This thread has been locked.

If you have a related question, please click the "Ask a related question" button in the top right corner. The newly created question will be automatically linked to this question.

ISO1050: ISO1050: Communication fail issue

Part Number: ISO1050
Other Parts Discussed in Thread: ISO1042

hi, TI team

We meet a issue about CAN communication,you can find the detail in attachment

hope your response,and any clue will help us quite a lot .

Many thx in advance.

CAN Failure isse_240112a.pdf

  • Hello Li,

    Thank you for reaching out.

    In the attached document it looks like your team has performed and A-B-A swap. Where you have removed the "bad unit" and replaced it with a new good unit. After this swap the PCB is functional. Is this correct? 

    If the PCB works using a new unit, place the "bad unit" on a different PCB and see if the failure is following the unit. Let me know the results. 

    Otherwise, from a schematic point of view everything looks good. I did notice that the CAN-side 5V rail does not have any decoupling capacitors. A capacitor placed less than 2mm away from the supply pin is usually recommended for normal operation.

    Best,
    Andrew

  • Hi, Andrew

    I am very gald to receive your so rapid reply,many thanks

    yes,after we replaced the ISO1050 with new one ,the PCBA is functional。

    We we sway the ISO1050 between functional PCBA and failure PCBA ,the failure follow with ISO 1050(bad unit).

    About your warmly feedback on schematic , i add our VCC2 DCDC in attachment ,There are 0.1uF and 10uF CAP for

    decoupling ,as your information .

    DCDC spec only in Chinese,but I think the graph is enough for your information ^_^

    MORNSUN_B_S-1WR2.pdf

  • Hi, Adrew,

    I try more today  ,   the failure still follow ISO 1050 when sway ic between PCBA

    I'd like to  send  these bad ic samples to TI , could you help to arrange some Failure analysis ?

    or check whether these are fake or dummy one?

  • Hello Li, 

    First off, I just wanted to make you aware that this is a public forum, so if you have any concerns about sharing your schematic then you may consider taking that information down (or we can for you). 

    As for the abnormal behavior you are seeing, is this just a single device, or is this many devices that are having this behavior.

    Is it possible that there was some handling or ESD that caused damage to a single device instead of there being a systemic issue with the ISO1050 in your system? 

    As for checking for counterfeit devices, I would ask that you make sure to buy devices only through TI.com or through an authorized vendor of our products.

    https://www.ti.com/ordering-resources/faqs/purchasing-online/authorized-distributors.html  

    We can certainly check to see if its counterfeit if you can share a picture of the topside marking of the device.

    Best,
    Michael

  • Hi,Michael

    Share this part of Schematic is ok for me , which is in the 《CAN Failure isse_240112a.pdf》  attachment in original topic infor

    This abnormal behavior has been found about 15pcs ,  but never found in prvious batch .

    I think maybe not the handling or ESD issuse,because I use a new 1050 IC to change onto the failure PCBA ,but the PCBA is

    still failure after with the new 1050, I replace another new 1050 instead ,the PCBA is functional, work normally。

    so I think handling or esd factors can be ignored。

    many thanks for sharing the offical vedor list for me, I will forward these infor to our supplychain team.

    attach some failure sample marking in below。

    BTW,Michael, can we setup a communication with e-mail?My E-mail address:chengli@gtsonic.cn

    Many thanks and  Be sincerely.

  • Hi Li, 

    Let me review this information and I will get back to you tomorrow. 

    Best,
    Andrew

  • Hello Li, 

    Unfortunately, we are unable to confirm the validity of these lots. Please include the following:

    1. High quality pictures of the bottom & top of the device(s).
    2. High quality pictures of the 2D bar code labels.
    3. Copies/pictures of shipping documents.
    4. Can you please provide which distributor you are using? 

    Best,
    Andrew

  • Hello Li, 

    Sorry I didn't bring this sooner, but I overlooked some critical points in your original attachment (see below questions).

    • The failure in your original post is unclear. The picture to the left looks normal, can you please circle the area that is failing? 
    • Also, when reviewing the schematic again, is the side two powered using VCC2 = +5V and GND2 = -5V? 
      • If this is true, then it the -5V rail would be violating absolute maximum conditions, and the device would be damaged. 

    Best,
    Andrew

  • Hi,Andrew

     

    Thanks for your detail check and response.

     

    For your 2nd question

     

    HW engineer just use netlabel "-5V"instead of "VCC2 GND"the actual voltage of it is 0V

     

    I add this part of schematic and DCDC spec in attachment in below

     

    For your 1st question

     

    If left signal wave looks normalbut this PCBA can't setup communicaiton with other CAN node deviceattach some

    information from CAN TOOL in below.

     

    Many thx again for your warmly support

    金柯_B0505S-1WR2.pdf

  • Hi,Andrew

    For item1

    I put 2 more high quality pictures in below, double side .

    For item4

    We purchase the whole PCBA  from JiaLiChang ,website for your information

    www.jlc.com/.../vcontact.html

    for item 2,3 I will check with my vendor.

  • Hi Li, 

    Just to recap the issue so far: 

    Your PCB has an issue with the CAN communication not connecting. You replace the ISO1050 (bad unit) with a fresh device and the board works. The bad unit is then placed onto a working PCB and then the PCB fails. This established that the failure follows the unit. 

    Here is what we have checked so far with my suggestions for each below:

    • The schematic looks good, and we confirmed that the device has the proper supply rails.
    • From what we have seen so far these seem to be valid parts. However, I cannot verify with our counterfeit team until I have items 2 and 3.
    • The ISO1050 appears to be functional. I am assuming ISO1050 is operating normally since it seems that TXD is transmitted to the CANBUS (TXD and CANH-CANL waveforms match).
      • Since the device looks to be functional, I am assuming the issue is not with ISO1050. To continue your debug, I would recommend looking at either the MCU/controller on the PCB or one of the other nodes. 
    • The whole PCB is assembled by JLC PCB
      • It sounds like this company is the one that is sourcing the parts for your board. 
      • You will have to reach out to JLC to see if they are sourcing their parts from an approved vendor.
      • Unfortunately, I cannot provide further support if these parts are not purchasing these devices direct from TI or an approved vendor.

    I hope this helps.

    Best,
    Andrew

  • Hi, Andrew

    Is it possible the failure was introduced from software ?

    Protocol or communication data relevant or some setting issue

    Then design is not robust enough or some buffer is quite tight which cause to the PCBA have to pick up some particular 1050

    unit ?  In spite of  during A-B-A swap, it seems the failure follow IC,but the root cause is other function ?

  • Hello Li,

    From what I can see in your wave forms, the CAN signal is always following the TXD signal. This would mean that the device is not failing. If I have misunderstood, then please show me where the device is failing on the wave form.

    If the device is operational and the failure still occurs then, yes, it is possible the failure to be introduced by software.  

    Best,
    Andrew

  • Many thanks  ,Our team will check the issure further more,

    and any update ,I will synchronizate the resule here。

  • Hi,Andrew

    We find some new infor ,it seems 1050 go into DTO mode, I summarize the wave and debug infor in attachment,ples help take a check.

    thanks

    CAN Failure isse_240122a.pptx

  • And I check more ,find that  for next frame, DTO alway activated among 420um(4bit)

    and according to Spec,Ttxd_dot show from 1.2 to 4ms ?

  • Hello Li, 

    Thank you for the detailed information your patience in solving this issue. reviewing your information and discussing with my team it looks like 

    The ISO1050 has gone through a revision and the datasheet was updated last year (October 2023). The current datasheet is on TI.com which you have been using. These changes are not applicable to your devices and will still follow the specs of the previous datasheet.

    You can find the previous revision attached here: slls983j.pdf.

    In the power point you mentioned the data rate of your application, 10kBits/s, which is too slow according to the previous datasheet. As you can see the typical value of 450us now lines up with your measured value of 420us. See below table from previous revision: 

    I hope this helps and sorry for the inconvenience. 

    Best,
    Andrew

  • Hi, Andrew

    We change bit rate up to 50KHz,the CAN communication works well.

    The root cause is DTO value definitely.  Many thanks for your team support recently.

    But our device need to work with 10KHz accoding to our customer requst and product definition.

    1050 meets the requirements of ISO11898-2 ,  should support low speed (10KHz,15KHz,20KHz)

    So I think DTO issue of  previous 1050 should define as a bug , Am I right ?

     The otherwise , How could I purchase the latest version 1050 with longer DTO (1.2--4ms) ?

  • Andrew

    Could you help me to setup purchasing channel about latest 1050?

    How can I describe this lastest version to distribtor? Lots number? Version number?or New ordering number?

    Our company need to purchase the latest 1050 correctly.

    Many thanks in advance !

  • Hello Li, 

    ISO 11898-2 is a high-speed CAN transceiver which ISO1050 complies to. There is a separate sub-section of the standard ISO 11898-3 for low-speed CAN communication and therefore this is not a bug.

    The PCN process will release an updated datasheet first and then new devices will begin to ship after a waiting period. Currently we are in the waiting period and the newer devices have not started shipping yet. 

    Since you need something in the near future, I would recommend looking at ISO1042 Isolated CAN Transceiver which has a DTO of (1.2 - 3.8). 

    Best,
    Andrew

  • Hi, Andrew

    Many thanks for your information sharing,our team will study ISO1042 intead of ISO1050,  It's a little pity

    they arenot Pin2Pin compatible.

    Could you give me the purchase information about new version ISO1050 when it started pre-ording and shipping

    period ?(My E-mail address : chengli@gtsonic.cn)

    Several of our product have passed the certification, can't change CAN tranceiver easily.  Really need

    the newest ISO1050 for manufacture and mass production.

    best wishes , and thanks again about you and your team warlmy support recently!

    Chengli

  • Hello Cheng, 

    We are happy to help. I will update you when I have more information on the new material, and I will go ahead and close this post for now. Feel free to open a new thread if you need anything else.

    Best,
    Andrew 

  • hi, Andrew

    Got it .

    Thanks quite a lot , be sincerely.