Mercedes Tech Talk Discuss general technical questions and issues about your Mercedes-Benz. Moderated by a certified MB Tech.
Sponsored by:
Sponsored by:

Need help with 2000 S-class CAN Issue, car at dealer

Thread Tools
 
Search this Thread
 
Rate Thread
 
Old 05-04-2006, 04:55 PM
  #1  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Need help with 2000 S-class CAN Issue, car at dealer

Ok so recently my moms 2000 S class was brought into the shop because it wouldnt start. it would power up, but not crank or start. Now ther car is under a extended warranty, so we really werent concerned. so We had it towed in to the dealership. Now I am very good friends with a tech there, so we specifically asked for him to work on the car.

well he found the car wouldnt start because apparently there was a CAN network failure caused by a faulty ESP module. well they installed a new ESP module and the car worked fine.

Now that same day my mom and I went to pick up the car and as soon as we started driving away from the dealer, an error message came on the instrument cluster for DISPLAY DEFECTIVE VISIT WORKSHOP. Now my mom was kinda pissed, but she took it back and left it again.

Now this all happened last friday. well we left the car again at the dealer on saturday and they would check it out monday. on monday the dealer called my mom and said the tech has been working with the car all day but have ran into a very strange problem and they have so far been unable to figure it out.

Well my mom, as weird as she is, went up to the dealer, I went with, and the shop foreman explained to her that my they cant communicate with several control modules on the car and that problem is only allowing the diagnostic equipment to view 11 control modules on the vehicle. Now after talking a bit with my friend, he said if everything is working properly the diagnostic equipment should access all control modules on the vehicle which is well over 30 modules, but he can only view 11. he said there is a major CAN failure, but they cant locate it. He said as of now, if I wrote it down correctly, the diagnostic equipment cant communicate with the transmission control module, the electronic selector module for the shifter, the ESP control module, and the EIS control module. he said they checked all fuses and all seemed to be working. he said they also checked power and ground to each control module and it was fine. then he said they disconnected each control module on the engine CAN and when they disconnected the fuel computer, he re-ran a quick test and everything started communicating again. then he said he hooked the fuel computer back up and everything was still working. Now the car is still at the dealer as they are making sure nothing is wrong, they have a tech specialist looking at it along with the shop foreman, but they are stumped.

Now they called back yesterday and said the problem re-appeared and they are trying to fix it, but are unsucessful as of yet.

I was looking for some help from a mercedes tech on her if there is one. any help would be appreciated as to what has caused this problem or what may be causing it.

Last edited by craft1; 05-04-2006 at 04:59 PM.
Old 05-05-2006, 07:04 PM
  #2  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
The 'display defective' is the key here- that means that the fuel computer is sending out faulty information on the Can. Usually it is caused by the wrong coding, but can't be your instance here, I wouldn't think (although I have seen instances where modules like the ESP get coded wrong and it corrupts everything)

It is fairly easy to narrow down using a lab scope- the SDS is far from being all powerful!! If all the coding is right, my guess is the ME (fuel) computer- but I would be scoping it to confirm before I changed anything.
Old 05-05-2006, 07:43 PM
  #3  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MBTech21
The 'display defective' is the key here- that means that the fuel computer is sending out faulty information on the Can. Usually it is caused by the wrong coding, but can't be your instance here, I wouldn't think (although I have seen instances where modules like the ESP get coded wrong and it corrupts everything)

It is fairly easy to narrow down using a lab scope- the SDS is far from being all powerful!! If all the coding is right, my guess is the ME (fuel) computer- but I would be scoping it to confirm before I changed anything.
what kind of equipment does the dealership use to "scope" the vehicle? and how does it work. I mean how does a "scope" narrow down the problem? I mean if this is something you know about, why doesnt the dealership do it, or better yet why isnt this dealership doing it. and how do you tell if the coding on the module, or modules is correct? and what exactly is coding?

the vehicle is still at the dealership. I talked with my friend earlier today and he said they still havent narrowed it down. Maybe I should ask him abut the lab scope.
Old 05-06-2006, 08:57 AM
  #4  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
You would use a lab scope to check this, if the tech doesn't have his own, then the shop would have the Hermann scope (although it is a little slow for engine CAN, it still works).
A scope displays voltage over time in a graphical display, so you can see the electricity as a picture. If I had my scope here, and not at work, then I would attach a picture of what a proper engine CAN signal looks like. Basically, it is bursts of voltage, as the control units talk to each other. It is supposed to be a certain voltage and speed- if it's not the CAN is corrupted. You would hook up to the CAN at the bus-bar, also known as an X30 connector. If it's corrupt, then you can disconnect each unit from the CAN at this bar- the second you disconnect the faulty unit(or branch of the CAN), then the signal will start to look like it should. Coding on the other hand will not impact the CAN signal, so that's what lead me to say that this is fairly easy to track- if the signal looks good on a scope, move on to coding.
Coding is the way that MB tells a control unit what it needs to operate correctly, ie: what type of car it is in, what country it is in, what protocol (language) the CAN uses!!, etc... On some control units- there can be as many as 100 or more settings that can be changed by the SDS. The worst part is, the only way to know what the proper coding is, is experience a lot of times- the SDS is wrong about production breaks,etc a lot, then we have those wonderful pages in German, that apparently didn't get translated. For example- on an instrument cluster, if you code the engine CAN protocol wrong- it can be 310 or 312, and the tip the SDS gives you is wrong- then you get the same 'Display Defective' message- with a lightning bolt picture in the cluster.
As far as why I know this, and they haven't done it, that is a whole different story. If you were to ask 20 techs, I bet 1 of them would actually own and/or use a lab scope. According to MB, fully half of all the techs in the country now came out of Elite, which means the most experience they could have is 8 years or so, if they were in the first classes. Worse, a lot of shop foremen and team leaders are picked for that position due to political nonsense, which means the guys who should be teaching and fixing the tough problems aren't in the position to do so. We all have our things we know best, and sometimes need help, but its a dog eat dog world in the flat rate world, and if you aren't getting paid to help/teach, why would you? Everywhere is not like that, but it's a lot more common than you would think.
Hopefully they got it narrowed down, because the National shop foreman meeting is this coming week in Alabama, and the tech specialists will be there too, so the help will be absent!! Email me if you need some more info, or if I can help.
Old 05-07-2006, 05:30 PM
  #5  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MBTech21
You would use a lab scope to check this, if the tech doesn't have his own, then the shop would have the Hermann scope (although it is a little slow for engine CAN, it still works).
A scope displays voltage over time in a graphical display, so you can see the electricity as a picture. If I had my scope here, and not at work, then I would attach a picture of what a proper engine CAN signal looks like. Basically, it is bursts of voltage, as the control units talk to each other. It is supposed to be a certain voltage and speed- if it's not the CAN is corrupted. You would hook up to the CAN at the bus-bar, also known as an X30 connector. If it's corrupt, then you can disconnect each unit from the CAN at this bar- the second you disconnect the faulty unit(or branch of the CAN), then the signal will start to look like it should. Coding on the other hand will not impact the CAN signal, so that's what lead me to say that this is fairly easy to track- if the signal looks good on a scope, move on to coding.
Coding is the way that MB tells a control unit what it needs to operate correctly, ie: what type of car it is in, what country it is in, what protocol (language) the CAN uses!!, etc... On some control units- there can be as many as 100 or more settings that can be changed by the SDS. The worst part is, the only way to know what the proper coding is, is experience a lot of times- the SDS is wrong about production breaks,etc a lot, then we have those wonderful pages in German, that apparently didn't get translated. For example- on an instrument cluster, if you code the engine CAN protocol wrong- it can be 310 or 312, and the tip the SDS gives you is wrong- then you get the same 'Display Defective' message- with a lightning bolt picture in the cluster.
As far as why I know this, and they haven't done it, that is a whole different story. If you were to ask 20 techs, I bet 1 of them would actually own and/or use a lab scope. According to MB, fully half of all the techs in the country now came out of Elite, which means the most experience they could have is 8 years or so, if they were in the first classes. Worse, a lot of shop foremen and team leaders are picked for that position due to political nonsense, which means the guys who should be teaching and fixing the tough problems aren't in the position to do so. We all have our things we know best, and sometimes need help, but its a dog eat dog world in the flat rate world, and if you aren't getting paid to help/teach, why would you? Everywhere is not like that, but it's a lot more common than you would think.
Hopefully they got it narrowed down, because the National shop foreman meeting is this coming week in Alabama, and the tech specialists will be there too, so the help will be absent!! Email me if you need some more info, or if I can help.
now I was talking with someone over AIM yesterday and he said that on the mercedes, the interior can has bus bars or X30 connectors where all the control modules connect to, but he said that the engine CAN is different, that there isnt a main bus bar, he said they connect at common splices, now is this true, or is it different? He also said with finding engine CAN issues, that you start by disconnecting each control module on the engine can. And if say you disconnect the ESP module and re-run a quick test, and everything goes back to normal, then you would suspect a faulty ESP module. Now if disconnecting each one doesnt have any effect, then you would most likely have issues in the wiring itself. Let me know if this all sounds about right.

Last edited by craft1; 05-07-2006 at 05:32 PM.
Old 05-08-2006, 07:29 AM
  #6  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
That is true about the bus bar- your car has a z-splice in the LF door sill. I would still check it with a scope as opposed to just disconnecting modules- like I said, if the CAN signal is normal, then that is eliminated and I'm checking the coding on every thing.
Basically what he told you is correct, but I would assume the dealer did this.
Old 05-08-2006, 04:56 PM
  #7  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MBTech21
That is true about the bus bar- your car has a z-splice in the LF door sill. I would still check it with a scope as opposed to just disconnecting modules- like I said, if the CAN signal is normal, then that is eliminated and I'm checking the coding on every thing.
Basically what he told you is correct, but I would assume the dealer did this.
Just out of curiosity, what does the Z-splice look like? the guy told me the interior can has a bus bar that has several connectors attached to it and you can disconnect each one if needed after the cover is removed, now will the Z-splice be the same way, or is it different?
Old 05-08-2006, 05:02 PM
  #8  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
No, they are twisted together and shrinkwrapped together. No reason you couldn't cut them if needed, as long as they are put back together properly.
Old 05-08-2006, 05:08 PM
  #9  
Super Member
 
sprink49's Avatar
 
Join Date: Nov 2005
Location: West Palm Beach
Posts: 788
Received 16 Likes on 11 Posts
2004 S500
Is this a great forum ...or what? Ten~15 years ago you would have a better chance of seeing Jesus walk in the front door than get this kind of feedback on a technical problem. This is great!!
Old 05-08-2006, 06:05 PM
  #10  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MBTech21
No, they are twisted together and shrinkwrapped together. No reason you couldn't cut them if needed, as long as they are put back together properly.
is there a reason for this? I mean isnt using a main connection a better idea like the interior can, thats what I would think. Just wondering how each control module would communicate with each other on the engine can if all the can wires were twisted together. you dont have a pic of the point do you? and what would you connect to to read the can signal?

and yes thanks for all this help, it is very good to have someone here who knows about this stuff.
Old 05-09-2006, 02:50 AM
  #11  
Junior Member
 
babason's Avatar
 
Join Date: May 2006
Location: london. united kingdom
Posts: 28
Likes: 0
Received 0 Likes on 0 Posts
C230 komp (202)
Hi everyone. First time user. Been looking for a website where i could chat, make friends and exchange technical info with technicians that know what they are talking about. Respect and regards to Mbtech 21, you are on the ball.
I am also not too confident using forums so any help, and advice would be helpful.

Advice and help regarding Mbs, willing to help. will say i dont know or unsure when in doubt, willing to try and find out, time permitting

Love and peace to everyone
Old 05-09-2006, 06:45 AM
  #12  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
I guess they figure it is a more secure connection, rather than a plug-in deal. As far as hooking to it, I use a device called 'bed of nails'- it has tiny needles on the clip that make self healing holes in the wire insulation (much like phone techs use). It doesn't matter with a can system how the wires are joined- they are all just together in a star type pattern. If a module needs the info it sees, it uses it, if not it is ignored. There is a High and Low circuit- redundant opposite signals, so if one side is open for example- the other side gets used.

The pic won;t upload for some reason- I'll try again later.
Old 05-09-2006, 04:42 PM
  #13  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
Originally Posted by MBTech21
I guess they figure it is a more secure connection, rather than a plug-in deal. As far as hooking to it, I use a device called 'bed of nails'- it has tiny needles on the clip that make self healing holes in the wire insulation (much like phone techs use). It doesn't matter with a can system how the wires are joined- they are all just together in a star type pattern. If a module needs the info it sees, it uses it, if not it is ignored. There is a High and Low circuit- redundant opposite signals, so if one side is open for example- the other side gets used.

The pic won;t upload for some reason- I'll try again later.
a pic would be great, thanks
Old 05-10-2006, 05:30 PM
  #14  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
The MB site won't let me copy the picture and upload it, and as luck would have it- I was staring at your connection today ( for another - Interior Can problem) and both my digital camera and phone were at home, getting ready for my trip to Alabama
Best way to describe it- take the drivers floor carpet out- there is plastic trough that is under there that the top snaps open on- the Z splices are about even with the front of the door opening- look for green wires (about 7, I think) all shrinkwrap together in a but style connection (like house wiring)-

How is the car coming anyway??
Old 05-11-2006, 05:54 PM
  #15  
Junior Member
Thread Starter
 
craft1's Avatar
 
Join Date: Apr 2006
Posts: 39
Likes: 0
Received 0 Likes on 0 Posts
well we finally got the car back. the problem was tracked down to a faulty fuel computer, or ME-SFI as the dealership calls it. I talked with my friend and he said they basically did everything MBtech21 said to do. they tapped into the engine can to begin with and at first he said everything was fine. he said signals all were good. then they checked it again when the issue came back, but again he said the info on the can looked good. But they still were not being able to communicate with anything. so again they disconnected the fuel computer from the can and everything starting communicating, but again upon hooking it up, it stayed fine for awhile. they finally decided to replace the fuel computer as they believed there was a intermittent short circuit inside of it. they said once they replaced it, the issue never came back. And we have it now and everything seems fine.

thanks again for the help.
Old 05-16-2006, 05:27 PM
  #16  
Senior Member
 
MBTech21's Avatar
 
Join Date: Mar 2006
Location: Dublin, OH
Posts: 302
Likes: 0
Received 0 Likes on 0 Posts
05 ML350, 85 300D
Happy to help..

Currently Active Users Viewing This Thread: 1 (0 members and 1 guests)
 


You have already rated this thread Rating: Thread Rating: 0 votes,  average.

Quick Reply: Need help with 2000 S-class CAN Issue, car at dealer



All times are GMT -4. The time now is 02:05 PM.