VEMS Community Forum
VEMS => Software => Topic started by: bart on September 30, 2010, 10:58:23 am
-
Hi all.
It´s sad that the first post i make is of a negative kind, problems, but disregarding that I will feed you some info & hope you can help me out?! I have searched for a solution & tried most of them, but none of them describe the problem I seem to have.... :-\
Installed the latest Vemstune version & was about to upgrade to 1.1.81 from 1.1.76 when something went wrong. Config saved & upload began, suddenly stopped & nothing.....
I checked for info & have already tried the "pin 2&3" solution and then running Terminal.exe to check, & this is what I get:
S,p,v -> AVREFI1
p -> S
When I try this without the "pin 2&3" before startup I get no respond at all in Terminal!
When starting up with "pin 2&3", double checking with terminal & then starting up Vemstune it´s still not possible to run a Firmware upload. I only get a message saying: It seems that there is some problem with Vems communication. Check power, cables & serial port connections.
On the log page I constantly get a message saying that I should connect ECU....
So from here I really don´t know what to do :'(
If any more details are needed, just let me know. Computers & software are not my strong points....
Thanks for a great forum :)
/bart
-
When you stay in bootmode Vemstune can't work, it's correct (if no, correct me..)
Anyway, you can download http://www.vems.co.uk/VEMSTech/VemsMT1.0.78rc2.zip, in boot mode (2&3 connected during power-up) try to flash this old 1.0.78rc2 FW to Vems from boot mode :
Note: check/modify COM port nr. in that BAT's
1. generate-config-16x14WithDefaults.bat
2. upload-firmware-boot-16x14.bat
3. after flash is done well, try to run MegaTune attached in that VemsMT1.0.78rc2.zip
This method works for me if the firmware gets wrong and only bootloader mode stays working.
-
Hi & thanks for the tip Pet, i´ll try it tomorrow!
Just a few things first....
The info I got from one of the support guys in Sweden, was that the firmware upload function should work in boot loader mode! Or have I managed to misunderstand this part maybe?!
As i mentioned in my previous post, anything regarding computers & fancy terms is just not my cup of tea, sorry :-\ it´s just like trying to read chinese.... ;)
When you say "flash", this means what? I just downloaded VemsMT & checked the files & folders, but sorry, I do not know what to do next really :-\
some more detailed help please ;) otherwise I just might make a mistake & the box will go up in flames... ;D
-
I had no trouble "reviving" a box with corrupted firmware a few weeks ago, I just use the short rx/tx trick and off we went to upgrade to the new firmware (with a good working serial port this time). VemsTune did in fact notify me that the ECU was in bootloader mode.
Make sure you use the latest VemsTune release, latest is from 2010-09-08 and usually there is a new one every other week - we are waiting for a release with some major changes which takes a while to test/develop.
Download here : http://www.vems.hu/download/VemsTune/
You can see which version you got from the Help menu -> About.
-
Hi Mattias!
Thanks for your support! I tried everything you mentioned & i searched this page for other tips, but nothing has helped so far...
As i mentioned earlier, when running Terminal.exe to check this is what I get:
S,p,v -> AVREFI1
p -> S
so, what should I make of this ??? positive or negative?
pssstt, Mattias, you know I will send you the box if I don´t manage to get it started myself ;)
-
What sort of Serial/RS232 port are you using - these can cause so many problems that its best to check that its a good one
One of the main problems with USB to RS232 ports is that they do not install their drivers correctly, or they pick-up the wrong driver. If your Com port changes each time you plug-in to the USB then you probably have this issue.
-
Hi Rob & thanks for your attention!
I used a usb cable before but had problems every now & then so I skipped it & now use a laptop with serial port. I have been using the 9 pin serial port since a while back & so even when performing this procedure I have used the serial port.
-
anybody, anything ??? no :'(
-
Sorry, theres not really much that I can suggest for you to do. If Mattias can help then I would speak to him.
-
If you get those answers from terminal mode then you should have no problem connecting with VemsTune and it should verify that the ECU is in bootloader mode and allow a firmware upload.
Remember, after successfully confirming that it is in bootloader mode you won't be connecting anything to the ECU until you quit the terminal program as it occupies the serial port.
-
Hi,
First check that you disconnected the terminal program as Mattias suggested above, but if that fails it seems like we have a situation VemsTune has problems recovering from. We'll check it out and we will try to identify the problem.
But for now we can use the old way to recover from a failed firmware upgrade.
Start by remvoing the fuses to the coils and injectors, the ECU can missbehave a bit during a firmware upgrade.
Download: http://www.vems.co.uk/VEMSTech/VemsMT1.0.78rc2.zip
Unzip to a folder of your choice and open upload-firmware12x12.bat in notepad.
Find this line: wintools\megaloader.exe firmware\vems.hex -wtE -c com1
Change to this: wintools\megaloader.exe firmware\vems.hex -wE -c com1
(-wtE changed to -wE)
Also change the com port to match the one you use.
Save the modified file and close notepad.
This makes the old style upload tool expect the ECU to be in bootloader mode (Something VT normally detect automatically.). Now force the ECU to enter boot loader mode by shorting pin2 and pin3 in the serial connector of the ECU while turning on the ignition. Remove the short and double click upload-firmware12x12.bat to run the script. Follow the instructions to "upgrade" the firmware.
If all goes well the ECU will now have a working 1.0.78 firmware, but one without a proper configuration. This is not a problem as this is only an intermediate step.
Now cycle power to the ECU and go back to VEMStune to upgrade the firmware to 1.1.81. When the firmware upgrade is done you need to upload the config file you saved from 1.1.76 before doing the upgrade. DO NOT use any config you saved from 1.0.78RC2.
I hope this helps.
Jörgen Karlsson
Gothenburg, Sweden.
-
Jorgens old way solved similar problems for me too. Vemstune has issues with recovery in bootmode.
-
hey guys! I will try Jörgens tip first thing tomorrow morning. Thank you Jörgen for describing it in baby steps for me, I really need that ;D so hopefully I will manage on my own tomorrow?!
ANd regarding Terminal, yes I did disconnect the program before starting up Vemstune, so you see I´m not totally lost ;)
I will try Jörgens tip & return with an answer ASAP!
thanks again!!
-
wehey!!! :D :D :D :D
My ECU is alive again!!! I followed Jörgens tip & everything went according to plan!
Another thing now instead occured....the car will not start at all, nope! Same values as before when the car did start, but now dead!
In the upper fields I can see a red bar indicating sec tr fault or similar, so I tried to perform an analyze/record trigger log, but can not manage to get any values to show?
Procedure: ignition on, Vems online/analyze record trigger log/V3/start V3 & turn key to crank engine & recieved bytes increases as time passes. I stop turning & stop V3, then press show results.....but get nothing. The files do get saved but how can I watch what problems i might have?
I will try to search a few more times & see if I can mangage to find something on it...
Thanks again for all the help :D
-
If the ECU has "secondary trigger errors", then check out the settings for that. Most likely the min/max angles are wrong, set these to 0 and 359 and to the default event in case of a bad trigger position.
I don't remember what kind of trigger you have but some have not been supported yet. Missing-tooth has always worked. Simple coil-type has been known to not work.
-
Hi Mattias!
I have the 60-2, the same one as Fredde on his spq, the one thats mounted between the flywheel & the block on the crank.
these are the fault codes I get when the engine is running:
Sec. tr. pos. bad
Too many pr. trig
I saw that the min/max angles you mentioned were set to 359/359, so I changed min to 0. The Afterstart turns on & off, BUT this may be a result of the triggering error I have. I get spark cuts now & then and the indication (when earlier checked at a local company) that my primary trigger looses signal & the rev indication drops down to cranking revs (below 299rpm)& this seems to make the car enter cranking settings making the car jump between 30 & 9 degrees every time I get the spark cut.
In other words, it seems as if i loose the primiray trigger signal for some reason & this causes the erratic behavior with the spark, but I can not run the analysis of the triggers. Right now I am waiting to check everything with an oscilloscope to figure out where the fault might be....
As I asked earlier, why cant I get the alaysis of triggers to function??
-
It should work out ok for you.. Just to make sure, install VemsTune 2010-10-01 (most recent today) which I used yesterday to successfully make a triggerlog with the same trigger setup that you have.
http://www.vems.hu/download/VemsTune/
You may find that you should change the trigger edge of the secondary trigger to offset it a bit, because it must never occur at the missing teeth of the primary trigger. Here you see a valid position for the cam sync (secondary trigger) from yesterday:
(Audi S2, 3B engine with 60-2 Hall sensor + stock cam sync in distributor )
(http://www.savarturbo.se/~mattias/motor/vems_data/vemstune-dev/triggerlog-60-2-and-camsync.png)
-
ok, domnloaded the latest version & will install right away!
Ok, now I at least know how the log should look ;) I will make a try myself & update here later during the day!
-
Ofcourse nothing should be easy, so why should it work this time....nope, I had to do the procedure all over again with bootloader & terminal & firmware upload just as Jörgen mentioned earlier.....booring! At least now I know how to!
Well, here is my trigger log, nice isn´t it ;)
(http://forumbilder.se/images/d91120101111599e47.jpg) (http://forumbilder.se)
maybe you guys can see the missing spark i get now & then, other then that I can´t see that my secondary trigger should be causing any problems, right?!
So, i need to figure out why i get these misfires. You notice them much better the hotter the engine gets since the revs get lower. At start I got about 1250 & @ warm engine about 900. SO yes, the lower the revs get the harder it is for the engine to stay alive when misfiring.
Is there any way I can analyze what is wrong with a trigger or must I wait for my friend to bring his oscilloscope?
-
Why did you have to do the procedure with the "bootloader & terminal & firmware upload" again ?
You already had 1.1.81 firmware uploaded to the ECU, right? Why upload again ?
What follows has nothing to do with why you created the thread in the first place, but we can continue to discuss..
Make the triggerlog at cranking rpm (~ 200 rpm). Press WOT to use flood clear to disable the injectors, and then crank the engine 4-5 seconds.
If you make it at idle (1000 rpm) then the tooth frequency is too high, communication speed to the tuning laptop/computer is too low to handle it, and data will be missed. VemsTune should tell you this is happening, but it doesn't - maybe in the future.
In case you did make the triggerlog while cranking the engine, from what I can see, at times you have a total lack of signal from the primary trigger. No need for an oscilloscope to tell you that.
Is the sensor of Hall or VR type ?
If Hall, what voltage source do you feed the sensor?
Which pin on the ECU? You should use EC36-pin28 (not 29 which is TPS only).
-
??? I have no clue why...maybe i screwed something up while updating to the newer version of Vemstune, but everything workes so far & I got more practice doing it ;)
Mattias, I´m away working now for 2 more weeks so I will give you an update on your questions as soon as I get home. I can at least answer the following.
the trigger log was made during idle speed. I will perorm another one during cranking.
The trigger is of a HALL type, the same one Fredde has on his spq, if you know which one I am referring to?! The sensor is connected to a separate pin, not the one used for the TPS.
Before unbolting my gearbox & clutch I just want to be sure I do not have any problems with the cables or any distortion of any other kind, therefor the use of an oscilloscope.
Have a nice week :)
-
Regards goes to Jorgen from me as well. Had to use his procedure when upgrading the box, and out of the blue just totally crashed.
-
Jorgens method is more direct than VEMSTUNE. If things go bad with this method, then just serial adapter or board can be the reason (if board supply is OK).
In such cases I tried 4 combinations with -wE, -wtE and board in usual and in boot mode. And different USB adapter.
Beware any signals in trigger inputs. FW changing is not possible with trigger impulses.
-
So you need to disconnect the triggers as well? No where i could read this until now.
Never the less, in the end it was successfully upgraded with the trigger cables connected.
I always use direct serialport, i dont trust USB converters..
-
Real signals, not wires. It is built-in protection.
Sometimes things goes bad with native serials ports too.