Bli medlem i Norsk elbilforening og støtt driften av Elbilforum. Som medlem får du i tillegg startpakke, medlemsfordeler og gode tips om elbil og lading. Du blir med i et fellesskap som jobber for mindre utslipp fra veitrafikken. Medlemskap koster 565 kroner per år. elbil.no/medlemskap

Dead CAN bus

Startet av strandte, søndag 16. oktober 2016, klokken 23:55

« forrige - neste »

strandte

Hi,
I have for some time tried to breathe new life in my 2010 Think City Gen2 with lithium batteries. I bought it right after a workshop had found out that it needed a new PCU. I thought that I might figure out what was wrong with the PCU, and bring it back to life. So far I have learned a lot, but not yet found the source of the problems.
Maybe someone could give me some clues where to look next?
The history is like this:
The car started to need several tries on the key to finally start. In the end it didn't start after 100 full sycles of the key (This is what the old owner tells me).
The workshop investigates and tells the owner he needs to buy a new PCU for NOK 37.000.
The owner sells the car to me.
I use about a month to get hold of the Think diagnostic tools, but I'm not able to get communication with the PCU/ECU.
I pick apart the PCU and look for broken fuses. I onl find one broken fuse, and that is the FU7 which seems to belong to the Air Condition which my car does not have. That is strange, but I change it anyway, and put togehter the PCU and does a new test with it in the car. The symptoms are all the same: drive lock on (blinking red key on top center dashboard), red wrench indicator and no green car symbol.
I pick the PCU apart again and take out the processor board, since this is where the multi connector for communication between the car and the PCU sits.
I look for fuses on the circuit board (other than the obvious big red one which was not broken) and finds two small fuses near the multi connector. Both of them was OK. I try to do some diagnostics on the board without having been able to retrieve test procedures for the inside components of the PCU. This was not easy for me, but I find that only a small part of the processor circuit board, near the 12VDC power inlet, has power over the different small surface mounter capacitors (while the card is out of the PCU). I have also ohm'ed the +12 connection to be good all the way to the processor card, so I know that this board should get 12VDC.
I decide to buy a new processor board.
With the new processor board installed I still have the same symptoms!
Since I didn't get communication with the PCU/ECU even after changing the processor board, I decided to look into the PCAN tool to see what I could find out with that one.
The same thing happens again. The CAN bus seems dead. When I use the Comm Tool together with the PCAN Viewer I can see that I send CAN commands from the Comm Tool, but there are no answers from the devices which should communicate on the CAN bus. I can not see any traffic on the bus by using the PCAN viewer either.
I have a good indication that the diagnostic tools work, since the led(s) in both the Omnitech device and the PCAN-usb is lit and flashing. The drivers is reported to be working in the device manager.
I currently do not have the correct connector to connect to the connector underneath the fusebox in the car, so I do not know if this tool could give me any clues to what is wrong?
If someone knows what could bring down the CAN bus, I guess we would be on to something.
Any good tips on where to start is appreciated.

PRensel

#1
Hi,

Let me try to help you with this.

First of all the CAN bus is always dead :-)
You see the can bus is nothing more then a pair of twisted wires terminated with two R120 resistors.

Each component using the CAN bus send or waits for traffic on the bus, all in turn, meaning serial.
If you dont see ANY traffic at all on the CAN lets first measure the CAN itself.

Use a decent multimeter to measure the resistance of the bus, it should be R60. Easiest is to pull the plug from the PCU and measure directly on the plug, there are two pins used for the CAN, nothing else.

If it's not R60, but your measuring just R120 then the bus is not terminated on one end. If you measure anything else then R60 or R120 then one of the components might be jamming the bus.
If the latter is the case start pulling all CAN components from the bus and keep measuring until you find the jamming component.

But first try to see if the CANbus itself is properly terminated etc.

Let me know if you need  help.

Regard,

Paul



===============
My personal EV collection:
----------------------------
- Audi A2 Electric
- Citroen Saxo Electrique 2-doors
- Vectrix VX1 (in different colors)
- CityEl Fact4
- MiniEl Targa/Cabrio/Basic: red, red
- Th!nk A266 PIV4 (also for sale)
- Th!nk A306 City (cars, parts, service)
===================

strandte

Thanks for your answer!

Do you know which of the pins in the multi connector that are for the CAN bus? I can't find any documentation on that connector.

The documentation is a bit confusing as well since it seems to be for gen1 since it is a bit different than in my car.

I guess I could just measure the CAN bus in the ODB2 connector in the main fuse box? It's to late to do it today, so I have to look into it tomorrow.

Regards
Terje

PRensel

#3
I have the wiring diagram for the gen2, which one do you have ?

On the OBD2 connector pins are:
14 - CAN L
6 - CAN H

16 - +12V
5 - GND

7 - K-line
5 - Sig GND

On the PCU (62 pin connector) its:

7/62 - CAN L
8 /62 - CAN H

If you take of the PCU cable and look into the cable/connector with the bigger pins on top then pin 1 is lowest left, pin 62 is upper right (big pin nr 6)
So your CAN signals are on the second row from below, last 2 pins on the right, nr 7 and 8.

By taking of the PCU connector your disabling one R120 so you should measure R120 on the pin 7 and 8.
After installing back the PCU plug and measuring at the OBD2 pin 14 and 6 you should measure R60.

Take both tests and if other R values are coming up your problem is with the CANbus itself.
If both values R120 and R60 are ok, your problem lies in one of the components.
So next thing would be analysing the CAN traffic.

Regards,

Paul
===============
My personal EV collection:
----------------------------
- Audi A2 Electric
- Citroen Saxo Electrique 2-doors
- Vectrix VX1 (in different colors)
- CityEl Fact4
- MiniEl Targa/Cabrio/Basic: red, red
- Th!nk A266 PIV4 (also for sale)
- Th!nk A306 City (cars, parts, service)
===================

strandte

Thanks for great info regarding the multi connector!
I think we start to home in on the real issue. The PCU side of the CAN bus had 120 Ohm, but the cable side had only 84 Ohm. When I connect the multi connector to the PCU and measure from the ODB2 connector it shows 49 Ohm.
I just found the wiring diagrams for Think gen2 on a ftp server link found in the Yahoo USA group. I will start the troubleshooting tomorrow if I get time. Thanks for helping me in the right direction!

Regards
Terje

strandte

I have been disconnecting one by one of the devices on the CAN bus and measured the resistance on the bus. I thought I had found the source of the problems when I disconnected the VCU and the resistance on the bus increased from 85 Ohm to about 120 Ohm with the PCU disconnected. I connected the PCU and did a control measure on the ODB2 connector, which by now showed 60 Ohm. I thought I would get traffic on the bus when I powered up without the VCU, but that was not the case.
Does anybody know if this is normal? Do I need to have the VCU connected to wake up the systems?

By the way, I can't find the TPMS, SRS airbag ECU and the RAC (mindbox). Could someone indicate the location of these devices?

As I understand the wiring diagams the PAS, ABS and BMS is all connected on the PCU side when the multi connector on the PCU is disconnected? I measured 120 Ohm on the PCU side, so I have not investigated these systems.

Are there anyone that has some tips on further troubleshooting?

Regards
Terje

PRensel

Sitat fra: strandte på onsdag 19. oktober 2016, klokken 19:49
the VCU and the resistance on the bus increased from 85 Ohm to about 120 Ohm with the PCU disconnected. I connected the PCU and did a control measure on the ODB2 connector, which by now showed 60 Ohm. I thought I would get traffic on the bus when I powered up without the VCU, but that was not the case.
Does anybody know if this is normal? Do I need to have the VCU connected to wake up the systems?

By the way, I can't find the TPMS, SRS airbag ECU and the RAC (mindbox). Could someone indicate the location of these devices?

As I understand the wiring diagams the PAS, ABS and BMS is all connected on the PCU side when the multi connector on the PCU is disconnected? I measured 120 Ohm on the PCU side, so I have not investigated these systems.

Are there anyone that has some tips on further troubleshooting?


The RAC/mindbox was only installed on fleet car i believe. The TPMS i dont know if its fitted standard also, dont think so. Airbag is probably hidden under teh dashboard near the fusebox.

The moment where you measured R60 when was that ? what did it cause going from R85 to R60 ? Removal of the VCU ?

Do you have the Peak CAN bus adapter ?
If so, what do you see with pcanview (with the right settings) ?

===============
My personal EV collection:
----------------------------
- Audi A2 Electric
- Citroen Saxo Electrique 2-doors
- Vectrix VX1 (in different colors)
- CityEl Fact4
- MiniEl Targa/Cabrio/Basic: red, red
- Th!nk A266 PIV4 (also for sale)
- Th!nk A306 City (cars, parts, service)
===================

Orla Pedersen

The airbag controler is here I believe:



The carpet has to come out for access.
Mvh
Orla Pedersen

strandte

Thanks for informative picture Orla, I probably wouldn't have found the SRS ECU without it  :)

When I measured 60 Ohm between pin 14 and 6 in the ODB2 connector, the widest of the tree contacts on the VCU was disconnected. When this connector (on the VCU) was connected again the resistance fell to about 48 Ohm (When everything I know about is connected).
When I connected the PCAN_USB tool and looked with PCAN Viewer at 500kB I could not see any messages on the bus. When I at the same time start the Comm Tool also at 500kB, select a diagnostic file and click Get Data --> I see in the PCAN Viewer the messages that the Comm Tool sends on the CAN bus, but no answers.
Would someone be able to test if the CAN bus will work without the widest contact of the VCU connected? I believe it will work, but the reason I ask is that I see the wiring connections markes "wakeup" in the wiring diagrams. Could it be the VCU that wakes up the other devises on the CAN bus, so that the CAN bus will be dead without a working VCU?

Regards
Terje

Orla Pedersen

Sitat fra: strandte på torsdag 20. oktober 2016, klokken 22:09
Would someone be able to test if the CAN bus will work without the widest contact of the VCU connected? I believe it will work, but the reason I ask is that I see the wiring connections markes "wakeup" in the wiring diagrams. Could it be the VCU that wakes up the other devises on the CAN bus, so that the CAN bus will be dead without a working VCU?

Without one of either charge-wakeup (PCU J1-57), keyrun (PCU J1-47) or discharge request. everything is sleeping and no messages will be send.

I don't think termination is the problem......
Mvh
Orla Pedersen

strandte

#10
It has been a long time since I posted anything on this issue, but today I had some time to troubleshoot this. I found Chapter "G.7.5 Programming new EEC-V" in the Think Service Manual. It seems pretty simple to exchange to a new VCU/EEC-V, so I decided to try the VCU from my working Think in my Think with the dead CAN bus. The reason I wanted to test this is that with the old VCU the CAN bus didn't measure 60 Ohm.
With the working VCU in place the CAN Bus was still dead when I used PEAK Viewer and had the ignition key turned to run. The symptoms on the dashboard is just the same. The wrench lamp is still on and the PATS key is still blinking quickly. I even tried to program the VCU like it was described in the service manual, but since the CAN bus is dead I guess this would not work anyway.
I also tried the VCU from the nonworking car in the working car. The working car did now also show the quickly blinking PATS key light. My guess was that if the VCU was OK I would be able to program it by using my two keys one at the time, but this did not work. I wonder why? Have anyone been programing the VCU and know that this works? Would you please explain in details how you did it if it is not exactly like the procedure below:

disconnect 12v battery
remove old VCU, and connect new VCU
connect 12v battery
cycle one key to run and remove (out of the car)
cycle the other key to run

I must also add that the CAN bus on the working car did work with the VCU from the nonworking car, so it seems the problem with the CAN bus is not related to the VCU. I'm at least happy that I didn't just buy a new VCU :-)

VWspirit

Did you have your new main board reprogrammed for your key?
Bosted:Steine, Vestvågøy...*Heggedal, Asker
-----------------------------------
Jaguar I-Pace 2019
Tesla S100D Pearl White 2019
Think 2011 Sort 2+2 Li (bygd om fra zebra)
*Think 2009 Sort, Li (bygd om fra zebra)
*Think 2002 Sort, 3x25A 230V hurtiglader
*Citroen Saxo 2002 Blå met.
*VW Golf Citystromer 1998 56600km

© 2024, Norsk elbilforening   |   Personvern, vilkår og informasjonskapsler (cookies)   |   Organisasjonsnummer: 982 352 428 MVA