• State Resolved
  • Locked Locked
  • Replies 7 replies
  • Subscribers 34 subscribers
  • Views 295 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.

THVD1424: stops working above 80degC

Jim Pitman
Intellectual 410 points

Part Number: THVD1424

I recently respun a design to sub in your THVD1424RGTR RS485 device because it has programmable termination. That is the only change. This effort required changing the layout copper due to pinout differneces.

The previous design used a simpler device from Holt industries. The previous rev board with Holt makes it to 120degC.
That is, pinging this device, it returns a proper understandable respopnse.

Now tha we have changed to your device, it does not reurn a proper response above 82degC even though the spec claims 125degC as the upper limit. It works fine below this temperature.

Above 82dewgC it does respond, however some character in the response string is subtley different above 82degC.

Termination or not makes no difference.

Please advise.

 

  • 0
    TI__Mastermind 26032 points

    Hi Jim,

    Thanks for reaching out on E2E.

    Sorry to hear about this problem, THVD1424 should operate fine at ~80C. Is this behavior repeatable for other THVD1424 transceivers? Or is it only one transceiver that is behaving this way? 

    Couple other general questions:

    • Could you send a schematic over?
    • What data rate are you operating at?
    • This sounds basic, but what is your soldering procedure? Sometimes greatly exceeding our reflow rating while soldering can damaged the device.

    • Could you provide a waveform of the A-B pins alongside the TX, RX pins so I could see exactly how the transceiver is handling the response at high temps?

    Best,

    Ethan

  • 0 in reply to Ethan Sempsrott
    Intellectual 410 points

    We made up six assemblies. They pass out anywhere from 80 to 105degC. No tone makes to 125degC.
    I'm working on getting you waveforms. See attached schematic.

  • 0 in reply to Jim Pitman
    TI__Guru 72522 points

    Hi Jim,

    Today is a holiday. Please give some time for some feedback, thanks.

    Best Regards,

    Michael.

  • 0 in reply to Michael Ikwuyum
    TI__Mastermind 26032 points

    Hi Jim,

    I am catching up on E2E questions, but I will get a response to you tomorrow. Thanks for your patience!

    -Ethan

  • 0 in reply to Ethan Sempsrott
    TI__Mastermind 26032 points

    I do not see anything concerning with your schematic. Thanks for sending that over. 

    Is this behavior consistent across other units?

  • 0 in reply to Ethan Sempsrott
    Intellectual 410 points

    Hi Ethan. I was running on some disinformation. Apparently our existing design with competitor's RS485 device was also passing out above 80degC. Our new design uses your THD1424 and a higher temperature rated processor.

    We found a solution to our problem that may interest you and any other customers operating in a high temperature environment. I bought a software "sniffer" so we could observe the comm data in real time. I found the first four bytes of response data were getting corrupted whereas the remaining 32 that are a part of a proper response were not. These are an important four (part of a handshake) This was very consistent only and above 80degC. I hypothesized that when our processor switched from the receive mode to the transmit mode that not enough time was allowed before transmitting. Sure enough inserting a 1ms delay in the firmware after switching modes (disable receive/enable transmit) before transmitting solved the problem. We now make it up to 125degC consistently. This suggests that the there is some latency or change in rise time perhaps that increases with temperature. If you allow for it, all is ok. Feel free to archive and pass along. BTW other manufacturers' RS485 exhibit the same issue.

    Thankyou

    Jim Pitman
    GPMS Int'l

  • 0 in reply to Jim Pitman
    TI__Mastermind 26032 points

    Hi Jim,

    That is very interesting. Thanks for highlighting this. This actually seems very similar to a common RS-485 glitch which is present even at room temperature. Essentially at room temp, significant capacitance creates that same "handshake" error when switching from transmitting to receiving. I would theorize that removing capacitance on your RS-485 bus might also help with your initial transmission. 

    Here is the link to that app note I was referencing in case it helps: https://www.ti.com/lit/an/slla660/slla660.pdf 

    Best,

    Ethan