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.

SN74AVC16646 running hot

Other Parts Discussed in Thread: SN74AVC16646
 
I have a customer who is having issues with a SN74AVC16646 - it's running hot.
 
A little background. Originally they were using the Pericom PI74AVC+16646, unfortunately that device has gone obsolete. The SN74AVC16646  looks like it is a drop in replacement with the exception of a couple of timing parameters which does not affect them. However when the parts are placed on their production boards they run 20-30oC warmer. We see this with an increased current draw and we can measure it accurately with a thermal camera. There are a total of seven transceivers, 4 are on a data bus and 3 are on an address bus. The 4 on the data bus run hot, the ones on the address do not. We originally thought it was some sort of data contention but it does not appear to be that.
 
I know using the TI cross reference tool there is no cross for the pericom part but looking at the datasheet it's pretty close other than the timing parameters. We have done a lot of investigation and have many plots and data points but I thought it best not to overwhelm you with data - at least not now, incase there is an obvious reason we are missing.
 
Any help would be appreciated.
 
Cheers
Calum
  • Calum is correct.  We put some Texas Instruments SN74AVC16646 transceivers in place of the previous Pericom PI74AVC+16646 transceivers on an engineering board.  Our card drew about 0.9A from a 12V rail with the Pericom part, and now draws 1.1A with the TI part.  We noticed the TI parts were hot to the touch, so we shot the board with a thermal imaging camera and observed a 30 degree C rise over ambient.  The transcievers sit between an FPGA and some flash memory devices.  We are able to run read and write tests to flash memory without any failures, which tells us the TI transceivers are functionally working correctly.  The question is, why are they consuming so much current and thereby running so hot? 

    We can post oscilloscope screenshots of the clock input, control signals, or data input/outputs if it will help.  As Calum said, any help would be appreciated.  Thank you.

    Derek Mitchell

  • Hi Calum and Derek.

    I would like to see the plots of the input signals and get more details like frequency and are there any floating inputs.

    Can we take this to Email so all the right people can be copied and we can tranfer info easier . C-cockrill@ti.com.

    Thanks

    Chris