• State TI Thinks Resolved
  • Locked Locked
  • Replies 9 replies
  • Answers 3 answers
  • Subscribers 14 subscribers
  • Views 501 views
  • Users 0 members are here
Support feedback
Options
Options
Related

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.

ISO1044: Chip burns often

Ricard Serra
Prodigy 30 points

Part Number: ISO1044

Hi,

I started to use your isolated can bus transceiver and it works fine but I needed to replace it time to time cause during startup it burns, I used the recommendd protection in the data sheet but I still have this issue, It seems when the board is powerup, depending of the other boards already powered connected to the bus causes the chip to burn.  This only happens at power on. If power on is success then all is working fine for hours or days.

I finalized the prototypes and I need to enter in a mass production but I'm really concerded about this issue. Attached find the schematic.

Please, help.

Manny Thanks,

Ricard.

ISO1044BD.jpg

 

  • 0
    TI__Mastermind 34580 points

    Hello Riccard, 

    Thank you for reaching out. The schematic looks good, and I do not thing the damage is an issue with your schematic. 

    You mentioned that the ISO1044 is more likely to heat up and damage when many nodes are connected shortly after power on. This leads me to believe two possible causes:

    • (1) There is a high in-rush current from your power supply that is potentially damaging IC's.
    • (2) The LDO that supplies the 5V rail is failing and applying 12 or 24V (from the LDO) input to VCC2 which then damages the device. 

    I would suggest the following next steps:

    1. Would you be able to monitor VCC2 with a Current probe to monitor the inrush current? 
    2. Can you please share the power supply schematic that connects to the 5V rails? 
    3. Please also share how man devices are connected to the 5V rail and the higher power rail that the 5V is generated from. 

    Best,
    Andrew

  • 0 in reply to Andrew J
    Prodigy 30 points

    Hi,

    Thanks for your response.

    Attached find shematic of the power supply. The 24V input is using the same power supply for the other CAN devices not isolated. U8 provides 5V to VCC2 of the TI CAN transceiver with its groud referenced to the ground of the other CAN devices cause 0V is common. U3 is isolated from the U8 and provides 3.3V to VCC1. Datasheet of U8 says no capacitor is needed. Can be the problem here?

    Many Thanks for your help... :)

  • 0 in reply to Ricard Serra
    TI__Mastermind 34580 points

    Hello Ricard, 

    Thank you for including the power stage schematic. I am not an expert on these parts, but the schematic looks okay here as well. The grounds appear to be properly isolated. 


    I think it is necessary to check if any of these power IC's are heating up or failing as well. Please check this in addition to my previous comments and tests from my first reply. 

    Best,
    Andrew

  • 0 in reply to Andrew J
    Prodigy 30 points

    Hi Andrew,

    I think I found where the problem is. Testing the board in home using 24V 1A power supply seems the chip never burns but when testing the board in the machine using a common power supply 24V 20A the ISO1044 chip burns often.

    My question is. It's possible that the inrush current goes throught the U8 and reach the can transceiver?

    If it is, can I put an NTC to the U8 input,what value you recommend?  this regulator is only used by the isolated transceiver.

    Or maybe different recommendation?

    Many Thnaks for your help.

    Ricard..

  • 0 in reply to Ricard Serra
    TI__Mastermind 34580 points

    Hello Ricard, 

    Yes, I think the inrush current is a possible cause for the damage. TI has a few documents and causes and Managing Inrush Current (Rev. A). That may be helpful. Most inrush current is caused by high capacitance on the power rails when turning on the system. The amount of CAN nodes that are being powered would contribute to a larger inrush current.

    Protection devices (like and NTC) should help; however, I would first recommend measuring the inrush current to confirm that it is the issue. 

    Best,
    Andrew

  • 0 in reply to Andrew J
    Prodigy 30 points

    Hi Andrew,

    With an osciloscope and a shunt resistor I can see up to 1.8A of insrush current at power up. So seems the problem is here.

    I have read the recommendations and a think I found a solution. I want to repalce U8 by a soft-start power supply like this one:

    https://www.digikey.com/en/products/detail/torex-semiconductor-ltd/XC6702D501ER-G/10175672

    What do you think?

    Many Thanks,

    Ricard.

  • 0 in reply to Ricard Serra
    TI__Genius 16360 points

    Hi Ricard,

    Thanks for confirming the root-cause issue in power stage.

    As Andrew has already mentioned above, we're not the right experts to confirm the selection of right power supply IC. If you have a TI power supply device in your selection list, then you can raise a new thread on e2e with the particular TI device name, and it will direct you to the right experts across TI.

    We definitely cannot comment on a non-TI device.

    Regards
    Varun

  • 0 in reply to Varun Kumar
    Prodigy 30 points

    Hi Varun,

    I tried but not found a TI soft-start 24V to 5V power supply. Anyway, I will try again.

    Many Thanks,

    Ricard.

  • 0 in reply to Ricard Serra
    TI__Genius 16360 points

    Hi Ricard

    You can raise an E2E query for the same as well with the closest available device from TI that you might have seen 

    It will still help you to reach the right experts and they can confirm whether TI has a power supply device with your exact requirements or not

    Regards
    Varun