very specific, like for idle or boost control tuning, and that is a big step away from how MegaTune works (or really didn't work at all).
The main operation in the 2D and 3D table is through QWER keys (big dec, small dec, small inc, big inc). If you want something different there is only Shift + arrows used to select bins that can collide. Ctrl and Alt are left to be used with arrow keys, a good complement to QWER - it should be added!
Other than that.. I personally always map spark in 2D, but each to his own I guess. I must see the number, not a 3D vector line that can be spun around.
Datalogging.
MegaLogViewer support is important as long as there isn't an internal function that is better. We brought this up already last week that it was not working, I have to check up on the progress. Either way the big thing which VT brings is the SIPR protocol and higher serial port speeds which will allow for higher datalogging frequency, at 19200 you're already at 20+ Hz. This could be improved a lot but there is a lot about how datalogging is handled that would require change to make this frequency even higher if anyone actually needs it. Some systems only datalog things like coolant temp a few times each second, while lambda is nice to have at a faster update frequency. By the way, you can always see the frequency in the status bar.
Aside from datalogging, but related, you will probably like the Real Time data acquisition system. It will compare to target lambda and in each VE bin suggest a correction depending on how many samples and deviation. I don't know how far it's come but there is definately progress and there was good code written by a member from the UK that is a member on this forum.
As far as manual/"mechanical" or automatic tuning I'm personally afraid to use EGO correction for anything but tuning the low load and cruising part of the VE table. You have to be pretty close to the final tune to be able to trust it and even then a mis-fire can wreak havoc given too much free control for the EGO correction algorithm. I usually run open loop and work out everything in real-time and when the load is higher it's all datalogging, and I'm pretty sure my methods will change once the Real Time data system is tested and functional as that will make VE tuning into a really quick procedure if you have a dyno load cell.
Until then, manual tuning is always needed. The 3D and 2D view of all the tables is very nice! You can do more than I've seen in 99% of tuning software. You can select all or parts of the table and do all of the following operations :
- increase/decrease (large + small)
- multiply/divide
- interpolate
- set value(s)
- lock/unlock cells
In addition, in 2D mode you can "paint" with a value by pressing shift and then moving around to other bins with the arrow keys.
You guys bring up many valid points, most are easily solvable and that was what I found to be the case during my visit - a short list of changes was given to the programmers everyday and progress was instant. I'm not sure on the order within the the priority list but the main problems that the programmers are working on right now are :
- datalogging
- real time statistics
- internal log viewing
- and simply removing developer crud and things normal users don't want.
There is very little project management to deal with development efforts in my mind, I hope that we find a way to channel the thoughts of everyone in a better way. The old VEMS Wiki is a joke in this respect but some of it's pages are still very valid for what's recent and actually working but the usefulness is kind of lost in the noise.
I am not able to adopt VT hotkeys like on MT. (coarse tuning: Ctrl+Shift-up/down)
- You would have to alter your hot keys to do this, as ctrl- is used for copying of data from cell to cell, and shift for highlighting a area on the table. All hotkeys can be edited, and there is a possibility to set up multiple profiles for this.
Some 3D rotating problems. Especialy for spark map. Horizontal axis rotating is limited for some reason.
In VT arrow directions is coded against 3D window not graph itself (just checked big M... directions are linked to graph). It sits in my fingers: right arrow=more rpms; up arrow=higher load...
- I have to check this, I usually use 3D table only for VE tuning, and 2D for the rest
How to equal all selected cells let's say to 24?
- Use shift-button to tag a square, and enter number.
But main reason is limited Mega Log Viewer support. It is very reasonable tuning tool: fast, simple, reliable...
Msq export still do not work.
Live datalog do not work. (Ctrl-T in MLV)
I did not find key to mark some event (e.g. knocking) during logging and later find this mark in MLV graphs. Converting to csv is slow.
- Have to test this as well, I do a lot of log converting and have never thought it was slow. never even tried the msq conversion, will return about that.
VT itself do not offer any reasonable .vemslog file analyzing. What in reality is reason to playback file and watching on 10 gauges? IMO it is historical thing from old DOS tuning programs.
- The automatic playback will be removed, but you will still have the option to play it back.
- There is a built in log viewer being developed at this moment, will probably be there in next released test version.
Now visual representation of huge amount of logged data in desired shape is the key. Let's say to plot knock value map in 3D based on average or max values. Or Ego correction depending on MAT. I know - it is so hard to create good software tool. But why then drop compatibility with MLV?
- There is a configurable statistics function being made. Currently can show a table of "Count", "Min", "Max", "Mean" and "Variance" in a table form.
- Is compatibility with MLV really dropped, or is it not finished yet?
- You are still able to use megatune with vems if you like that program more. Megatune is a bit limited on other things, so all settings can not be easily worked with. I have used megatune to do actual logging and vemstune for changes waiting for the log viewing / analysis of vemstune
I know many dyno operators prefer manual tuning. But on VEMS VE tuning is just mechanical work - VE table really represent Volumetric Efficiency. Let computer do it automatically! Manual tune is necessity for lambda table, spark table and many constants and curves. Even on dyno I allow MLV to do the job - there are other important things for attention. Close to perfect Ego correction allow it!
- There will be a builtin function similar to autronic's "Mixture Table" to start from
Then output folders is some kind of mystery. VT saves files without asking where. And then it is hard to find where it is: documents, vems documents, C: , ProgramFiles...
- You can choose any "Working Directory", where each ECU-directory is created, and logs / configs that you save will be stored by default in this ECU specific directory.
I hope my chart isn't too depressing! And some of these things are solvable with ease.
- definately not depressing, some of the things are already solved and some are in the works.
EDIT: one more thing - MT in 3D graph by hitting Z gives top view of graph. Very usable when set table bins.
- never thought about this use of that function, I have only been annoyed when accidentally pushed that button. I use the 2D table to set the bins. Right click on the RPM or KPA bins and you can "Insert" (actually steals highest bin from that axis and moves it to your chosen position), "Move" a bin (and automatically recalculates table values), and you can also "Edit" your bin (Changes number without doing recalculation), though the idea of seeing the table from top in 3D will be added to the list of functions.
//Emil
Well done and keep us updated on anything new regarding the new VT versions.
I too believe that a megalogviewer is needed as it helps do tuning so easy
as far as idle PID and idle tuning i have contacted Marcell and it has been posted on the vems wiki page about my thoughts esp. for the audi users.I have read and managed to fiddle around the PID settings a lot, giving me now a very stable idle.
if you want i can email you my thoughts on the matter and see what can help you
One thing that we could do here on the forum is actually post what we find with these mystery settings, this could then allow the findings to be digested and either pasted directly onto the wiki or sent to who ever is doing the documentation. I stopped developing any documentation a while ago because things kept changing without any notice - maybe if there is stability in the release procedure documentation wont be such a pointless, thankless task.
This forum could be by far the best tool to keep people updated on the status of new features/vemstune/etc.
About VEMSTune, it is getting a lot better. But unless you are a mind-reader, you would never know it!
Because if you go to the vemstune wiki page, http://www.vems.hu/wiki/index.php?page=VemsTune , it tells you there have been no updates since 8-28-2009.
For example, we could start a new thread today and sticky it, and title it: "VEMSTune Updates",
And if you logged in you could see about the new features like:
"WB and TPS Calibration Tools",
"Output Channel Visual", the improved layout, and more.
And also, it would be easy to find out things like, the datalogger does not work, etc.
You are right I couldn't understand why the page wasn't updated yet if you click the download link you see the new versions
http://www.vems.hu/download/v3gui/
Quote from: mattias on November 26, 2009, 05:45:29 AMSome stuff is just plain obvious, things that normal users never really touch - like wideband PID settings, only the calibration is necessary in practice. I know that configuration handling has been greatly improved, some old 1.0.x are not supported in a good way but it warns about all mis-matches when you upgrade !
My thougts about 1.0 an 1.1 branch - i think more and more functions, settings, constants are more and more self explaining oposite to 1.0.x where some constants was deep mystery - hard or even not possible to understand. But of course hitting F1 is much more nicely than digging in wiki to find nothing :)
QuoteHow VT handles files might have been strange in the early stages, you now set a working directory. For each new box that connects with VT a new unique folder named by you is where all config and log files will be dumped, for anyone that wonders how? The serial number of the box is used along with a name of your choice.
Is it changed during last two relises? I have folders with box serial number and given name on C: directly all empty. And same in My Documents\VEMS_Files\ Both are not my desired places where to store data. When I connect new board VT asks just for project name, but not location....
Quote
Seeing the important data in one screen is important, me and Emil worked on a decent default gauge layout with 3D and 2D table views of the main stuff, the moving target and actual lambda gauge with a graph below is a real neat feature that makes datalogging less of a necessity in many situations. There is always a need to custom-make things that are very specific, like for idle or boost control tuning, and that is a big step away from how MegaTune works (or really didn't work at all).
Can you also take care to not destroy previous settings when installing new version of VT. It happened time to time with older versions.
One more issue is screen resolution of my current laptop 1400x1050. I see 5 and 1/2 gauge in each row :) But it is not problem if my file is not overwrited during upgrade.
QuoteOther than that.. I personally always map spark in 2D, but each to his own I guess. I must see the number, not a 3D vector line that can be spun around.
This point is important: there are many succesive tuning strategies around. Some tuner feels comfortable just with numerical tables, some can't imagine life without 3D edditor (me), some use mouse w/o problems, some hate it and so on.. up to small but important details (later example). So sharing experiences and suggestions between users is urgent during creating of good s/w tool (it is why VEMS is successive - many experiences/knowledge in one resulting product). And this is the point where VT limped on both legs. Up to yesterday! :)
Datalogging.
QuoteMegaLogViewer support is important as long as there isn't an internal function that is better.
agree with that. But as I wrote, it is tool coded by deep talent (Phil Tobin). And it will be hard to overtake. It is very handy today. Are someone notice, how nice is average statsistics tool in bottom of graphs accesible by dragging mouse along datalog part?!
I recently tuned MBE unit http://www.sbdev.co.uk/Display.htm. Its datalog tool seems quite advanced. But some things killed me - perceptibility, migrating over datalog, numerical data in side window apart from graphs... I filled myself like with no fingers. Case when badly designed features kills simplicity of simple using.
QuoteWe brought this up already last week that it was not working, I have to check up on the progress. Either way the big thing which VT brings is the SIPR protocol and higher serial port speeds which will allow for higher datalogging frequency, at 19200 you're already at 20+ Hz. This could be improved a lot but there is a lot about how datalogging is handled that would require change to make this frequency even higher if anyone actually needs it. Some systems only datalog things like coolant temp a few times each second, while lambda is nice to have at a faster update frequency. By the way, you can always see the frequency in the status bar.
I have 24 samples per second in latest 1.1.53 datalog with MT, but lambda value changes just every ~4 samples. I think because of some filtration inside board. Most important datalog rate is for knock logging.
QuoteAs far as manual/"mechanical" or automatic tuning I'm personally afraid to use EGO correction for anything but tuning the low load and cruising part of the VE table. You have to be pretty close to the final tune to be able to trust it and even then a mis-fire can wreak havoc given too much free control for the EGO correction algorithm. I usually run open loop and work out everything in real-time and when the load is higher it's all datalogging, and I'm pretty sure my methods will change once the Real Time data system is tested and functional as that will make VE tuning into a really quick procedure if you have a dyno load cell.
When wrote
mechanical I tried to describe fact that VEMS have exact VE table for given engine. If it is tuned by hand or tuned by software - no difference in results. It has no any creative/artistic part where tuner care is needed.
Belive me, no any problems to use lambda correction over all loads/speeds during tuning. Yes it require some care for smoothness in strat map and driving. But it works wonderfully on VEMS. I use EgoC support even at 2.5bar boost! BTW MLV do not need EgoC switched on.
It is one significiant difference between real time automated tuning and on datlog analysis based. Second one have one important advantage. Very important! It do not change any cell during run. If manual or VE learning alike tool works in real time it cripple table around tuning point. It makes everything around this point unstable, especialy if entry table is far from output table. But when using datalog based kind - there is no such problem, because routine do not create problem for itself - table stays as is. And if nothing bother egocorection (accel/deccel) output is very clean.
In first look real time method seems faster. But it is oposite in fact. No matter - dyno or especialy road tuning is used. I respect my sentences about different strategies wroten higher: it would be nice to keep both routes by choice :)
Quote
You guys bring up many valid points, most are easily solvable and that was what I found to be the case during my visit - a short list of changes was given to the programmers everyday and progress was instant. I'm not sure on the order within the the priority list but the main problems that the programmers are working on right now are :
- datalogging
- real time statistics
- internal log viewing
- and simply removing developer crud and things normal users don't want.
From my point of view one of main points should be MLV comparability. It would tear out main reason why to use MT. And as result much higher user comunity.
Gints
Quote from: [email protected] on November 26, 2009, 06:01:32 PM
I am not able to adopt VT hotkeys like on MT. (coarse tuning: Ctrl+Shift-up/down)
- You would have to alter your hot keys to do this, as ctrl- is used for copying of data from cell to cell, and shift for highlighting a area on the table. All hotkeys can be edited, and there is a possibility to set up multiple profiles for this.
I thik problem is different: Ctrl+Shift-arrow change window hight of 3d edditor. It is hardcoded and IMHO unnecessary.
Here is my 3d hotkeys http://www.vems.hu/files/GintsK/VT/3DTableEditor_GintsK-0.zip
QuoteHow to equal all selected cells let's say to 24?
- Use shift-button to tag a square, and enter number.
It cahnges just one cell of marked area. Or I do something wrong. Mark desired area-> hit Enter -> write desired value -> hit Enter.= just one cell... (in numeric table)
QuoteBut main reason is limited Mega Log Viewer support. It is very reasonable tuning tool: fast, simple, reliable...
Msq export still do not work.
Live datalog do not work. (Ctrl-T in MLV)
I did not find key to mark some event (e.g. knocking) during logging and later find this mark in MLV graphs. Converting to csv is slow.
- Have to test this as well, I do a lot of log converting and have never thought it was slow. never even tried the msq conversion, will return about that.
It depends to what comparing. In MT/MLV variant open datalog (if it is not live and already running) take 2 seconds (otherwise 0 seconds), then saving msq and opening it take another 3 seconds. Then let's say 30 seconds for analysis (datalog size dependent), 10 seconds for backing up undesired areas, 5 seconds for export and opening loading new map in MT. About 50 seconds. It is pretty fast! Now try to repeat it on VT :) :). In overall 20 minutes for finished VE map!
No matter dyno or road!
QuoteNow visual representation of huge amount of logged data in desired shape is the key. Let's say to plot knock value map in 3D based on average or max values. Or Ego correction depending on MAT. I know - it is so hard to create good software tool. But why then drop compatibility with MLV?
- There is a configurable statistics function being made. Currently can show a table of "Count", "Min", "Max", "Mean" and "Variance" in a table form.
Ouch! Perfect! I will not sleep tonight! :)
Quote- Is compatibility with MLV really dropped, or is it not finished yet?
- You are still able to use megatune with vems if you like that program more. Megatune is a bit limited on other things, so all settings can not be easily worked with. I have used megatune to do actual logging and vemstune for changes waiting for the log viewing / analysis of vemstune
Here is another problem: no Megatune for newest versions. And VT now seems promising in many areas.
QuoteThen output folders is some kind of mystery. VT saves files without asking where. And then it is hard to find where it is: documents, vems documents, C: , ProgramFiles...
- You can choose any "Working Directory", where each ECU-directory is created, and logs / configs that you save will be stored by default in this ECU specific directory.
I will try to sort it out!
QuoteEDIT: one more thing - MT in 3D graph by hitting Z gives top view of graph. Very usable when set table bins.
- never thought about this use of that function, I have only been annoyed when accidentally pushed that button. I use the 2D table to set the bins.
Different tuners, different maners. Personaly it is hard for me find too close or too far axis on numerical table, but I know someone without such perception problem.
QuoteRight click on the RPM or KPA bins and you can "Insert" (actually steals highest bin from that axis and moves it to your chosen position), "Move" a bin (and automatically recalculates table values), and you can also "Edit" your bin (Changes number without doing recalculation), though the idea of seeing the table from top in 3D will be added to the list of functions.
It is what we missed. Just one suggestion: it wold be nice possibility if "move" allows move axle over others (1200->4600). It is needed more often than trash out last axle.
Thank you guys for your push and writing here. Beneficial!
Gints
Latest firmware is out, but no separate cranking and afterstart as said on the Wiki.
Is this a .ini file problem?
(http://myndasafn.bmwkraftur.is/d/79563-1/noseparatecrankafterstart.jpg)
As long as Vemstune has compatability problems with (eg Prolific) USB to serial converters it cannot, in my view, be considered even as an equal to the old, clunky and above all RELIABLE Megatune.
I simply don't care how fancy (and good) the interface and innovative tuning helpers have got if it fails at square one - simply talking reliably to the ECU.
I really do like the features and tools in VEMS tune, but with reliability and connectivity issues, it's a non-starter for me unless I have an ECU on my bench-tester connected to a desktop with a real serial port.
Fix that, and I will delete Megatune from my laptop with pleasure!
Quote from: gunni on November 27, 2009, 03:19:49 PM
Latest firmware is out, but no separate cranking and afterstart as said on the Wiki.
Is this a .ini file problem?
Use todays VT beta build, it does have files for 1.1.63.
I really like the "arrow" buttons at the bottom of each tuning window!
With Autronic, this would be done using the "Page Up"/"Page Down" keys. Very useful!
I was going to suggest, can we get the "Review Mode" back from MT? It allows quick browsing thru the entire calibration menu.
Are we now limited to 4x4 tables for the cranking and afterstart enrichments?
I would place a "strong wish" on the following:
- Simple WheelSpeed vs. Boost Duty (open loop)
- Injector Dead Time Table
- Individual Injector Trims
- MLV Compatability
Suggestion - you could put a setting for "Beginner Mode" and "Advanced Mode" - the "Advanced Mode" would have all of the latest tricks.
You could lock out the "Advanced Mode" with a password as well.
This would be an alternative to having 5 versions of basically the same software / ini files
Than you for the updates!
Kevin
Quote from: lugnuts on November 28, 2009, 12:29:25 AM
I really like the "arrow" buttons at the bottom of each tuning window!
With Autronic, this would be done using the "Page Up"/"Page Down" keys. Very useful!
VT have same hotkey as Firefox migrating between tabs: Alt-Tab. IMO better than buttons for mouse.
Gints
I would like this thread to focus on things that DON'T WORK, rather than "I want this feature, I want that feature".
Things like having functional datalogging (internal or MLV), and trouble-free serial port communication are the two biggest issues to me right now.
Firmware features can be discussed in another thread. Some of Andreys work should be looked at and integrated, I'm just not sure of how to make that happen - hopefully Marcell is reading this. But like I said, that should wait until the tuning software actually works to most peoples satisfaction.
New Beta version is out, if you install it into a different directory from default and then in vemstunes set your default directory it shows your default directory greyed out
If you save a file it uses the default setting
If you open a file it always goes to the folder defaultfirmwareconfigs directory instead of the set default
One more thing in VT seems not work:
graphical editor for 2d graphs for warmup enrichment and others.
Another, maybe not so important:
Key shortcuts to dialogs.
Like Alt+T-A - acceleration enricment.
Some of this work, but most part not because of same underlined letters used or no underlined letters at all.
Quote from: mattias on November 28, 2009, 05:28:00 PM
I would like this thread to focus on things that DON'T WORK, rather than "I want this feature, I want that feature".
Things like having functional datalogging (internal or MLV), and trouble-free serial port communication are the two biggest issues to me right now.
- OK I agree totally. I am 100% for having MLV compatibility as soon as possible.
- Probably 50% of my sales would not have happened without this excellent data logging/graphing ability.
- Bugs:
- The new VemsTune (11-27-09) hangs up when I got to close it out about 50%-75% of the time
- Several of the tuning windows open up to a very small size and must be expanded to see the settings.
<<<Firmware features can be discussed in another thread. Some of Andreys work should be looked at and integrated, I'm just not sure of how to make that happen - hopefully Marcell is reading this. But like I said, that should wait until the tuning software actually works to most peoples satisfaction.>>>
- OK, thanks.
Ther appears to be no way out of full screen mode while in full screen mode and I cannot change it back even if I shut VT down though program manager
Just re installed VT and found that Shift-F11 toggles between, It may be usefull to have Escape as an exit from full screen mode as if like me, you never made note of the short cut ??? ::)
I would like to see a 36-1 default for primary trigger settings :D
And, I would also like to see the return of the Expensive Misc outputs. These will give SO much more versatility, and satisfy evaporative emissions purging systems ;)
Quote from: Sprocket on November 29, 2009, 12:12:01 AM
Ther appears to be no way out of full screen mode while in full screen mode and I cannot change it back even if I shut VT down though program manager
Just re installed VT and found that Shift-F11 toggles between, It may be usefull to have Escape as an exit from full screen mode as if like me, you never made note of the short cut ??? ::)
LOL I did that when I 1st tried Vemstune, for me to exit should be ctrl+esc or alt+enter which is the universel default for exiting full screen.
Quote from: Jamo on November 29, 2009, 01:39:03 AM
Quote from: Sprocket on November 29, 2009, 12:12:01 AM
Ther appears to be no way out of full screen mode while in full screen mode and I cannot change it back even if I shut VT down though program manager
Just re installed VT and found that Shift-F11 toggles between, It may be usefull to have Escape as an exit from full screen mode as if like me, you never made note of the short cut ??? ::)
LOL I did that when I 1st tried Vemstune, for me to exit should be ctrl+esc or alt+enter which is the universel default for exiting full screen.
You learn something every day :) I only have ever used Esc to exit anything in full screen mode ::)
there is an new version of vems tune online:
http://vems.hu/download/v3gui/VemsTune-Install-2009-12-04-testing.exe
does anybody know the changelog and whats new?
I can try to write down some changes since last one:
- Help text for most settings dialogs (F1 when showing a settings dialog)
- moved some settings to more logical placement
- Shows current ECU or open file in status bar
- Change multiple tagged cells in 2d table by entering a new number
- View and Edit mode for gauge groups, edit mode is to move around and add/remove gauges.. still shift-alt + arrows is needed to change size of gauges
- Doesnt play back log files automaticcaly when loading
- Test version of integrated log viewer, hidden in gauge group Dev2 (hotkey shift-8)
- Experimental "VE Tune by statistics" in tools menu, not good yet but sort of shows the idea, key "T" might be good to know when playing around in the table
- Drag and drop config files, now not only for tables but for all settings dialogs. Just drop a config file on any settings window and only the data in the dialog is imported
- Changed serial port code, may work better with bad usb-rs232 adapters, please report if someone that had problems before has more luck with this version.
Probably missed a lot, but this is at least something
With this version, also support for firmware 1.1.63, which has these important changes
- linear Analog input calibration for showing in logs / vemstune, gauges are still accidentally called MCP3208 channel 0,1,2,5,6 and 7. Channel 3+4 is EGT inputs, and not shown here because of that
- Cranking and afterstart is now split into 2 separate, but smaller tables
- Acceleration enrichment dotrate is now shown in the curve
- "Flags" available for N2O output and Injector stage2
- Warning light
- "expensive" misc outputs enabled
- Separate idle threshold for ALS active
Issues I found so far:
- VE tune by statistics is heavily under construction and crash the program easily, also does not have very good statistics yet.
- Shows a somewhat unusable gauge group on startup
- Bad size of log viewer, at least in 4:3 screens, you'll need to resize in edit mode with shift-alt+arrows, remember to right click and save
- Log viewer does not automatically load new files, you'll need to close it and reopen (shift-8 + shift-8 for example) to see new file
Just installed it and the help files are 100% better, top job there
I tried this version and my serial problems are still that exact same.
Quote from: gunni on December 06, 2009, 05:32:19 AM
I tried this version and my serial problems are still that exact same.
It would be nice if you could describe your problems,
I have seen two types of problem
- Cannot connect at all
- Work with engine off, computer hangs with engine running
The connection problem has been fixed with latest prolific driver and possibly by turning off "AIM enable" in Display settings
I haven't been able to reproduce the hanging computer with my own, but with another guys laptop and usb-serial converter placed very close to the ECU
The exact problem is
The ecu will become disconnected anywhere between 20second up til 5minutes from connecting, inbetween that time everything is fine.
There is no pattern to when it disconnects.
Second problem is corruption of the data coming from the ecu, sensor values and readings are totally crazy and random, tables will look corrupted and out of context, while a LCD will show the values reported to the LCD to be just as expected.
This can happen anywhere between 20seconds up until re detecting the ecu. If I re detec the ecu everything is back to normal for a little while until it starts again.
I have zero problems with firmware updating so it seems to be related to information coming from the ecu data stream
What version of your prolific driver do you have?
I have left my computer (core2duo lenovo laptop) on for 24 hours connected to ecu without problems, both with real rs232 and usb.. This didnt work on older drivers
I was even able to reproduce this corrupted data problem using terminal program with old prolific drivers
Thing is as well .
That I tried the same ecu and usb adapter on another laptop,
There where no problems on that.
My driver version is 3.2.0.0
Just updated the driver, will report back. version is now 3.3.5.122
Just tried with my Prolific USB device, and it still doesn't work at all - what version of drivers did you test with Emil? I'll try those.
It still mostly works with my FTDI usb device, but it will get itself into a cycle where it connects and instantly drops sometimes. There seems to be no reason why - engine running or not makes no difference.
dnb, do you have AIM protocol enabled?
it seems driver or vemstune gets a bit dizzy about this stream of data coming before the port is opened the first time
If you have AIM enabled, try to connect with any terminal program and send "A", then, without restarting ecu try to connect with vemstune again
Once connected, try disable AIM and try to connect to a restarted ecu again
I can reproduce this behaviour here with prolific cable and fw 1.1.63 with AIM enabled
works every time with AIM disabled
//Emil
Quote from: gunni on December 07, 2009, 11:54:07 PM
Thing is as well .
That I tried the same ecu and usb adapter on another laptop,
There where no problems on that.
My driver version is 3.2.0.0
Just updated the driver, will report back. version is now 3.3.5.122
the 3.3.5.122 just crashes my laptop and does not at all find a ecu .
So back to the 3.2.0.0 and it connects , but AIM is disabled on this ecu
sorry, forgot, im running XP with driver 2.0.13.130
new driver package was released 2009-12-02
Can you post that up for download so I can test that?
Thanks.
http://www.prolific.com.tw/support/files//IO%20Cable/PL-2303/Drivers%20-%20Generic/Windows/allinone/PL2303_Prolific_DriverInstaller_v110.zip
I tried to find a changelog for the drivers, but failed..
Same problem.
No changes with the prolific drivers you posted.
And it crashes VT.
The package you linked to is driver 3.3.10.140 for Win7.
So maybe it´s my windows that´s at fault here.
Here it only hangs if i leave AIM protocol enabled and using USB adapter
Workaround for this is to start vemstune with correct port configure (com3 in my case), but with serial cable unplugged
Then, after vemstune said "ECU has not been detected on port: COM3", plug the serial cable into my usb adapter
for windows 7, prolific say that windows version 7100 or newer has to be used
tried to run vemstune in administrator mode?
I can´t do anything if it doesn´t run in administrator mode.
I´ve just installed VT on my just up and running XP desktop and will see how long that lives, that has serial port as standard even though it´s a brand new type motherboard(can you belive it ) . I don´t think it´s going to be a problem on XP.
I´m going to be swithing over to a XP laptop real soon and this will not really be a problem for me after that. As VT worked on that one.
But it will require sorting as Win7 is fast replacing XP and Vista for people
Im leaving vt running tonight online over usb, to see what happens in the morning
So far the only difference i can see is that update frequency is ~1hz lower (21..22hz instead of 22..23)
The desktop is already staying alive and working 100% as I would have expected.
So first try on the XP desktop is running 15minutes without a hitch.
I will also try to install the USB adapter and test it on my XP machine, and report back.
Maybe this is really a Vista / Win7 problem
EDIT Update.
I´m now using my desktop and the usb adapter with the drivers Emil provided.
And there are no signs at all of any problems , already running 10minutes which is by far better then using Windows 7.
So it´s not VT and not the USB adapter but Windows 7 that is causing the problems on my laptop.
Some weeks ago wiki suggested some new-tested prolific driver. I tried it and trash out it - I had a lot more problems than with old driver. You can find my post in wiki. Seems I have no connection issues with VT (not tested on last VT).
if i try to run Firmware web tool, VT chrash every time
/Skassa
Works for me and it's found version 1.1.64 for download!!!! what the :o
anybody got any details on 1.1.64? changelog does not reveal anything.
I tried the new VT last night. It's no better at connecting than before.
I really fail to see why it's such a problem. Megatune works just fine. My GPS software works just fine. Even my old Subaru data logger (written in 2001 when I had a real serial port) works just fine.
The only thing that doesn't work is VemsTune. I must therefore conclude that the problem lies within VemsTune and NOT within my USB to serial devices. Changing drivers and generally messing around with computer settings is not a good solution - especially when no other software I use seems to need it - so please either explain why it's so hard and why VemsTune is so revolutionary in its use of the serial port that I need new drivers for a device which to my mind works perfectly well. Or even better, just fix Vemstune!
Sorry for the rant - I'm getting angry about VemsTune now that it's costing me customers and making support very difficult.
I agree, it's ridiculous some of the things I have seen with the VEMS software, firmware, and hardware.
I miss the Autronic days where the thing just worked, 100% of the time, and that was that.
Example 1:
Firmware "updates" often have 5 new features, but they also seem to have 5 new bugs!???
Why can't the "old" be left well alone, and the "new" just added and tested?
If a customer were to read the "Firmware Updates" notes/comments on the wiki - they would be hard pressed to fine one that would assure any confidence at all.
Example 2:
VEMSTune is released with nothing but bugs, software crashes, some people have connection issues, at first it did not have offline editing (useless), does not work with MegaLogViewer, so the datalogging is useless, etc.
Just stop with the "updates", and release something that works!!!
Example 3:
Hardware - Why is it even an option or an afterthought that the speed input is protected for over 5 volts, or the analog inputs are not protected from 12v ruining the mcp3208 chip, etc.
It's a support nightmare trying to fix all of these problems. Customers make mistakes. If I can spend $0.50, $5.00, or even $50.00 in hardware to prevent a problem from happening, please give me that option!
I wonder what kinds of backgrounds the developers have.
What other interfaces have they used before?
What are they used to using, that makes VEMSTune seem like something useable?
What computers and OS do they use? Since they don't seem to have any problems with VEMSTune?
It's just that....working with VEMS, I often get the feeling that I am on a different planet.
I'm beginning to wonder what i've let myself in for :D
Is there no way to get megatune to work with the new firmwares as a workaround for users?
I have Megatune working (but not exhaustively tested yet - only I have used it in anger) with 1.1.63 so don't worry too much!
Any details of how you've done this?
I just edited the vemsV3.ini file to match the header (*.h) file for the 1.1.63 firmware. All the data except the instructions is supplied with the firmware.
Send me an email and I'll reply attaching the file. It's not guaranteed to be perfect, but I am fairly sure it's OK.
I have had some trouble recreating the connection problems with the USB adapters recently. Can those of you who have problems connecting try post some information about your system? We really want to recreate it so that we can find and eliminate the problem.
Something like this:
VEMSTune version (development release date)
Laptop brand, model: Lenovo X61
Processor: Intel Core2Duo T7300
OS (Also 32 or 64 bit): XP 32 bit.
USB-RS232 adapter brand: ST Labs (From Webshop)
USB-RS232 driver version: v2.0.13.130 (XP driver) Released 2009-12-02
Problem description: Current configuration not tested yet.
Jörgen Karlsson
Gothenburg, Sweden.
Hi,
Thanks for helping us test the software but please keep in mind that VemsTune is an early development release, the current version is 0.10.7. We hope that the first release candidate will be released 2010-01-28 or 2010-02-04.
I'm preparing for the next HU trip, I hope that we will have a RC1 when I leave HU the next time.
It's interesting to see that Mega Log Viewer is so highly regarded. I only thought that a very small number of users used it. I have failed to find a version that support VEMS at all out of the box.
I hope that our log analysis will provide a superior integrated analysis solution which eliminate any reason to use MLV. It must be preferred compared to an export-analysis-import solution like MLV. The vemscfg format used in VEMSTune is a development of the native config and tables files used by VEMS and when megatune goes out the door the msq format may become problematic to support.
Jörgen Karlsson
Göteborg
Thanks for replying to the thread Jorgen :) I'll start with the fault reports:
VEMSTune version: All versions up to and including 0.10.7 (2009.12.04)
Laptop brand, model: Fujitsu Siemens Amilo
Processor: Intel Core2Duo T5250
OS (Also 32 or 64 bit): Vista 32 bit.
USB-RS232 adapter brand: Unbranded Prolific
USB-RS232 driver version: 3.3.2.105 Released 2008-10-27
Problem description: Cannot connect to ECU. (even though Megatune can)
Same laptop with FTDI USB device: (will get version later)
Problem description: Connects, but drops connection repeatedly with defined time period until VEMStune is restarted. Sometimes it's 30 seconds, sometimes 10 minutes - this changes each time Vemstune is started.
Please make it reliable before making it pretty! ;)
<<< It's interesting to see that Mega Log Viewer is so highly regarded. I only thought that a very small number of users used it. I have failed to find a version that support VEMS at all out of the box. >>>
- I'm a big fan of the MLV, but I am excited to see Datalogging being regarded as a high priority with the VEMS developers.
- I sell more ECU's based on the quality of the MLV when compared to other ECU brand's datalogging.
- It puts me in a very bad position to be losing this feature (MLV) in search of newer firmware / features.
- What problems do you have with MLV and VEMS? (I used from 1.076 to 1.53 firmware so far with MLV)
My favorite MLV Features:
- Can scale any log channel min and max (useful to visually line-up several lines in the graph)
- Having several graphs total (up to 4 total, keeps the clutter, uncluttered)
- Can convert MAP to Boost(Psi)
- Ability to customize log channel Names (by modifying VEMS MT ini files)
- Can hide the Dashboard, gauges, and Field Select for un-obstructed graph viewing
- Can keep the cursor centered
- Can adjust the amount of data that takes up the screen horizontally
- can choose any color for graph lines, background, etc.
General Data Logging Requests / Recommendations:
- "One Key Touch" when PC Logging (save the file naming for later, after the log is completed
- Internal SD Logging (like to use this ASAP)
- ECU setting to automatically turn internal Logging ON/OFF (I use 3,500 rpm with other ecu brands)
- Ability to "zero out" or "shift" the time line in the log display (for example, when the car launches, make this "0.00 seconds"
- Integrated logging within the ECU interface (It is good that VEMSTune logging will do this)
<< I hope that our log analysis will provide a superior integrated analysis solution which eliminate any reason to use MLV.
It must be preferred compared to an export-analysis-import solution like MLV. The vemscfg format used in VEMSTune is a development of the native config and tables files used by VEMS and when megatune goes out the door the msq format may become problematic to support. >>
- This is great. I just need to have my MLV until this is fully ready!
Quote from: Jorgen on December 11, 2009, 02:00:36 AM
It's interesting to see that Mega Log Viewer is so highly regarded. I only thought that a very small number of users used it. I have failed to find a version that support VEMS at all out of the box.
Jörgen Karlsson
Göteborg
It means you never tried it? IMO any of last two years versions work with VEMS out of the box. LT401Vette (http://195.159.109.134/vemsuk/forum/index.php?action=profile;u=196) is author of MLV he has just 10 posts here but result is awesome. I have little modified MSGraph.properties file for supporting other tables like Boost table. But main tables and analyzer is compatible as is.
Jörgen, you realy need try how it works. To overcome this piece of software is hard to reach goal. Especially in ease of use, visual representation, speed and stability.
I can afforce good points of MLV:
-easy to read information about current, minimal maximal values for each chanel
-easy to migrate over all datalog, easy to catch any point in datalog under cursor
-current engine working point representation on numeric tables
-possibility to compare two datalogs with overlying
-live logging: MLV can read file currently logged by tuning program
-averaging calculations of marked area of datalog for each chanel
-possibility to create calculated channels
-possibility to define range of any channel (automatic by default)
-save necessary/desired chanels combination (alpha-N, boosted, n/a S-D engine).
-not laggy
-
no bugs!And here is no word about VE analyser. Just viewer.
Gints
I have narrowed my problems down to the OS being Vista or Windows 7.
Absolutely NO problems with XP based OS´s on desktops or laptops serial port or prolific usb adapter.
VEMSTune version - ALL
Laptop brand, model: Packard Bell
Processor: Intel T1600 Core duo
OS (Also 32 or 64 bit): Windows Vista first and then Windows 7 32bit
USB-RS232 adapter brand: SEWELL 1103
USB-RS232 driver version: All driver versions
Problem description: Random "ecu not connected" , occurs between 0-10minutes, mostly within the first 2minutes, Input stream from
ecu is corrupted and tables become corrupt if refreshed. I have configured the vemstune.exe to run on a single processor but that did not help things.
In MLV i have designed a graph line that represents the correct VE based on gve, lambda readings, lambda target, air temp compensation, EGO control. This makes VE tuning literally an hours job or less and makes datalog based tuning ultra easy.
Using the MAX and MIN, I make GVE and VE correct have the same max and min, all I have to do is look at those lines and see how far off they are, the closer they are the less fine tuning or any tuning is needed.
I have used MLV for all my tuning , both live tuning and offline tuning. It´s real nice to be able to jump into MLV that is trailing a live file and review something quick that you missed during a conversation with the driver. So there is no need to repeat the same load site.
Especially things like MAT and CLT that you may have missed while looking at other gauges.
I have even made MLV work with a ecu that outputs a format no where near VEMS and MS I like it that much.
So datalog related I think it´s important that a very easy to manipulate (custom graph lines through mathematical functions combining many data inputs) program is used within VT - I reccomend you guys take a look at the Race Technology datalogging software and how it allows you to create complex data channels and helps in making very complex formulas from simple table inputs, such as temp at 30C = 2v , 60C = 5v , 90C = 5.5v . This makes changing hard Voltage numbers into meaningful figures that are accurate.
I have also never had to do anything to MLV to make it work with any megatune log file ms or vems. It even works better with Vems then MS because of the lambda target and lambda channels that I use for Gve calculations
I have yet to be able to tune with VT so I can´t comment on that, but I think the
lambda target / lambda bar gauge to be very clever and will probably be very helpful.
One important aspect I think needs to be addressed is the use of h[2] table names and any other hex values to be bad.
but I´m sure this will disappear eventually.
Hi Jørgen. To me it looks like that vemstune do not like the Core 2 Duo processor. I have to modify vemsTune.exe, vemsTune-c.exe, and VemsLogger.exe with imagecfg.exe. So that they use only one core.
Uploaded imagecfg.exe and imagecfg.bat files to make it easier for everyone to test.
http://www.vems.hu/files/ArvidOlausen/IMAGECFG-2.zip
Computer nr1.
Firmware: 1.1.62
VEMSTune version (2009.12.04)
Laptop brand, model:Hp NX9420
Processor: Intel Core2Duo T5600
OS : Vista 32 bit SP2.
USB-RS232 adapter brand: ST Labs (From Webshop)
USB-RS232 driver version: (19.11.2009 V 3.3.10.140 works ok) (31.01.08 v 3.3.0.1 also works on my computer)
Problem description:The ecu will become disconnected after 2-5sek.
Problem solution :I had to use imagecfg.exe to patch exe files to always only use 1 processor.
Computer nr2.
Firmware: 1.1.62
VEMSTune version (2009.12.04)
brand, model:MSI
Processor: AMD Penomen 2 X4 810
OS : Windows 7 Ultimate 64bit
USB-RS232 adapter brand: ST Labs (From Webshop)
USB-RS232 driver version: (19.11.2009 V 3.3.10.140 (31.01.08 v 3.3.0.1 )
Problem description:The ecu will become disconnected after 2-5sek.
Modify vemsTune.exe, vemsTune-c.exe, and VemsLogger.exe with imagecfg.exe does not work on this computer :-(
Arvid Olausen
Høyanger, Norway.
Thanks for the input
I would like to add one thing to the reports if possible,
in the folder where you have installed vemstune, a file called "filelog.txt" is created every time the program starts, might be only when using Vemstune-c.exe
Please copy this after a failed connection, and either post link here or email to me (
[email protected] )
thanks / Emil
Email Emil
Quote from: dnb on December 10, 2009, 04:40:50 AM
I just edited the vemsV3.ini file to match the header (*.h) file for the 1.1.63 firmware. All the data except the instructions is supplied with the firmware.
Send me an email and I'll reply attaching the file. It's not guaranteed to be perfect, but I am fairly sure it's OK.
Sent a mail the other bud did it get through ok?
Quote from: Jorgen on December 11, 2009, 02:00:36 AM
Hi,
Thanks for helping us test the software but please keep in mind that VemsTune is an early development release, the current version is 0.10.7. We hope that the first release candidate will be released 2010-01-28 or 2010-02-04.
I'm preparing for the next HU trip, I hope that we will have a RC1 when I leave HU the next time.
It's interesting to see that Mega Log Viewer is so highly regarded. I only thought that a very small number of users used it. I have failed to find a version that support VEMS at all out of the box.
I hope that our log analysis will provide a superior integrated analysis solution which eliminate any reason to use MLV. It must be preferred compared to an export-analysis-import solution like MLV. The vemscfg format used in VEMSTune is a development of the native config and tables files used by VEMS and when megatune goes out the door the msq format may become problematic to support.
Jörgen Karlsson
Göteborg
I find myself slightly discouraged by that statement :(
I've just bought a VEMS recently and will be running a pirate version of 1.1.63 and my car is booked to be mapped the start of Jan
But if VEMS tune is still in development release and no R1 candidate is being released until the end of jan why is there no official megatune support until the R1 is out?
I hope I do not find myself in a situation where I will be out of pocket because the vemstune will not be stable enough to use as I know there are other issues such as the lambda values being incorrect at present.
You can always use a older firmware version.
As they do have almost all the features and have run thousands of cars now.
Quote from: Jamo on December 12, 2009, 08:01:15 AM
Sent a mail the other bud did it get through ok?
I replied too. I assume it hasn't arrived, so I will send it again.
BTW it fixes the lambda reporting error too ;)
Quote from: gunni on December 12, 2009, 03:43:47 PM
You can always use a older firmware version.
As they do have almost all the features and have run thousands of cars now.
I didn't buy an ecu with 1.1.63 to downgrade it heh
Bys the by DnB has come up with a great option if vemstune isn't up too the job for me ;D
My email seems to not be working at the moment. Looks like the ISP has "upgraded" its server again...
I have put the file here: http://www.aqyz05.dsl.pipex.com/vemsv3.zip
The file should be put in the mtcfg folder of the project you wish to use and MT should read it.
It should work with the main features including logging and table editing, but it may not scale some of the more obscure parameters correctly. Please tell me if I've made any mistakes or omissions and I'll correct them.
Today there was a new test version published,
There is a new settings called "inter write delay", if you had problems with usb-rs232 adapters, try to change this to 40 mS to start with, and see if it works better. Vemstune has to be restarted for this value to be active. This has proven to solve the loosing connection problem on some computers
I had a setup which refused to connect at all here,
found another fix for this that will be included in next test version
There is still a bug in prolific's driver that is hard to get around, it crashes with AIM stream on before port is opened.
Recommendation for people with AIM enabled is to first start vemstune, then turn ECU on, this will keep the driver from crashing.
thanks / Emil
Still no go with the new VEMStune and prolific adaptor (as expected). I will try the FTDI adaptor and the dropping comms issue in the morning.
Win7 40ms and there are no more problems.
Prolific driver for Windows 7 is 2.0.019 by Prolific and comes from the link emil put up.
At 50ms when I changed the TPS the Battery voltage would change... Which is very strange.
EDIT.
Still going strong 50minutes, So this is resolved for me at least.
http://vems.hu/download/v3gui/VemsTune-Install-2009-12-15-testing.exe (http://vems.hu/download/v3gui/VemsTune-Install-2009-12-15-testing.exe)
Now with another patch that at least fixed my problem with not connecting at all.
EDIT: I was a little bit quick in posting the link, wasnt complete on the web server, but now it is.
dnb: Hope this test release solves your non-connecting problem
One small problem with todays release is the default setting for write delay, it is 5 mS,
I would recommend to start from 30, if loosing connection try higher (40?)
Success on all counts!
No connection problems, and no intermittant connection drops with either usb device.
The forum works ;D ;D ;D
Now I can get logging and stuff, I have some questions:
GWarm (the warmup enrichment) settles at 290% when the engine is warm (CLT >70C) when it should be 100%. Megatune's log from my logging hack for firmware 1.1.63 shows 100% as it should.
Similar for GBaro - it sticks at 128%, but it seems to do that in Megatune logs too. (I'm sure it used to do something a long time ago!!)
Given the car runs OK and hits lambda targets like it should do, I am thinking it's logging related and nothing serious.
BTW, it's worth noting that the crank fuelling and warmup seems to have changed in 1.1.63 to 1.1.59 and my Megatune hack does not have this, so you can't edit a couple of tables.
Quote from: dnb on December 17, 2009, 05:30:17 AM
Now I can get logging and stuff, I have some questions:
GWarm (the warmup enrichment) settles at 290% when the engine is warm (CLT >70C) when it should be 100%. Megatune's log from my logging hack for firmware 1.1.63 shows 100% as it should.
Similar for GBaro - it sticks at 128%, but it seems to do that in Megatune logs too. (I'm sure it used to do something a long time ago!!)
I checked my last logs with the 2009.12.14 VT build and GWarm behaves itself, it reads 100 with a warm engine. This is with 1.1.64 firmware. GBaro is stuck at 128% in all my logs it seems but I usually have barometric correction disabled for a number of reasons.
Jörgen
I checked some old logs I did with a 1.1.63 developer only release and the GWarm was at 290 with a fully warm engine then. It was higher while the engine was warming up. Obviously a log issue back then. I'm now running 1.1.64 in the test cars and it seems to be fixed.
My earlier logs with the internal SD card logger show 100%GWarm as expected even back with 1.1.61.
Jörgen
Thanks. It had all the hallmarks of logging - the engine would have hydrolocked if it was really enriching by 290%...
I have 1.1.64, so I might give that a go when I get back home unless there is an obvious mistake in the INI file.
Just started looking at 1.1.64 in anger. Most of it looks good, but there are a couple of little things:
1. Gamma enrichment appears to be misreported - it seems to be 2.55 times too big. (I'm sure that's just a logging issue)
2. The tacho output (I use the P259 on my test car because I have no other spare outputs) now seems to not work at all.
ETA: Tacho output does work, but I couldn't get the config to properly enable it until I used the "Assign Output Channels" dialog and restart VEMStune. I can't get the fault to happen again, so it was probably an isolated comms problem. VEMStune crashes a lot when I used the trigger player - could be related...
Quote from: dnb on December 17, 2009, 05:30:17 AMBTW, it's worth noting that the crank fuelling and warmup seems to have changed in 1.1.63 to 1.1.59 and my Megatune hack does not have this, so you can't edit a couple of tables.
So you can't access them through MegaTune? That's obvious.
There are two tables (1x4) now for cranking and afterstart . Warmup is the same size and table as earlier.
You could edit them with Megatune, but I haven't time to rewrite the ini file to sort it out - VEMStune now works properly.
The tables are actually 2x4 - there is a temperature and a fuel scale. (Same temperature scale for both tables)
In fact, you can access the tables through Megatune, but not in a nice way. The crank table in my modified ini file uses the same addresses as parts of the new table.
Quote from: dnb on December 27, 2009, 06:15:38 AM
Just started looking at 1.1.64 in anger. Most of it looks good, but there are a couple of little things:
1. Gamma enrichment appears to be misreported - it seems to be 2.55 times too big. (I'm sure that's just a logging issue)
2. The tacho output (I use the P259 on my test car because I have no other spare outputs) now seems to not work at all.
Sami and I saw the gamma enrichment issue, in his test bench it would bottom out the gamma enrich in the real time data when a trigger was not applied. He applied an audi trigger, I use a simple signal generator to stimulate the ecu with a simple trigger input. I still had the problem after applying trigger.
What kind of trigger do you use?
There is also a range problem with the reported enrichments, there is a new scale going up to 376% and there seem to be a missmatch somewhere. Christmas has stalled development, we will hopefully be able to sort this out shortly.
Jörgen
Trigger is standard 36-1 on the crank. No cam trigger. The gammae issue seems to be there with or without the trigger.
Whatever the problem is, it doesn't stop the car running and it doesn't stop the car achieving lambda target even when using a narrow ego control window. I am therefore fairly convinced it's logging/reporting related.
Let me know if there's any testing I can do - I'm off work until the 4th Jan so I have plenty of time.
after dnb's recomendation to convert the v8 to 1.1.64 and to use VT (thanks emil for the link)
and after the probs i had/having with 1.1.44 i am going to go ahead,
been reading the help files,
but bit confussed on how to save or mod a msg file?
is that the projects bit on VT?
will be setting the new version up on the v8 in the next few days but wanted to have a play with the current msq to get use to the functions/settings of VT,
The file format is a bit different in VT - it doesn't use a .msq file any more. The easiest way is to connect your ECU to the laptop and save a Vemstune cfg file before you flash the new firmware. You can then write this back on after upgrade and change the bits you need to - it'll complain about the things it thinks are daft.
Quote from: dnb on January 06, 2010, 01:58:37 AM
The file format is a bit different in VT - it doesn't use a .msq file any more. The easiest way is to connect your ECU to the laptop and save a Vemstune cfg file before you flash the new firmware. You can then write this back on after upgrade and change the bits you need to - it'll complain about the things it thinks are daft.
Such as the Lambda and ingition tables, which are a different size?
I am going this route when the engine goes back in the car, but its looking like the back end of Feb or march even, so hopefully the development team will have everything sorted, progress is steaming along!!!
in firmware upload
device: v3
version: 1.1.64
type:???? ps2 or nonps2?
The lambda and ignition tables should already be the right size since the 1.1.5x firmware I put on with you.
Why's the engine out again??
Quote from: dnb on January 06, 2010, 05:51:39 AM
The lambda and ignition tables should already be the right size since the 1.1.5x firmware I put on with you.
Why's the engine out again??
?? ??? Im running 1.1.47 and I did it myself :-\
And the engine is out because of this :D
(http://www.turbominis.co.uk/uploads/i/wh909-oe1/sprocket-1795.JPG)
Sorry - thought it was Tony taking the engine out again. ::) I can't read, you know ;)
Ha ha no worries :D But from a recent previous post Tony said he is running 1.1.44 ???
i am running 1.1.44 :D
and yup engine maybe coming out to sort this over run oil burning fault.
as im snowed in today i may have a go at converting over to 1.1.64 and VT
what is the ps or non ps bit about?
what do i need to set to?
when i save the current msq VT and load new firmware then upload the dloaded msq or vemscfg what ever its called will it set to 12x12?
and i gather i also need to do firmware validate etc after upload?
non ps version has speed senzor.
Does 1.1.6x still have switchable maps?
i'll go with ps version then as have no speed sensor.
hoping to dig the garage out of the snow today can then have a go at mucking the ecu up :D
I think the map switch is still there Rob.
Tony - it doesn't matter whether you go for PS2 or non-PS2. (I prefer non-PS2 because I was never really happy with the reliability of the PS2 keybord code)
The oil use issue is probably the cheap and nasty way Rover chose to implement the PCV system... Easy to fix with a catch can.
it only burning oil on 2 cylinders mate, (plugs black)
gona change the two inlet seals on them two before we go any further
I see your point... There's always the indian rope trick for getting the collets out of the valves.
Does anyone know an easy way to add new clt and mat definitions so that Vemstune picks them up nicely (ie without me messing about too much)? I have a few to add, so would prefer not to use the trick of renaming and replacing existing definitions. Thanks.
HELP!!!!!
warmed the car up to make starting easier with new firmware, plus wanted to make sure car had a good battery after sitting in garage in freezing temps,
down loaded the config as per VT,
uploaded new firmware and config and validated, came up with a list of errors,
took ages for car to start and when it did it ran as if on half cylinders wont rev just pops and bangs and stinks of rich fuel,
its idling lower than it should and also the its higher up the kpa scale now too.
any ideas, any one need to look at any files to see what i have got wrong?
did you review all the windows to be the same in most aspects?
send me the cfg file and I'll have a look. I had these sort of issues too for a while.
i have noticed the temp and intake temp settings are wrong,
and some tables are 12x12 and some are 14x16,
would it be best to go 14x16?
and a lot of conflics appear when you click the validate config button.
Cheers dave, cfg file on its way to you now.
The spark table should always be 12x12 - you can't change this. Only the VE table is 16x14.
I have found some problems with uploading patched firmwares with my USB to serial converters. I don't know why yet, but it could be they don't like sub zero temperatures!
The conflicts are usually just warnings. It shouldn't be too hard to sort out.
Oh right,
well the cfg file has been sent, also sent the old msq if you wanted to cross ref,
Here's my tweaked firmware. It has the Rover sensors built in to it (no patching required - in case this crashes your computer too...)
http://www.aqyz05.dsl.pipex.com/firmware_1.1.64_dnb.zip
Just unzip and copy into the vemsprojects directory over the top of the existing firmware_1.1.64 directory.
then upload the firmware to ecu again yeah?
or do i wait till the cfg files has been checked/tweeked?
Upload now. You should have the new cfg too now..
oops
for some reason the ecu didnt like that upgrade,
i cannot now connect to the ecu,
is there a way to reset it? and start again???
Did you select PS2 or non-PS2? (It should really be non-PS2 since that's what's on mine)
A power cycle should allow VEMStune to detect the ECU again after a bad flash.
Failing that, you might need to short the rx and tx serial lines together during ECU power-up to force it into bootloader mode.
yup,
and tried the power up and also shorting the rx and tx
still nowt
says bad communication check power, leads etc
all seem ok,
has it died?
right its detected but keeps flashing ecu bootloaded detected, what now?
nope now gone back to not being detected on com port1
right if i try to upload the firmware you sent Dave it goes beserk and wont conect
just uploaded the old 1.1.64 firmware (saved it you see incase of probs)
and its re loaded the old firmware and is conected on the pc,
when you try to upload your firmware Dave it says,
erasing all pages be patient,
erase ok
then it lists this load off errors on seperate lines,
fatal error out of sync unable to send address at page6
page7
page8
page9
page10
page11
page12
page13
page write check sum error at page 14
fatal error out of sync unable to send address to page 15
page16
page17
page18
page19
page20
error detected ecu left in boot mode
then it constantly displays ecu has not been detected on com port 1
Interesting. It uploaded to mine without incident and the car even ran. The coldest start it's ever done at -4C too.
how strange,
wonder why it wont load onto my ecu?
right been back round to the car,
still wont start, if you hold it continuosly on the key it tries to run, let go and it stops,
if you touch the throttle it dont even try,#
leave throttle alone (closed) and it does the above tries to fire pops and bangs a bit then turns over, then tries to catch etc etc,
anyone fancy taking a look at my cfg file?
failing that i may have to put MT and 1.1.44 back on cos at least it ran with the msq file i got saved from that setup.
thanks guys,
Have you been able to connect with the ecu again?
Are you able to log what is happening now and possibly against older logs of similar starts?
Mainly to see if the pulsewidth is same or very similar.
i'll do that tomorrow mate,
yes connecting to ecu ok now,
how do you do a log with VT?
Tools - start new vemslog
doh! cheers
what i could do, is do a log with the new firmware 1.1.64 where it dont start cold,
then re flash the ecu with the old 1.1.44 firmware and msq and config, and datalog that on cold start then someone who understands the datalogs could possibly take a look for me?
Well post up the logs and the configs.
And somebody might have a look.
Have you looked at the cranking and afterstart tables? They were completely corrupted in the file you sent me, so I copied over the ones from my TVR. These will most likely not be right for your car.
i did but they are different between mt and vt and im a bit confussed over setting them up,
i have a cold start log on 1.1.64 and cold start log on 1.1.44 and it running,
i then uploaded 1.1.64 again and got it to fire and run when hot and also have a log of this if anyones interested in having a look for me?
files are here:
http://www.capriracing.co.uk/datalog-09-01-2010.zip
i have put the files for 1.1.44 and 1.1.64 in seperate named folders
the 1.1.44 files are an msq and xls file as they were done in megatune not vems tune.
thank you....
In 1.1.44 there is prime pulse hot and cold but still no cold start enrichment. Vemstune - The reason that the engine doesnt start in 1.1.64 is that it uses the VE value. That VE value needs to resemble the warm start pulse width, and then tune the cranking enrichments table for cold starts as a percentage, 100 being the VE value set above, and obviously the pulse width depends on the size of the injectors. So you have prime pulse AND cold start enrichment.
That there is the problem
I am really surprised that the engine starts with just prime pulse
Perhaps with the prime pulse as big as it is, with the cranking enrichments, the engine is now flooded? Try it by removing the prime pulse value, check the engine and clear it out if it has, stick the plugs on the gas burner to clean them off, but do not cook them.
the car starts every time on the 1.1.44 settings,
even after i tried starting it on 1.1.64 and it failed,
where are these settings in VT mate?
sorry im still learning VT only used it this week
im an old MT user you see :-(
once warm the car starts on 1.1.64 although it did seem to kick back a bit but it seems to run better when hot on .64 than it does on the .44 firmware and revs up a lot crisper too,
had all these probs with .44 but Damir managed to sort that prob out for us,
oh injectors are bosch greens, 42lb/hr 440cc/min if i remember correctly,
required fuel is set to accomodate the larger injectors,
They can be found in the starting and idle menu, next to the tuning menu
Just try it without the prime pulse, put a zero in for the prime pulse, then see what happens, again check that the engine is not flooded. If Dave uses the same injectors as you and the engine configuration is of similar nature, it should start, but the settings may need tweaking only a little. If however Dave uses different size injectors, then you need to adjust the cranking VE to suit.
have set Prime P to 0
anything else i need to change?
as going round to the car in the morning while its cold,
if it fails to start what should i do then?
Yes, I have very different injectors. Mine are specified for alcohol based fuel (even though it's not available here yet).
Prime pulse should always be zero - I have never found it necessary. Use the cold crank and afterstart enrichments to do the fuelling for starting.
as i said this side of things is all new,
as the ecu came with a working config many moons ago,
its only recently with new firmwares i have had cold start probs, and now started to learn,
my cranking enrichment curve is:
-40C = 300%
-10C = 220%
20C = 164%
70C = 120%
After start Curve:
-40C = 240%
-10C = 170%
20C = 140%
70C = 120%
its all a bit different in 1.1.44 and megatune,
thus why i asked for any help on what sort of value i should be using etc.
It's difficult to guess exactly what the value should be - too many variables. I would suggest increasing the afterstart -10C bin by 20% to 30% if the engine is trying to fire but not suceeding. If there's nothing at all then you should increase both the crank and afterstart -10C bin.
Quote from: dnb on January 10, 2010, 04:15:56 AM
It's difficult to guess exactly what the value should be - too many variables. I would suggest increasing the afterstart -10C bin by 20% to 30% if the engine is trying to fire but not suceeding. If there's nothing at all then you should increase both the crank and afterstart -10C bin.
is that meant to read increase the afterstart -10 bin by 20-30% from 170% it is now?
Yes.
nope still no start today,
how can i wipe 1.1.64 off the ecu so i can put 1.1.44 back on as need the car next week,
have tried firmware upload in 1.1.44 but it wont allow
keeps saying no hello from avr
right have got it to upload 1.1.44 in MT again,
looks like im going to have to wait to go over to 1.1.64 for now :-(
Quote from: Tony C on January 10, 2010, 10:15:44 PM
right have got it to upload 1.1.44 in MT again,
looks like im going to have to wait to go over to 1.1.64 for now :-(
Do a datalog of a cold start and post the log, it contains the config so that's all we need.
I have used 1.1.64 successfully and started at 0 C down to -25 C, with some difficult as I underestimated how much fuel it actually needs.
Quote from: Tony C on January 09, 2010, 07:02:13 PM
files are here:
http://www.capriracing.co.uk/datalog-09-01-2010.zip
i have put the files for 1.1.44 and 1.1.64 in seperate named folders
the 1.1.44 files are an msq and xls file as they were done in megatune not vems tune.
thank you....
this what you want mate? ;)
I am bit confused about 1.1.63 cold start too:
My cranking curve on 1.1.53 was up to 400% @ -40degC. Now it seems 304% is maximum.
Does injection of additional fuel at first cranking cycles really works in this code?
It seems to me that prime-pulse scaling down not works. Does 0% means no pulse? I have pulse.
Can anybody comment?
Gints
Quote from: GintsK on January 13, 2010, 02:10:41 AM
I am bit confused about 1.1.63 cold start too:
My cranking curve on 1.1.53 was up to 400% @ -40degC. Now it seems 304% is maximum.
From what I can see, the 100% cranking enricment is the VE value set in
priming cranking and afterstart. That VE value is % of required fuel set in
base setup.
Making the cranking VE% larger and reducing the curve % will achieve the same.
I don't understand a warm cranking value on the curve of 120%, why not just increase the cranking VE by 20% and set the warm value on the curve for 100% and then adjust the remainder of the values apropriately. makes everything more representative when everything else is biased at 100% anyway.
My warm crank value in curve is always 100%. and warm cranking is adjusted by setting cranking VE value.
What I have to do? lowering my 100% to 75%.
It seems I have not enough fuel at -15..-20deC. Engine starts just after few cranking attempts.
Those cars with 1.1.53/54 seems have no problems. At least no one called me.
Quote from: GintsK on January 13, 2010, 02:10:41 AM
I am bit confused about 1.1.63 cold start too:
My cranking curve on 1.1.53 was up to 400% @ -40degC. Now it seems 304% is maximum.
Does injection of additional fuel at first cranking cycles really works in this code?
It seems to me that prime-pulse scaling down not works. Does 0% means no pulse? I have pulse.
Can anybody comment?
Gints
cranking=cranking enrichmnet*warump enrichment
Lowering below 100% does not work. Higher than 304% not allowed.
According to my experience and this graph (http://www.vems.co.uk/VEMSUserManual/CrankFuelling.png)
at 0degC 4..6 times more fuel needed compared to fully warmed.
Graph shows even about 15 times richening at -20degC! I have -15C here!
Does 1.1.64 injects additional fuel at first cycles? Previous it was configurable.
Quote from: MWfire on January 14, 2010, 02:59:13 AM
cranking=cranking enrichmnet*warump enrichment
at 0degC it is OK. (May be). 135%*300=405%
But when temperature goes lower...
145% is my WU at -15degC (engine idling fairly good, WBO2 reads 0.9...1.0). But then 1.45*300=435% - far too low!
My WU graph usually looks very similar:
http://www.vems.hu/wiki/index.php?page=MembersPage%2FAndrey%2FCranking
(http://www.vems.hu/files/MembersPage/Andrey/CrankSettings/Warmup.PNG)
Quote from: MWfire on January 14, 2010, 02:59:13 AM
cranking=cranking enrichmnet*warump enrichment
Are you sure about that? Please explain :-\
Cranking * warmup would give massive pulse width
I was always led to believe that cranking is what gets the engine started, pure and simple pulse width value, once the engine has started it is warmup*VE + afterstart and then to just warmup*VE to just VE
Where cranking enrichment is cranking VE* cranking temperature scaling
Seems MWfire is right. In accordance with Andrey.
Problem is not too big pulse, but too short if inject additional fuel at cranking is discarded.
In beginning cranking value was used for multiplicaion twice (square). Then 145% * 300%^2=13 - close to perfect!!
Something else to consider, is setting the injectors to fire all banks at cranking, that will add more fuel, reducing the cranking curve peak. I couldnt get enough pulse width on 1.0.73 with individual fire at cranking, but all banks at cranking made all the difference :)
Quote from: GintsK on January 14, 2010, 04:56:56 AM
Seems MWfire is right. In accordance with Andrey.
Problem is not too big pulse, but too short if inject additional fuel at cranking is discarded.
In beginning cranking value was used for multiplicaion twice (square). Then 145% * 300%^2=13 - close to perfect!!
i think in 1.1.64 isn't twice, that was in older version.
My cold start with 1.1.64
http://www.youtube.com/watch?v=23q2jm2mnyE&feature=channel
MWfire: I am getting a bit skeptical about the cold start tuning in 1.1.64..
I have trouble getting enough pulsewidth (maxing 304%) and it begins at -10 C.
I've been using alternating injectors up till now, and will try firing in banks/semi-sequentially starting tomorrow.
Seems firing strategy of injectors do not change amount of fuel a lot.
Batch ~3ms 4 times, semi ~6ms 2 times, seqv ~12ms. 1time per cycle.
Mattias, can you concertize possibility to enable inject additional fuel at cranking in 1.1.63/64. Is it just VEMStune problem or firmware?
It is urgent for me. We have first race in Saturday. But I put these firmwares on two fast cars. They are about 200km far from me. I trying to solve problem from distance.
Gints
You can try cranking car with fast squeezing throttle(but no more than 67%, becouse more than 67% tps is fuel cut). Then vems will inject more fuel.
BTW sandgren start car with 1.1.64 at-20C without problems. I can send you cranking values from his car.
Quote from: MWfire on January 14, 2010, 03:19:39 PM
You can try cranking car with fast squeezing throttle(but no more than 67%, becouse more than 67% tps is fuel cut). Then vems will inject more fuel.
BTW sandgren start car with 1.1.64 at-20C without problems. I can send you cranking values from his car.
Yes, please send! Gintsk at inbox.lv
Quote from: GintsK on January 14, 2010, 07:53:48 PM
Quote from: MWfire on January 14, 2010, 03:19:39 PM
You can try cranking car with fast squeezing throttle(but no more than 67%, becouse more than 67% tps is fuel cut). Then vems will inject more fuel.
BTW sandgren start car with 1.1.64 at-20C without problems. I can send you cranking values from his car.
Yes, please send! Gintsk at inbox.lv
sended
Quote from: GintsK on January 14, 2010, 01:53:52 PM
Seems firing strategy of injectors do not change amount of fuel a lot.
Batch ~3ms 4 times, semi ~6ms 2 times, seqv ~12ms. 1time per cycle.
Mattias, can you concertize possibility to enable inject additional fuel at cranking in 1.1.63/64. Is it just VEMStune problem or firmware?
It is urgent for me. We have first race in Saturday. But I put these firmwares on two fast cars. They are about 200km far from me. I trying to solve problem from distance
Your best option is to raise "cranking VE" x1.5 to 2.0 to temporarily solve the problem. This will rarely cause problems with the warm start, just press WOT to flood clear if that is suspected. At least it will solve the cold start.
I have confirmed that I need this scale-up at -10 and lower temperature, maybe at warmer temperature on some engines. The cranking table values for the cold coolant temperature for my engine are maxed out.
There is no "additional fuel" as in Andrey's initial implementation in 1.1.51, I'm not sure where this option got taken away. It is not in 1.1.63 or .64.
I will suggest to the firmware developers that the values are scaled up at least x2 because my car needs 7-8 ms at warm engine, at -15 it can barely start at 60 ms and would need 80-90 ms to start immediately.
Quote from: MWfire on January 14, 2010, 03:19:39 PMBTW sandgren start car with 1.1.64 at-20C without problems. I can send you cranking values from his car.
"Without problems" is not true, it does start but the the values are bottomed out and it needs a few seconds of cranking to start, always starts though - rarely requires a second attempt.
Hi.
I have problem with trigger sync. I use 60-2, own made trigger wheel.
Engine run on low rpm correctly. But always, when engine go to ~3000rpm vemtune show trigger error and have samothing like rev limiter. All revilimters are set to ~7000rpm
What should I check? It may be wrong polaristation of VR sensor? Or maybe trigger wheel is wrong made?
Quote from: GintsK on January 14, 2010, 01:53:52 PM
Seems firing strategy of injectors do not change amount of fuel a lot.
Batch ~3ms 4 times, semi ~6ms 2 times, seqv ~12ms. 1time per cycle.
Mattias, can you concertize possibility to enable inject additional fuel at cranking in 1.1.63/64. Is it just VEMStune problem or firmware?
It is urgent for me. We have first race in Saturday. But I put these firmwares on two fast cars. They are about 200km far from me. I trying to solve problem from distance.
Gints
I thought that 'Fire all banks while cranking' gave you four times as much fuel during cranking, if you are using sequencial injection, and twice as much with semi. Is there anyone still running batch fire injection these days?
I personaly only use fire all banks while cranking which means that even with the older firmwares, I was no where near mxing out the pulse width that then, was a max of 25ms
Sprocket: With the old system using only a cold/hot pulsewidth - changing from alternating to all banks would result in as much more fuel as the number of cylinders you had. So if you had 4 injectors and 4 ms warm start pw, you would get 4 ms x 4 cyl = 16 ms on each 4-stroke cycle. All injectors went from firing once each cycle to firing at every ignition/injection event, the pulsewidth stayed the same.
With the new "enrichment" strategy the pulsewidth is calculated and you're given the option of dividing it over the 4-stroke cycle, so you only get the fuel more often but the total amount is always the same. That makes it possible to use the same cranking tables regardless of your injector strategy for cranking.
My engine starts just barely better when I don't use alternating, simply because fuel is injected more often so the chance is higher that the engine gets some on each intake stroke. Yes, I would've liked to see true sequential fuel injection implemented soon as well to see what difference that can make for both cranking, idle and low load quality and power.
Well there you go ::)
These things need to be explained better as those upgrading from earlier firmwares could well become unstuck, when no one describes what has changed.
I wont butt in now as its clear the differencies in firmwares are huge, and what has gone before relates little to what is availble now, still without real explanations of differencies.
The post above is a real explanation, and made it 100% clear. Thanks :)
Today I wonders how easy startups new engine with 1.1.54 strategy from Andrey.
1.1.64 have good thing: afterstart enrichment is in separate table. I think it is necessary.
I hope lost additional cranking fuel is just bug. And not dropped on purpose.
Why did they reduce the cranking fuel table from 8x8 to 4x4???
Is it going to remain that way?
Does 1.1.6x have the same Boost DC per Gear as the 1.1.47??
main reason was e85 at low temperatures where cranking enrichment requirements are way different than aftersart enrichment.
Are you expect some difficulties with 4 point graph?
Quote from: lugnuts on January 16, 2010, 01:06:14 PM
Why did they reduce the cranking fuel table from 8x8 to 4x4???
Is it going to remain that way?
Does 1.1.6x have the same Boost DC per Gear as the 1.1.47??
becouse now there is crank curve and afterstart cure
1.1.6x doesn't have Boost DC per Gear as the 1.1.47, but i'm working on that.
BTW 1.1.65 has fix for afterstart enrichment
Quote from: Sprocket on January 16, 2010, 05:46:39 AMThese things need to be explained better as those upgrading from earlier firmwares could well become unstuck, when no one describes what has changed.
I wont butt in now as its clear the differencies in firmwares are huge, and what has gone before relates little to what is availble now, still without real explanations of differencies.
Until now there has been no good documentation built-in to the system, and none of the firmwares we are talking about have ever been "official" although advanced users are using them.. or rather "testing" them on their own risk.
To make the upgrades easier we will add "configlets", or sub-pieces of config that will apply sane settings and tables for a sub-set of the config. So if you want to convert you can simply apply cranking and afterstart enrichment that will automagically work if you set the req fuel value using the standard equation.
Even now VemsTune applies sane defaults when you upgrade, so it's not as bad as you may think to upgrade.
Quote from: Sprocket on January 16, 2010, 05:46:39 AMThe post above is a real explanation, and made it 100% clear. Thanks :)
Thanks, I will add my explanation to the built-in "Press F1 for help" dialgos.
Quote from: GintsK on January 16, 2010, 02:49:28 PM
main reason was e85 at low temperatures where cranking enrichment requirements are way different than aftersart enrichment.
Are you expect some difficulties with 4 point graph?
When myself and Emil were in Hungary we enforced the split of the 1x8 cranking table into two tables for both cranking and afterstart (as of 1.1.63).
While E85 is a good reason even regular gasoline also has a "knee" in the graph where demand raises pretty dramatically, it's just shifted down in temperature. With only four cells you can focus enrichment on the most important temperatures, more points would be nice but it really doesn't need much more than this. It's radically better than the two pulsewidths we had before.
Tuning the cranking enrichment is a really hard job for anybody! I can only imagine how much effort has gone into that kind of research at Bosch and others who make industry systems. I like the graph you presented above : http://www.vems.co.uk/VEMSUserManual/CrankFuelling.png
You can imagine that with gasoline you can get a pretty small error with only four cells in the cranking table, example temperatures : -30, -10, 20, 70
I will make a good default table for my car and test on a few others, it will make it into the default configurations and configlets.
Quote from: MWfire on January 16, 2010, 08:56:30 PMBTW 1.1.65 has fix for afterstart enrichment
I'm pretty sure you mean cranking enrichment? We both, you and me, have noticed how we bottomed out the values in the table so we need to scale things up. It might make it into 1.1.66 real soon.
I agree, 4 point cranking enrichment is a big step forward, I was just worried because we had 8 points before.
What about using 4 points for the Warm-up Enrichment? To allow 8 points for cranking?
Can you guys make my simple "Boost DC Per Wheel Speed" in the next revision? It's very good for drag racing FWD cars, and very easy to set-up.
VEMSTune Questions:
1) What do you do for datalogging? What exact method is being used now on what version?
2) Is there any way to hack the logs to work with MegaLogViewer?
I think 4 points for WUE is too less. Currently I use WUE table for some protection purposes when engine comes overheeted.
That's a good point, I do that as well. I was just trying to think of a way to get the 8 cranking points back.
Overhead could be added to the "Safety" section as well.
What do you use for datalogging?
how will the engine benefit from more fuel during idle or normal cruising once overheated? For example if you go at 100C CLT and at that point you go to 105% of VE, how much reduction do you get on the CLT? On the other hand, isnt your lamda recalculating the fueling back to 1.00 on idle and cruising anyway? Unless you are doing this for drag and track purposes,in which case i can understand it. But for normal everyday use, i think it makes things more complicated.
I think the best way to avoid overheating is a large radiator and a lower degree thermostat.
On version 1.1.27 which i still use and works fine, you only have 2 cranking temps.So far,for the 440cc @3bar injectors i use, as well as 630cc which other have used on the audis, they had no problems with the cranking. So to go to 4 bin table is a good improvement i guess, but i wouldnt think anything more is needed.
Warmup enrichments do need to stay as they are, as well as the ability to manipulate the coolant bins as you want.I think that feature is very important,and has allowed me to control idle efficiently as well.
Personally, i would add a spark advance as well as retard feature,same as found on the motronic, so that when you use meth, or any kind of feature that can reduce the IATs and give better perfomance, the ECU would automatically climb to the higher advance. If you only have a retard and set to 0 the retard on the coolest IATs you get, then you are obligated to make a different map,via a switchable map feature to get any benefit from meth injection because with the addition of meth, you actually have a higher octane fuel + lower IATs(and therefore perhaps more advance can be added).
Hi.
Does anyone have a properly configured and well-functioning maps:
MAT/TPS Fuel Enrichement
MAT/TPS Ignition Retarda
RPM/MAP MAP Corr
??
and could share them?
mattias:
i mean to afterstart enrichment
lugnuts:
i'm working in embed boost DC vs gear(like in 1.1.47), but with alterboost for N20,water or what do you want :)
Hi all.
I.ve been offline to all vems things for a while. Now I see there has been changes a lot.
Being lazy (lot of text) and confused, could you guys awser me one small question. I've evolved from 1.0.73 megatune to 1.0.73 vemstune.
It looks very cool and got it working quickly. I tried 1.1.63 to but is seems like a gyand leep to mankind!
The firts thing I noticed that I lost 14 x 16 with spark and lambda map.
I have 1.3 bar boost, 60 - 2, 4cil, wasted spark and full seqential without camsinc.
I just want to use the better features from 1.1.6x And want to add the round gauche and egt.
So. What's the best software to confert to from 1.0.73 14 x 16.
if you read all the info on vems, you'll see that 14x16 table is only for VE now, and spark has 12x12 and lamda has 8x8. these are interpolating bins so you'll have to customise your previous tables to fit the new ones
<<< how will the engine benefit from more fuel during idle or normal cruising once overheated? For example if you go at 100C CLT and at that point you go to 105% of VE, how much reduction do you get on the CLT? On the other hand, isnt your lamda recalculating the fueling back to 1.00 on idle and cruising anyway? Unless you are doing this for drag and track purposes,in which case i can understand it. But for normal everyday use, i think it makes things more complicated. >>>
- In my opinion, it is meant to recude the cylinder temperature, and the chance for detonation. It won't do anything for reducing the coolant temp at all.
- Having this function in the "Safety" section makes sense in your opinion, so people can simply not use the feature. Less complicated then having to manipulate the warmup map.
<<< On version 1.1.27 which i still use and works fine, you only have 2 cranking temps.So far,for the 440cc @3bar injectors i use, as well as 630cc which other have used on the audis, they had no problems with the cranking. So to go to 4 bin table is a good improvement i guess, but i wouldnt think anything more is needed. >>>
- I disagree in part. 4 points is better than 2 of course. But we had 10 in MT and 8 in VemsTune. I don't like going "backward"
<<< Warmup enrichments do need to stay as they are, as well as the ability to manipulate the coolant bins as you want.I think that feature is very important,and has allowed me to control idle efficiently as well. >>>
- Yes I agree 100% on being able to set your own Coolant Bins.
<<< Personally, i would add a spark advance as well as retard feature,same as found on the motronic, so that when you use meth, or any kind of feature that can reduce the IATs and give better perfomance, the ECU would automatically climb to the higher advance. If you only have a retard and set to 0 the retard on the coolest IATs you get, then you are obligated to make a different map,via a switchable map feature to get any benefit from meth injection because with the addition of meth, you actually have a higher octane fuel + lower IATs(and therefore perhaps more advance can be added). >>>
- Another good idea because the fuel itself used in the Meth can allow for even more advance than the ecu will correct for based on the IAT.
- But then you should just make a dedicated "Meth Injection" Output so the Nitrous guys dont blow upo their motors, haha.
<<< Hi.
Does anyone have a properly configured and well-functioning maps:
MAT/TPS Fuel Enrichement
MAT/TPS Ignition Retarda
RPM/MAP MAP Corr
??
and could share them? >>>
- This is critical. if VEMS is to have commercial success, you need to include "base files" which have all of the outputs "Disabled", this will cut down on many support question.
<<< lugnuts: i'm working in embed boost DC vs gear(like in 1.1.47), but with alterboost for N20,water or what do you want Smiley >>>
- Sounds good, I will send you a PM
there is already a feature to reduce overheat and that is MAT/TPS fuel enrichment. The hotter the engine and IATs the more fuel you can add that way. No need to use the warmup enrichment bins in reverse!!
By the way, when GintsK says he uses that as safety for 'overheating' purposes, what does he mean? I thought he meant engine overheating, thus CLT going high(which would go up if you have high cyl temp anyway).
On high load, if you want to avoid knock, you reduce timing or add fuel. Warm up enrichment bins deal with CLT. not with engine's cyl heat and knock. So as far as i understand it the MAT/TPS table is there to deal with just that. Not the warm up enrichment bins.
you can add more fuel when clt is exceed 105C with warmup enrichment but allways(idle,wot..)
you can add more fuel when MAT is big, but with tps scalings(on wot more fuel, but in idle not).
you can add more fuel depending on egt
you can reduce boost depending on egt and mat
excactly, which is why i dont understand why the warmup enrich. scale has to be used for that as well.... ???
The MAT enrich should be KPA dependent or at least choosable between TPS and KPA.
same for MAT retard.
Safety from coolant enrichment when it steps out of bounds is fine to do with the wrongly named
Warmup enrichment table, should simply be called Coolant temp enrichment table.
VEMS should not use table positions as any references to other functions, so the second last cell in that table has some effect on
idle control or whatever, they should all be referenced against the numbers and never ever table positions.
-> A80Avant Overheated engines often have knocking danger. My ego limit is 10%, but WUE @130degC is +30%. and so it works.
BTW it is possible to overheat engine in cold winter when radiator is blocked by snow. Here MAT did not help.
Quote from: lugnuts on January 17, 2010, 06:39:17 AM
What do you use for datalogging?
I still use Megatune from
dnb. Thanks alot for this!
Quote from: MWfire on January 17, 2010, 09:45:40 PM
you can add more fuel when clt is exceed 105C with warmup enrichment but allways(idle,wot..)
you can add more fuel when MAT is big, but with tps scalings(on wot more fuel, but in idle not).
you can add more fuel depending on egt
you can reduce boost depending on egt and mat
Yes! All of these things currently works perfectly!
yes i understand that engine overheat will cause knock Gints, but are you saying that the rest of the options to avoid knock are not adequate?
@130C CLT, your coolant boils as well, so the engine should not work at all at those temps in the first place.
can you explain what you mean with the ego correction and the WUE as well? How does ego correction of lamda, is affected by WUE ??
My engine with a 80C thermostat and a good working cooling system only reaches 95-97C tops in a hot weather in Athens,Greece,in traffic on a 40C day. But i do aggree, as you say there may be times where different enviroments may call for drastic measures
coolant may be even not boils in closed system. But this is not about cooling but about saving engine when something with cooling goes wrong.
Rest of options are adequate and works. But this is additional protection which can or can not overlays with others. Coolant leak is case where not overlays.
If Ego correction is switched off WUE graph have direct influence on fuel dose. When Ego is on situation is bit more complicated. Therefore aggressive WUE ramp-up is needed to overcome ego correction.
ok i see what you mean with the ego correction.
anyway, more safety features is always better.I only hope there is a stable official version that will cure most bugs really.But i think it may end up in the 1.1.7x !!
keep up the good work guys in the development! we are counting on you!
Quote from: mattias on January 11, 2010, 06:13:29 AM
Quote from: Tony C on January 10, 2010, 10:15:44 PM
right have got it to upload 1.1.44 in MT again,
looks like im going to have to wait to go over to 1.1.64 for now :-(
Do a datalog of a cold start and post the log, it contains the config so that's all we need.
I have used 1.1.64 successfully and started at 0 C down to -25 C, with some difficult as I underestimated how much fuel it actually needs.
did anyone manage to have a look at this?
Tony, its not going to work without propper cranking enrichment values which you currently do not have. That is the difficult time consuming bit, trying to find those values. From this thread, it would appear that the cranking enrichments in 1.1.6x are a whole world appart from older firmwares.
Im sure Dave will be able to extrapolate from the values he uses in relation to the size of injector. The engines are similar enough to require similar amounts of fuel for cranking, its just working out the actual pulse width for the injectors you have, to give the same amount of fuel at cranking, then you can fine tune it from there.
Quote from: mattias on January 15, 2010, 07:53:31 PM
Sprocket: With the old system using only a cold/hot pulsewidth - changing from alternating to all banks would result in as much more fuel as the number of cylinders you had. So if you had 4 injectors and 4 ms warm start pw, you would get 4 ms x 4 cyl = 16 ms on each 4-stroke cycle. All injectors went from firing once each cycle to firing at every ignition/injection event, the pulsewidth stayed the same.
With the new "enrichment" strategy the pulsewidth is calculated and you're given the option of dividing it over the 4-stroke cycle, so you only get the fuel more often but the total amount is always the same. That makes it possible to use the same cranking tables regardless of your injector strategy for cranking.
My engine starts just barely better when I don't use alternating, simply because fuel is injected more often so the chance is higher that the engine gets some on each intake stroke. Yes, I would've liked to see true sequential fuel injection implemented soon as well to see what difference that can make for both cranking, idle and low load quality and power.
Im going to throw a spanner in the works,
If by setting the
Injector Firing Divider to '1',
Alternate Banks to '3..0', then
Cranking Fire Banks While Cranking to 'All Banks' and then
Cyl for Sequential, #Cyl/2 forSsemiSeq Cranking Fuel Divider to '1', will not add four times as much fuel at cranking???...
Quote from: Arthur on January 17, 2010, 06:34:07 PMI tried 1.1.63 to but is seems like a gyand leep to mankind!
The firts thing I noticed that I lost 14 x 16 with spark and lambda map.
..
So. What's the best software to confert to from 1.0.73 14 x 16.
I rarely see the need for 14x16, only on highly tuned engines with hot cams and 2 valve/cyls.
With interpolation and more tuning experience you'll see how it's more important where you put the bins and not how many they are.
I can't argue with you on table resolution being bad in itself, especially if the tuning software has methods to auto-tune or manually work with the larger tables. My opinion is still that it's mainly a marketing feature, to sell more ECUs and won't make your car run faster. "Look at my shiny rims! They're twice as shiny as that other brand!! They suck! HAHAHA"
How anal should you be in following the AFR/lambda target? Perfection is not really a personal goal of mine when I tune an engine, I shoot for a decent compromise. We can always discuss perfection, but know that it is utopia.
Example : Lets say you find a torque fluctuation just around the tuning frequency of the intake runners while the cam is coming on strong. Sweeping past and load holding in a dyno in this area reveal that you need 10 points in a 1500 rpm span to get a +/- 0.01 variation in lambda at most. This is great in a tricky area of the map and you can pat your self on the back once you've nailed the fuel table - and now do the same for task for ignition advance... Never mind the difference in tuning an engine that has flow metering electronics compared to the VE based approach.
In a real-life scenario where you drive the car - will it matter?
It's like injector fuel trim, very few have the opportunity to ever do any useful adjustment since they lack the sophisticated equipment and measuring tools - but as long as there is a margin of safety in the overall tuning and the plugs look fresh you can probably sleep well at night knowing that cyl #6 won't run too lean.
Convert to 1.1.64 or later versions, it will work out fine for you. The added control and features of the more recent firmware will make the engine run even more consistently good.
On a similar note.. On my wishlist is to have clearly written if it's TPS or kPA on the Y-axis on all tables.
Quote from: lugnuts on January 17, 2010, 09:29:56 PM<<< Hi.
Does anyone have a properly configured and well-functioning maps:
MAT/TPS Fuel Enrichement
MAT/TPS Ignition Retarda
RPM/MAP MAP Corr
??
and could share them? >>>
- This is critical. if VEMS is to have commercial success, you need to include "base files" which have all of the outputs "Disabled", this will cut down on many support question.
The "good defaults" are in the default firmware configurations that are delivered in the VemsTune install package. When you do an upgrade from an earlier firmware and the new tables and settings are missing in the old one, the default ones are used from the default configuration file of the firmware you're upgrading to.
At the moment they are pretty good but can get better, and some configlets to help individual settings/tables like cranking/afterstart/warmup for gasoline and ethanol will be added, but it takes an effort from people to test and make sure they are valid, which is what I'm working on now.
After the cranking table is scaled up (maybe also afterstart enrichment) I will come up with good defaults which should make it easy to start any engine as long as you set the req fuel value according to the correct formula, and then you only need to tweak the "cranking VE" and your VE table to make things work very well for your engine.
Quote from: Sprocket on January 18, 2010, 05:43:59 AM
If by setting the Injector Firing Divider to '1', Alternate Banks to '3..0', then Cranking Fire Banks While Cranking to 'All Banks' and then Cyl for Sequential, #Cyl/2 forSsemiSeq Cranking Fuel Divider to '1', will not add four times as much fuel at cranking???...
Like I said, the pulsewidth that is calculated from all the enrichments is divided over the 4-stroke cycle so the total pulsewidth is the same. To add fuel you need to alter the enrichement tables. The only thing you're doing with the cranking fuel divider is to decide how much to split the total pulsewidth over the 4-stroke cycle.
Hy all
I've upgrated ti 1.1.63 now.
With old HP ombibook 6100 no connection problems.
I still have some sanity check problems that I can't get rid off.
Output channel collision. Injector outputs collision with vvti channel
Warup_clt(9) Please check warmup
Tooth_wheel_twidth[0] Please check tooth_wheel_twidth[0,1]
Shiftcut and launch Please check shiftcut input
boost_egtdec Please check boost target reductio above EGT and fuel added above EGT Limit
I can't find vvti anywhere in VT.
Checked warup but can't find problem.
Tooth wheel. 6 deg on 60 - 2 Should be oké!
Shiftcut checked all not needed is diabled?
boost_egtdec no matter what, it go's back to -85 and 616
What does map corr do?
I really like the F1 help function. It's the VT itselfs that bugging me the most. I cant find how to increase complete map by a certain %.
Like hole map x 1.2
Quote from: mattias on January 18, 2010, 06:27:49 AM
Quote from: Sprocket on January 18, 2010, 05:43:59 AM
If by setting the Injector Firing Divider to '1', Alternate Banks to '3..0', then Cranking Fire Banks While Cranking to 'All Banks' and then Cyl for Sequential, #Cyl/2 forSsemiSeq Cranking Fuel Divider to '1', will not add four times as much fuel at cranking???...
Like I said, the pulsewidth that is calculated from all the enrichments is divided over the 4-stroke cycle so the total pulsewidth is the same. To add fuel you need to alter the enrichement tables. The only thing you're doing with the cranking fuel divider is to decide how much to split the total pulsewidth over the 4-stroke cycle.
Ok, fair enough but I then fail to see the reason for 'cranking fuel divider' whether it is 1, 2 or 4 times per cycle resulting in only the exact same amount of fuel being elivered during cranking.
I thought 'fire all banks while cranking' turned the cranking injection sequence from sequential into batch and the divider was used to then divide the calculated PW by the number entered in 'cranking fuel divider' 1 for sequential, 2 for semi, and 4 for batch.
Its all a mystery to me ??? why do we have settings that give the same end result. Its a lot of un necessary configuration for no gain.
Im not being awkward, i just cannot see logic lol.
Quote from: Arthur on January 21, 2010, 03:02:12 AMI really like the F1 help function. It's the VT itselfs that bugging me the most. I cant find how to increase complete map by a certain %.
Like hole map x 1.2
Ctrl+A, Q/W/E/R
Quote from: Sprocket on January 22, 2010, 05:59:38 AMOk, fair enough but I then fail to see the reason for 'cranking fuel divider' whether it is 1, 2 or 4 times per cycle resulting in only the exact same amount of fuel being elivered during cranking.
I thought 'fire all banks while cranking' turned the cranking injection sequence from sequential into batch and the divider was used to then divide the calculated PW by the number entered in 'cranking fuel divider' 1 for sequential, 2 for semi, and 4 for batch.
Its all a mystery to me ??? why do we have settings that give the same end result. Its a lot of un necessary configuration for no gain.
Im not being awkward, i just cannot see logic lol.
Because firing more OFTEN will guarantee that fuel will be available in the intake tract when the intake valve opens when cranking at 200 rpm. It doesn't make much of a difference anyway, alternate (sequential) works just fine for me.
I could be all wrong on the "banks" statement and what the divider actually does, but it makes sense.
For my car fire all banks while cranking help alto to start faster.
Quote from: mattias on January 22, 2010, 10:17:54 AM
Quote from: Sprocket on January 22, 2010, 05:59:38 AMOk, fair enough but I then fail to see the reason for 'cranking fuel divider' whether it is 1, 2 or 4 times per cycle resulting in only the exact same amount of fuel being elivered during cranking.
I thought 'fire all banks while cranking' turned the cranking injection sequence from sequential into batch and the divider was used to then divide the calculated PW by the number entered in 'cranking fuel divider' 1 for sequential, 2 for semi, and 4 for batch.
Its all a mystery to me ??? why do we have settings that give the same end result. Its a lot of un necessary configuration for no gain.
Im not being awkward, i just cannot see logic lol.
Because firing more OFTEN will guarantee that fuel will be available in the intake tract when the intake valve opens when cranking at 200 rpm. It doesn't make much of a difference anyway, alternate (sequential) works just fine for me.
I could be all wrong on the "banks" statement and what the divider actually does, but it makes sense.
I understand this which is why I am amazed that Tony C manages to start his engine on prime pulse alone!!!
Quote from: Sprocket on January 22, 2010, 03:16:49 PMI understand this which is why I am amazed that Tony C manages to start his engine on prime pulse alone!!!
My snow mobile does too :)
.. and I bet my car could be setup to do that as well (maybe not really cold).
Ctrl+A, Q/W/E/R
Oké thanx. That works. Where can you find these simple things. Not in f1
And the rest. Are these question suppit or hard to anwser?
How can I have a collision with vvti channel, while I cant find vvti anywhere in VT?
What's wrong with these's cooling settings. (apart from not beeing tuned jet)
-38 200
-26 160
-14 140
- 2 125
10 115
22 110
34 108
46 104
58 102
70 100
The Tooth wheel width should be 6 deg on 60 - 2 .....duh! Why does the warning stay?
My first tooth is physicly 120 btdc, so I set my trigger to 10 ad next to 30. Vems can't cope with 120 btdc!
What should I set in reverence table? "10 and 40" or "10 and 30"?
Shiftcut: I checked all is diabled but it keeps telling me to check shiftcut input.
Boost_egtdec No matter what, it go's back to -85 and 616 I don't understand how this works at all.
And what does map corr do. Why do you have to correct map/rmp when you have ve-map ???
Should I just ignore these warnings???
Map corr is only used if you set it up so.
Those coolant settings don´t represent the required amount of fuel needed.
Somebody posted a picture of this just a few days ago.
you may have missed the "Show / hide help F2" in the 3d tuning maps. That will show the possible shortcuts within that window
Quote from: gunni on January 23, 2010, 05:17:12 AMThose coolant settings don´t represent the required amount of fuel needed.
Somebody posted a picture of this just a few days ago
They're pretty good for a warmup/coolant enrichment table, could use just a bit more below freezing but depends on the engine.
I don't think he was referring to the cranking table, which the picture posted earlier showed.
Yes I know. the enrichment should probebly be raised. I have 0.9 pw batched (all bank) while cranking at 5 deg C. the pw at a running warm engine is about 1.9 full seq.
Somehow my cranking fuel divirder goes back to 51 in all the time in VT.
I'm afraid to start while there are still sanity issues.
Are the bins oké? I've seen warning that said: monotoriuos asending order. That's yust about the level where my englisch gets me.
Does that meen: start low and use the same steps to raise? Should it always stop at 70?
I found the problem with my F2 function. My laptop needs shift for F1 but doesn,t need shift for F2.
:-X :-X
I've told them about "monotonous" before. That roughly means "boring".
The word they mean is "monatonic", and it is nearly as you describe - a continuously increasing (or decreasing) function. The step size doesn't matter, but the next step should always be bigger than the last.
Quote from: Arthur on January 21, 2010, 03:02:12 AM
Warup_clt(9) Please check warmup
Tooth_wheel_twidth[0] Please check tooth_wheel_twidth[0,1]
Shiftcut and launch Please check shiftcut input
boost_egtdec Please check boost target reductio above EGT and fuel added above EGT Limit
I think you'll find these are all just warnings. They are simply there to make you double check for common config problems. They can be annoying, but if you have checked them and are happy with them then they can be ignored. Annoyingly they keep being reported.
I found some times with Megatune that even though some config values were disabled and greyed out, if the did not have sane values certain things still had an effect. I can only expect the same to be true with Vemstune.
I always try and put sane values into everything even though they will never ever be used.
Sprocket: "hidden" settings that belong to disabled functions are hopefully eliminated completely by the time VemsTune is official, I don't know of any apparent ones at this time.
Arthur: I would seriously check if your serial port is working like it should. Nothing but yourself should be able to set/change the "cranking fuel divider".
Quote from: Sprocket on January 18, 2010, 05:29:17 AM
Tony, its not going to work without propper cranking enrichment values which you currently do not have. That is the difficult time consuming bit, trying to find those values. From this thread, it would appear that the cranking enrichments in 1.1.6x are a whole world appart from older firmwares.
Im sure Dave will be able to extrapolate from the values he uses in relation to the size of injector. The engines are similar enough to require similar amounts of fuel for cranking, its just working out the actual pulse width for the injectors you have, to give the same amount of fuel at cranking, then you can fine tune it from there.
Dave has offered to pop up and see me next time he is on the main land to see if we can get 1.1.64 running on the v8,
will keep you all posted how it goes :-)
Can it be that the pwm idle solenoid is quieter with 1.6.3?
what do you mean quieter?
Quote from: A80Avant on January 26, 2010, 02:35:23 AM
what do you mean quieter?
The IAC valve rattles because of a firmware glitch. The pulses in the first channel (for a dual solenoid setup) gets truncated intermittently. Also, the implementation is a kind of PWM that's not suitable for an IAC valve.
(http://e23.bimmerclub.ph/images/vems/20090704-iac-scope.png)
Related threads here:
Idle PWM - rattling sound? (http://195.159.109.134/vemsuk/forum/index.php/topic,286.0.html)
Erratic IAC dual-solenoid output (http://195.159.109.134/vemsuk/forum/index.php/topic,917.0.html)
I haven't bench tested 1.1.6x yet to see if the issue is still there.
is that for the dual solenoid mode?
in audis this is not used and works fine on 1.1.27 with the use of i diode
Then your lucky. I have tested on a two wire ford ICV and it was terrible.
I´m hoping that the new firmwares will have static frequency.
at least add it on, dont mess around with a PWM that already works on some cars!!! :D :D
How do we contribute to source code? I was a hardcore C/C++ programmer and microcontroller geek back in the day so I may be able to help pinpoint the problem and a possible fix -- and of course, make sure current setups won't get messed up with the changes.
on the cranking/warm up/idle settings, when you change the coolant bins on the curves, they all match. However on the idle PWM, they always stay set. Which one of these 2 facts is telling the truth about the CLT temp?
if i alter the coolant bins on all those areas, will the idle PWM follow, or will it stay on the already set values under idle IAC PWM settings?
Quote from: A80Avant on January 26, 2010, 04:19:40 AM
at least add it on, dont mess around with a PWM that already works on some cars!!! :D :D
In fact every installation have this interrupts. But some valves and settings prevent mechanical noise.
If you set pwm frequency to 0 and manage throttle stop so that valve duty is close to 50% - most probably valve will be quite on current firmwares.
Andrey already has code for true PWMing. It should be implemented in some of next firmwares
Well with 1.0.73 i never wandered if the solenoid worked. I could here a soft brum
Now with 1.1.63 I can't here it but it seems to work fine.
The recent firmwares (1.1.65+) have the new PWM code for IAC valves that seem to work. I haven't tested it yet but will once the code for cranking enrichment changes to work with northern sweden winter temperatures, possibly in a few days.
Quote from: GintsK on January 26, 2010, 05:59:50 PM
... Andrey already has code for true PWMing. It should be implemented in some of next firmwares
Quote from: mattias on January 27, 2010, 06:09:17 AM
The recent firmwares (1.1.65+) have the new PWM code for IAC valves that seem to work. I haven't tested it yet but will once the code for cranking enrichment changes to work with northern sweden winter temperatures, possibly in a few days.
Excellent! I hope it fixes that nasty interrupt as well.
Welcome,
Knock at what works from version 1.1.58?
Quote from: dainiuxsky on January 27, 2010, 05:45:05 PM
Welcome,
I noticed that the 1.1.61 version of the left konck since 1.1.63 Boost controller left to inj-5 "there is no activity"
Here with me as one?
Wut?
I think the anwer shout be:
If the is with the, than the is me, is not. (with)
;D
Im interested to buy this ecu but reading through all this topic puts me off, I wanted to download vems programe but dont know which one to use, the wiki pages are not so useful, the hardware specs seem good, but what actually works in the software? All these codes for guys with no programing skills are a nightmare. Who actually runs vems?
When i use the latest vemstune it crash when i enter the Web firmware tool, it ude to do that in earlyer versions aswell, but have worked in between
/Skassa
Quote from: blown_g on February 03, 2010, 01:08:23 AM
Im interested to buy this ecu but reading through all this topic puts me off, I wanted to download vems programe but dont know which one to use, the wiki pages are not so useful, the hardware specs seem good, but what actually works in the software? All these codes for guys with no programing skills are a nightmare. Who actually runs vems?
It would be best for you to contact the nearest dealer/enthusiast/re-seller in your area.
Tell us where you are located and someone here should be able to help you out.
I am in the USA, I do ship overseas in my main market (VW/Audi). But if the application is different, I would help you find someone closer to your location.
I'm sorry. Just joking. ;)
I had the same problem you have. If you buy the vems you'll learn a lot. All the guys here where/are of a lot off help. ;D
So far no-one has died installing a unit.
Quote from: [email protected] on February 03, 2010, 03:10:26 AM
So far no-one has died installing a unit.
I think I lost a few brain cells though in the early days :D
Im in UK, i tried to call VEMS in Hungary but I couldnt find anybody to give me advice, emalied them no reply.
I do have experience with stand alone systems, I have dealt/tuned aem,neptune/hondata, sds. I have found that these systems are easy to use some more advanced that others but most important the information and support is there readily available and in a sorted order. I have read instalation pages, some bits of information missing.
To be honest I havent been able to use the vems software or practice with its feature yet, not sure if you can sample with no ecu connection?? at one stage I was stuck with the full screen :P had to reinstall ect.
As a buyer i want to see more consistency on guidance and support for this product from the manufacturer as lets face it most ecu's will have some sort of problem of some kind. AEM and SDS in particularly that I have dealt with have given me great support on problems I have encountered.
In terms of hardware on paper VEMS seems like a good system, but lack of some information or rather the way is presented will make some potential buyers out there to go for other options ie Megasquirt, DTA or other similarly prices systems.
Maybe Im not geek enough to handle VEMS yet ;D
What is your motor application?
Usually the only problem is trigger settings, after that things are usually always the same.
There are alot of known simple configuration that get pased around, while some other as more difficult,
Here is the location of some Vems users and tuners that can be of assistance.
http://195.159.109.134/vemsuk/forum/index.php/topic,923.0.html
its VW 16V ABA turbo. I was wondering if I can use the stock temp sensors for coolant and air, how well does the stock 3 or 2 wire ISV work with VEMS?
I have a standalone as it is, looking to switch to vems, main use for multiple fuel maps ie e85 and fuel. How many maps can you save?
So is tuning done through megatune which is what megasquirt people have developed? Going through megatune I found it ok to use, but the version 1.7 of vems cant get my head around it. Can someone point me to a working version of vems and can I play with it while offline?
Download this program
http://www.vems.hu/download/v3gui/VemsTune-Install-2010-01-08.exe
goto
file - project - create new project
name it something and hit Save and Load.
Now you can review all the possible options in that firmware and manipulate it any way you can. This will allow easy offline adjusting.
You can run 2 configs using VEMS.
Hey guys, which is the most stable version of the new firmware? I have Toyota style CAS, 24+1. I was told to check out the new firmware to try to get acceleration enleanment, which is one of the last things I feel I'm lacking with VEMS.
Thanks!
PS, blended Alpha-N and SD, which is very beneficial for cars with ITB's...I heard the new firmware supports this too?
I repeat what jrussel said in another thread .
Wait for a month or two for the official release. The current (1.1.6x) releases are for development and testing, not meant for widespread use.
HI
1.1.67 Super Boost controller , knock detection only after 7200 rpm ::) tried on 3 different cars Audi AAN, 3B, and 1.8T.
Knock Diff gauge always aktiv after 7000 RPM I tried various configurations, almost the same. ???
sorry for my bad English.
Quote from: dainiuxsky on February 09, 2010, 04:19:28 PM
HI
1.1.67 Super Boost controller , knock detection only after 7200 rpm ::) tried on 3 different cars Audi AAN, 3B, and 1.8T.
Knock Diff gauge always aktiv after 7000 RPM I tried various configurations, almost the same. ???
sorry for my bad English.
sd card loggin on? there is no knock detection when sd log on..
Quote from: z0tya on February 09, 2010, 06:50:08 PM
Quote from: dainiuxsky on February 09, 2010, 04:19:28 PM
HI
1.1.67 Super Boost controller , knock detection only after 7200 rpm ::) tried on 3 different cars Audi AAN, 3B, and 1.8T.
Knock Diff gauge always aktiv after 7000 RPM I tried various configurations, almost the same. ???
sorry for my bad English.
sd card loggin on? there is no knock detection when sd log on..
namely without SD card.
dainiuxsky, what you mean with super boost ctrl?
Is it working perfectly and easy to tune at same time?
Quote from: GintsK on February 09, 2010, 09:51:38 PM
dainiuxsky, what you mean with super boost ctrl?
Is it working perfectly and easy to tune at same time?
Older version of rails under linear control.
Current individual map, you can put there owerboost where needed and partly to speed limint. :o
Should still analog input 0-5V can be the total pressure regulation as an external boost controller. ::)
Okay, so I got the latest VEMSTune, uploaded the 1.1.68 firmware to an ECU, and took a step-by-step approach through...
So now I have a couple of WTFs:
MAT/TPS Fuel
MAT/TPS Ignition Retard
What are the X, Y scalars? X=MAT and Y=TPS?
How they are they scaled? MAT I'm guessing is -40degC to 102degC is TPS % or some notional value?
And the values in the tables... are they % biased around 100% or some other value?
What values should be used as defaults?
MAP correction, what are the axis units? What is represented in the table?
What values should be used as defaults?
Acceleration Enrichment, a lot of this seems self-explanatory, but... what is the ign retard amount? And the fuel film?
What values should be used as defaults?
Its looking good and has some good usability features.
I can upload my config tomorow, 1.1.68
MAT/TPS Fuel & Ign tables - yes, axes as described. It depends on if you use the flash table scaling or the 14 point table (which can go to higher temperatures IIRC) Usually -40 to 102c is OK.
Sensible values are the usual values from the ideal gas laws with a rich bias to guard against idle heat soak, plus a rich streak for safety at high temperatures at moderate throttle.
MAP correction
Table populated with 100 for everything in my case - not using forced induction with throttle bodies on an alphaN strategy!
Acceleration Enrichment, a lot of this seems self-explanatory, but... what is the ign retard amount? And the fuel film?
What values should be used as defaults?
Retard ignition while accelerating. We had a chat about that & turbos some time ;) Fuel film looks to be a time based smoothing filter for evening out lumpy steady-state running. Good for big cams. I have played with it but haven't made my mind up yet about it...
If you have your firmware and temp tables flashed, can you go into VemsTune and use the 14 point temp sensor table? Or do you have to upload firmware without temp tables?
The curve that you can define yourself, or use a good default from the dropdown (configlet), has 17 points.
It's just an option, if you want to use that curve or the curve in flash memory, no special firmware.
So where are my cranking, afterstart and warm-up enrichments?
They didn't seem to appear in the workflow, the < and > keys assist with the workflow of setting up the system.
Nicer still would be to have a standard config with everything set to 'off', I have been told by more than one tuner that they would like to be able to start with every feature set to 'off' or a value that does not have an effect on the primary settings.
Quote from: dnb on March 05, 2010, 04:58:28 AM......
QuoteAcceleration Enrichment, a lot of this seems self-explanatory, but... what is the ign retard amount? And the fuel film?
What values should be used as defaults?
Retard ignition while accelerating. We had a chat about that & turbos some time ;) Fuel film looks to be a time based smoothing filter for evening out lumpy steady-state running. Good for big cams. I have played with it but haven't made my mind up yet about it...
Sometimes knocking take place at tip-in. I hope ignition retard here can help.
But regarding fuel film - it works perfectly for acceleration. Andrey described it here: http://www.vems.hu/wiki/index.php?page=MembersPage%2FAndrey%2FFuelFilm
Since this acceleration becomes as it should be: no lag, no overflood, crispy.
Quote from: [email protected] on March 05, 2010, 02:39:18 PMSo where are my cranking, afterstart and warm-up enrichments?
You can't miss them. They have their own tables.
Here is my config for 1.1.69 for my BMW M20 engine : http://media.vems.se/mattias/bmw-m20-1.1.69.vemscfg
I have an ICV installed, other than that it's a pretty basic setup. Just drag and drop my configuration file over the dialogs and tables you want to copy and you should be set to crank it.
Analogue input chanels
Is there any way of getting displaying the spare ADC inputs representing temperature of the sensor connected to a spare ADC chanel?
For example.
A configureable five point temperature scaling for each volt of the ADC input in vemstune creating a crude temperature curve, or, similar to the coolant and MAT sensor dialog tables.
It would be nice to have two ADC chanels with this for monitoring. A single round (which I intend to purchase) could then display a believeable value that actualy represents tempreature, rather than volts requiring manual scaling to temperature.
I spotted that there is some sort of ADC calibration tables, but I am not sure how this would work with temperature sensors wich are non linear.
Could be for mesuring oil temp or pre intercooler temp for example
I really don't want to have to resort to a seperate datalogging system when the possability of doing it with Vems is real
Quote from: mattias on March 05, 2010, 06:36:29 PM
Here is my config for 1.1.69 for my BMW M20 engine : http://media.vems.se/mattias/bmw-m20-1.1.69.vemscfg
I have an ICV installed, other than that it's a pretty basic setup. Just drag and drop my configuration file over the dialogs and tables you want to copy and you should be set to crank it.
I have to point out the drag and drop function - its awesome.
Open the window you want, go to the configuration file that you want to use on your file explorer, drag the file onto the target window, drop and the values are updated.
Brilliant.
Viewing logs is a major pain.
When I hit play, nothing happens. so I´m forced to drag the slider around and that has a resolution all on it´s own.
will this be updated to fixed soon?
Quote from: gunni on March 08, 2010, 03:04:28 AM
Viewing logs is a major pain.
When I hit play, nothing happens. so I´m forced to drag the slider around and that has a resolution all on it´s own.
will this be updated to fixed soon?
Give me a step-by-step bug report and I will make sure that the developers know about it.
Step 1 - Open log file
Step 2 - nothing but a the initial screen. I can only hit Pause and Stop, so it looks like it´s trying to play but nothing happens.
Step 3 - move the time slider around and when it stops it shows the gauges at that time
Response:
--8<-----------------------------------------------------------------------
Now the log playing functionality is removed from VT (we can add it again if it
is needed)
We can implement the following:
Separate the show on startup functionality to online and offline mode.
Add a new option in gauge group properties:
Show on startup offline
Show on startup online
So the default setting should look like:
online mode: the "Main" group should open automatically
offline mode: the Multi Graph should open automatically
but the user can change it.
Nevermind,
I just realized that the csv file works with Megalogviewer.
And I´ll continue to use that until VT has a better internal logger reviewer.
Besides gauge watching while looking for anomalies is not really good, graph trends make alot more sense
I don't see why it needs to be in the tuning app anyway.
I agree in a sense, but it will be good when vems has it´s own specialized logger review program that will look and feel similar to what
vemstune does.
Quote from: [email protected] on March 11, 2010, 03:59:12 PM
I don't see why it needs to be in the tuning app anyway.
It is very convenient to have the logger in the tuning program - if done correctly.
Autronic is the best example of this. You can have the tuning window and the log each sharing half of the screen.
(http://www.vems.hu/files/KevinBlack/SM2VEandLog65.gif)
One key "Z"turns VE or log into Full Screen
One key "Tab" changes between log and tuning,
One key (page up or page down)changes from one tuning window to next,
Box highlights VE/Spark/Lambda table cell for RPM/MAP that is in the Log cursor (*not seen in this pic)
- VEMS developers - if you are reading this - please please please do not disable MLV compatibility!
- In developing a log program to integrate into VemsTune please use Autronic as example of lightning quick useability and versatility.
- There are a hundred reasons why Autronic is the best (I love the MLV) for quick and easy custom scaling and display of log data - I could demonstrate during an on-line (teamviewer) session perhaps?
- Who here actually reads gauges for data viewing?? I can't think of a worse way to do it.
so you want something like this as a default view?
(http://media.vems.se/images/log.png)
This log was downloaded from a ECU with builtin logging memory
hotkeys can be customized to have mostly the same assignments as autronic
Some features will probably be missing for some more time:
Currently you have to save a config and re-upload after working with a offline log, will be fixed maybe in next vemstune testing release
You can't zoom in on the table and scroll inside the table editor, it automatically fits the screen size
ECU log files doesnt wrap back to 0 or stop in a couple of minutes, they keep on recording for approx 400 hours, only makes a new file for each engine startup
Some things is already much better compared to autronic:
All fields are always available on ECU logs, even injector duty cycle.
You can choose if you want 2D or 3D view in any place
2D and 3D has same orientation (up is always up in the table, unless you want to flip the 3d upside down)
note: This is just an example of a screen view, you can choose any ratio between the log viewer window and the table editor, if you just want all graphs in one section or split up as in my example
Keep it up Emil, very nice.
I like the way you think Emil :)
I like the MLV and your way of multiple (2,3,4) rows of graphs. Keeping it optional will make everybody happy.
One thing we can do to improve on yours AND Autronic is to copy the way MLV does the "Hiding Field Select"
- This allows you the option of hiding the log display channels, but being able to easily access them if using the mouse.
(Autronic use of keyboard was quickest for me)
Another feature I would really like is the ability to have the time reference be "zeroed out" by the log viewer - this will allow easy time/track position referencing when you are at zero seconds at the starting line launch - for example.
Another thing I really like about MLV is the ability to user-configure the log channel min and max, and especially to create their own channel names. Example: I can turn "Misc 1" (0-5v) into "Oil PSI"
If you make this user configured - you would have less people asking you to incorporate new channel names later on.
I'd like to play with these prototypes if you feel like making them available.
Thanks for the great work.
this is all available already, http://vems.hu/download/v3gui/VemsTune-Install-2010-03-08-testing.exe (http://vems.hu/download/v3gui/VemsTune-Install-2010-03-08-testing.exe)
log viewer is in shift-8 view
actually time is not reset, but you can easily just mark a section in your log and see the time of that, i think thtas good enough
just need to hit alt-e with focus on the log viewer, or right click, to have a menu where you can add and remove data, and also rename if you want to
Logger review
#1.
it says rgba(x,x,x,x) but when I select a color it comes up as rgb(x,x,x) and the color does not get applied.
I´ve added the "a" and (x,x,x,200) just to make it work after have selected a color.
#2.
View ratio of the gauges seem to be strange, coolant line will be almost unvisible unless magnified 60times. I think
megalogviewers min-max setting would work alot better, where the program automatically ratios the gauge to show max at the top of the line chart and min at the bottom, or add the ratio option to the MultiGraph Desc editor.
#3. Gauge selection process is rather slow, again comparing to mlv where I can simply click on the lines name in the graph menu on the left and select a new gauge, all about 2-3seconds to switch gauges.
Overall it looks good and can improve, I think that each gauges configuration should be setup specifically for each gauge, but the gauge selection process should be alot simpler, i.e selecting a gauge with Q and W , then press a button and a list with all the available gauges comes up and you can select one from there. Putting in a new gauge in a specific trend graph could be that you need to have selected a gauge in that graph and then hit Alt+N and then select a new gauge from a list.
I think the people who have used MLV alot have gotten accustomed to the extreme ease of usage it provides.
I find myself creating custom gauges to aid in tuning and I use the min/max limits alot to display for instance the
lambda and lambdatarget trends in the same graph, this way it´s easy to just see how good the tuning is by how close the lines are.
I also made a re calculated Gve line(from all the gamma trends and lambda/lambdatarget ratio) where I can monitor automatically re-calculated Gve values, so I only need to look at the new value and put that in the MAP, no hand calculating.
Question.
I had setup quite a few turbo charged cars with a 0-5v boost controller adjustment. Now it seems
this really nice feature has been disabled without chance of enabling it.
Is there a way to bring back the 0-5v boost control option?
Mostly used with secondary pwm settings.
Updated somebody to 1.1.70 today.
Two things to look out for.
Anytrim is set to fuel trim and Ch0, the curve is then set to 0.78 everywhere. Disable if you don´t need it.
Also MAP sensor placement is - before throttle body, while the norm is behind.
Quote from: Sprocket on January 22, 2010, 05:59:38 AM
Quote from: mattias on January 18, 2010, 06:27:49 AM
Quote from: Sprocket on January 18, 2010, 05:43:59 AM
If by setting the Injector Firing Divider to '1', Alternate Banks to '3..0', then Cranking Fire Banks While Cranking to 'All Banks' and then Cyl for Sequential, #Cyl/2 forSsemiSeq Cranking Fuel Divider to '1', will not add four times as much fuel at cranking???...
Like I said, the pulsewidth that is calculated from all the enrichments is divided over the 4-stroke cycle so the total pulsewidth is the same. To add fuel you need to alter the enrichement tables. The only thing you're doing with the cranking fuel divider is to decide how much to split the total pulsewidth over the 4-stroke cycle.
Ok, fair enough but I then fail to see the reason for 'cranking fuel divider' whether it is 1, 2 or 4 times per cycle resulting in only the exact same amount of fuel being elivered during cranking.
I thought 'fire all banks while cranking' turned the cranking injection sequence from sequential into batch and the divider was used to then divide the calculated PW by the number entered in 'cranking fuel divider' 1 for sequential, 2 for semi, and 4 for batch.
Its all a mystery to me ??? why do we have settings that give the same end result. Its a lot of un necessary configuration for no gain.
Im not being awkward, i just cannot see logic lol.
I have been doing some bench testing, more of a looksie see to be honest. I noticed that upgrading to 1.1.70 and then uploading my 1.1.47 config worked really well. I am impressed ;D. I also noticed that the cranking setting I had in 1.1.47 as 'fire all banks while cranking' has transpired into 'fire all banks while cranking' and a cranking divider of 2, which would go with my original theory that it would double the fuel at cranking without the need for silly cold crank percentages ;)
Im off to play some more :D
Quote from: gunni on March 08, 2010, 04:10:48 PM
Step 1 - Open log file
Step 2 - nothing but a the initial screen. I can only hit Pause and Stop, so it looks like it´s trying to play but nothing happens.
Step 3 - move the time slider around and when it stops it shows the gauges at that time
This bug has been marked as completed on 31 March. I guess a new set of executables will be out soon with the fix applied.
I´m not worried, Since MLV can accept the CSV output files from VT I´m happy. Don´t really care for the gauge viewed log review either.
But will test out the internal logger reviewer when it´s better.
I saw a screen shot of their GPS integrated viewer - it plays the log back with a little pointer on a map showing where the vehicle was at the time..
Screen shots of my ride.
http://img230.imageshack.us/img230/68/gps2.png
http://img227.imageshack.us/img227/3446/gps.png
(gps has read one point wrong)
We upgraded from 1.1.27 to 1.1.70 on our m50b25 n/a bmw project, but we have gone into some trouble. The car has wasted spark coils, and blow out fuses maybe at engine die. With 7.5A fuse was perfect on 1.1.27 but now a 15A blowed out too.
The another problem is the engine start hesitate and missfire at about 2000-2200 rpm.
Examined the config and we have one fault because we set after throttle at Alpha/N menu but I think this is not the answer about fuse blowing out. Has anybody experience about 1.1.70 or try some older fw?
this is the config:
http://www.cosworth.hu/misc/bmw/bmw-m50-1.1.70.vemscfg (http://www.cosworth.hu/misc/bmw/bmw-m50-1.1.70.vemscfg)
(engine run in batch injector settings and inj FET E was wrong so using H replace E)
Thx
what did you put for trigger tooth width?
Quote from: MWfire on April 08, 2010, 04:11:36 PM
what did you put for trigger tooth width?
you are right!
it was 24 and missing was 72 in 1.1.27
but now 4614 in 1.1.70
we checked all dialog twice but missed it.
trying with good value....
I have one customer running a M20 with 1.1.70 and wasted spark coils and no issues there.
Your angular width of tooth is 4614.00 degrees. Which we know can not be right :)
Your fuel table is also very strange since the ecu is setup for map multiplication.
Your lambda table could be alot leaner as well
Also not sure why your injector output table is not "normal"
Quote from: gunni on April 08, 2010, 04:58:29 PM
I have one customer running a M20 with 1.1.70 and wasted spark coils and no issues there.
Your angular width of tooth is 4614.00 degrees. Which we know can not be right :)
Your fuel table is also very strange since the ecu is setup for map multiplication.
Your lambda table could be alot leaner as well
Also not sure why your injector output table is not "normal"
Yes I have seen your config. Our car was not fully mapped, but a "bit" rich I know.
Injector "normal" output table means semi-sequential?
Which will be the ideal shifting of the injector output table against the ignition?
(Because no sec trigger, so ecu don't know which cyl on tdc)