PicoScope 7 Automotive
Available for Windows, Mac, and Linux, the next evolution of our diagnostic scope software is now available.
j1939 v-can network
j1939 v-can network
i am having an intermittent issue with the can dropping out but it only affects the transmission. it flags a code for primary network interface drops out. the trans tcu has been replaced along with the controller i have completely tested the j1939 v-can network and all the wiring tests good. i have the pico scope connected to the v-can and do not see any issues, so i am trying to isolate control units to see if they are interfering with the network. can this be done with the pico scope?
Re: j1939 v-can network
Picoscope is an measuring device, not can jammer, so it will not isolate anything, unless I misunderstood your question?
-
ben.martins
- Pico Staff Member
- Posts: 543
- Joined: Tue May 16, 2017 1:02 pm
Re: j1939 v-can network
Hello jerp,
Thank you for your post and as liviu has stated, Pico is a measuring device which allows us to visualise the bus and the data but we cannot broadcast anything onto the bus unlike dedicated CAN logging tools.
However, we can get a better idea about what ECU's are on the V-can by utilizing the J1939 CAN decoder. You can see more in depth detail on this here - viewtopic.php?p=102564#p102564.
This decoder will allow you to split the header up into priority, PGN and source address. Whilst the PGN information is locked away in the J1939-DA document the source address information is available on the following website - https://www.isobus.net/isobus/sourceAddress
Make use of the math channels A+B and A-B where CAN H and CAN L are on channels A & B respectively. This can show you issues that are affecting the overal voltage levels as well as anything disrupting the actual packet the ECU see's.
An example of this can be seen below -
Here you can see that A+B is showing a disturbance when we should be seeing an almost straight line fixed around 5V. Some tolerence needs to be allowed here but you can spot the obvious interference. You then use A-B, seen here in pink, to determine if this issue is in fact going to cause a problem with decoding. By using A-B, in dark blue, this allows us to see the message as the ECU sees it. As you can see despite the physical data in both A and B being seemingly distorted, the power of a differential signal shines through and when removing one from the other we still have a clear edge to decode on. There is some interfernce though on the A-B so if this got worse would it trip the threshold levels that has been set by the ECU manufacturer?
As mentioned we can then use the J1939 decoder to determine if this issue is affecting just one ECU. Looking at the source address tells us where the packet came from which in this case is 2. Looking on the ISO BUS website we can see 2 is for the turbocharger. Scrolling through the packets we can attribute this interferece only being present each time the turbocharger is transmitting data. This gives some direction as to where the issue lies.
The best way to see if a component or ECU is affecting a network is to remove it. Make sure if removing an ECU that it doesn't contain the terminating resistor. Sometimes it's easier to trace the power supply and remove that rather than the disconnect it completely. Remember that components attached to the ECU have the ability to affect the CAN BUS as per the following case study https://www.picoauto.com/library/case-studies/no-audio where a shorted DAB aerial was pulling the bus idle voltage down.
Whilst PicoScope is not a dedicated CAN logger it is an does have some logging ability. I hope this helps and if you need help with the captures or decoding, please feel free to post your captures here.
Kind regards
Ben
Thank you for your post and as liviu has stated, Pico is a measuring device which allows us to visualise the bus and the data but we cannot broadcast anything onto the bus unlike dedicated CAN logging tools.
However, we can get a better idea about what ECU's are on the V-can by utilizing the J1939 CAN decoder. You can see more in depth detail on this here - viewtopic.php?p=102564#p102564.
This decoder will allow you to split the header up into priority, PGN and source address. Whilst the PGN information is locked away in the J1939-DA document the source address information is available on the following website - https://www.isobus.net/isobus/sourceAddress
Make use of the math channels A+B and A-B where CAN H and CAN L are on channels A & B respectively. This can show you issues that are affecting the overal voltage levels as well as anything disrupting the actual packet the ECU see's.
An example of this can be seen below -
As mentioned we can then use the J1939 decoder to determine if this issue is affecting just one ECU. Looking at the source address tells us where the packet came from which in this case is 2. Looking on the ISO BUS website we can see 2 is for the turbocharger. Scrolling through the packets we can attribute this interferece only being present each time the turbocharger is transmitting data. This gives some direction as to where the issue lies.
The best way to see if a component or ECU is affecting a network is to remove it. Make sure if removing an ECU that it doesn't contain the terminating resistor. Sometimes it's easier to trace the power supply and remove that rather than the disconnect it completely. Remember that components attached to the ECU have the ability to affect the CAN BUS as per the following case study https://www.picoauto.com/library/case-studies/no-audio where a shorted DAB aerial was pulling the bus idle voltage down.
Whilst PicoScope is not a dedicated CAN logger it is an does have some logging ability. I hope this helps and if you need help with the captures or decoding, please feel free to post your captures here.
Kind regards
Ben
Re: j1939 v-can network
thank you for the replies. i now have a better understanding of how this works. i will now try to duplicate this issue while reading data to see if i can narrow it down.
Re: j1939 v-can network
Feel free to upload your recordings and/or post some screenshots if you're struggling with this one. A few of us here know J1939 pretty well and the Pico employees on here (like Steve) are quite knowledgeable.