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 and PIC18F2xk80

Other Parts Discussed in Thread: ISO1050

Hey,
 
i have a big problem with the CAN but only with the PIC18F2xk80 and the ISO1050. The CODE and the settings are ok, they works on me testboard with a conection to the CAN Demoboard for the 18F or when i youse a MCP2551 as Transiver. But when i change this CAN Transiver to a TI ISO 1050 it dosent works. I need the isolation of the other part.
 
When i youse a dsPIC30F4012 in the same Layout the circut with the CAN works....
 

The TX Chanel is the problem, the flags are not ok. When i conectet a MCP2551 (onlyTX and RX in not CAN) on the same PIC togehter with an ISO1050 the ISO1050 works on the CAN side!


 
whats happen??? have same body anny idaers?
 
Thanks for help
 
Thorsten

  • Hi Thorsten,

    Can you please show an oscilloscope shot of what is happening on TX, RX, CANH and CANL when you say the ISO1050 isn't working? Also, can you share a schematic of your system? 

    Thanks!
    Jason Blackman 

  • Hey Jason,

    i attach a PDF with a lot of shots, when you need  special shoots plaes ask.

      I try a lot with RC but nothing helps. The Schematic is simpel, only TX to TX and RX to RX. When i youse the dsPIC30F4012 in the same Layout it works.

    Thanks for your help

    Thorsten

    ISO_1050.pdf
  • Hi Thorsten,

    We may need more information to help you debug this.  Schematically you show the 8 pin DW version of the ISO1050 CAN transceiver but no details on power supply or GND connections.  Is GND2, but CAN bus ground connected to the rest of the GND for the CAN bus?  Also, as Jason asked for we would need a 4 channel scope shot of TXD, RXD, CANH and CANL to really see what is happening on all 4 sides of the CAN Transceiver.  Also, is the output of this MCU open drain?  The CAN transceiver might have a weak pull up in TXD to keep the device recessive if TXD floats, however the ISO1050 with the isolation has a TXD input without a pull up.  Some times under nominal conditions the internal pull up on CAN transceivers is enough to meet CAN bit timing on TXD but across process, voltage and temperature corners it is not.  In the case of ISO1050 it will certainly not be strongly pulled up?  The second scope shot shows a very slow rise time on TXD which would imply open drain output on the uP.  Have you tried a 4.7k or 10 k ohm pull up? It is up to the uP CAN controller to provide the adequate bit timing input to the CAN transceiver.  This is also indicated in the final scope shot of the MCP2551 where the TXD input bits are highly rounded indicated likely dependency on the weak internal pull up of the CAN transceiver, normally on a push/pull output from the uP or a system with a proper pull up on TXD these would be very square at 250kbps. 

    -- Scott

  • Hi Scott,

    thanks for your insperation!!! Now it works.

    I try before only a pull up with 10k and it dosen't works, now i try 4k7 and it works.

    Thanks for your idea

    Nice Weekend

    Thorsten