Has anybody used these? We have two FPGAs on a board, a Zynq and an Artix7. We want to use the internal ADCs to read chip temperatures. We have been advised to ground the temp sense diode pins DXP and DXN "if they are not being used". Unless the XADC has a separate temp sense diode, seems to me that shorting the external diode pins might kill our ability to acquire temperature internally. I don't see any big downside to not grounding those pins. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
temperature sense diodes in Xilinx 7 series
Started by ●March 3, 2017
Reply by ●March 3, 20172017-03-03
On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin <jjlarkinxyxy@highlandtechnology.com> wrote:>Has anybody used these? > >We have two FPGAs on a board, a Zynq and an Artix7. We want to use the >internal ADCs to read chip temperatures. We have been advised to >ground the temp sense diode pins DXP and DXN "if they are not being >used". > >Unless the XADC has a separate temp sense diode, seems to me that >shorting the external diode pins might kill our ability to acquire >temperature internally. > >I don't see any big downside to not grounding those pins.After too much research, it seems that there are independent temp sensors, the external-pin one and another inside the XADC. -- John Larkin Highland Technology, Inc picosecond timing precision measurement jlarkin att highlandtechnology dott com http://www.highlandtechnology.com
Reply by ●March 4, 20172017-03-04
On Fri, 03 Mar 2017 14:00:17 -0800, John Larkin wrote:> On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin > <jjlarkinxyxy@highlandtechnology.com> wrote: > >>Has anybody used these? >> >>We have two FPGAs on a board, a Zynq and an Artix7. We want to use the >>internal ADCs to read chip temperatures. We have been advised to ground >>the temp sense diode pins DXP and DXN "if they are not being used". >> >>Unless the XADC has a separate temp sense diode, seems to me that >>shorting the external diode pins might kill our ability to acquire >>temperature internally. >> >>I don't see any big downside to not grounding those pins. > > After too much research, it seems that there are independent temp > sensors, the external-pin one and another inside the XADC.And they recommend grounding DXP and DXN to improve the ESD rating. IIRC there's no functional issue caused by leaving them open. Regards, Allan
Reply by ●March 4, 20172017-03-04
On 04 Mar 2017 10:52:06 GMT, Allan Herriman <allanherriman@hotmail.com> wrote:>On Fri, 03 Mar 2017 14:00:17 -0800, John Larkin wrote: > >> On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin >> <jjlarkinxyxy@highlandtechnology.com> wrote: >> >>>Has anybody used these? >>> >>>We have two FPGAs on a board, a Zynq and an Artix7. We want to use the >>>internal ADCs to read chip temperatures. We have been advised to ground >>>the temp sense diode pins DXP and DXN "if they are not being used". >>> >>>Unless the XADC has a separate temp sense diode, seems to me that >>>shorting the external diode pins might kill our ability to acquire >>>temperature internally. >>> >>>I don't see any big downside to not grounding those pins. >> >> After too much research, it seems that there are independent temp >> sensors, the external-pin one and another inside the XADC. > >And they recommend grounding DXP and DXN to improve the ESD rating. IIRC >there's no functional issue caused by leaving them open. > >Regards, >AllanRight, the esd thing makes no sense at all. -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●March 5, 20172017-03-05
On Sat, 04 Mar 2017 09:29:14 -0800, John Larkin wrote:> On 04 Mar 2017 10:52:06 GMT, Allan Herriman <allanherriman@hotmail.com> > wrote: > >>On Fri, 03 Mar 2017 14:00:17 -0800, John Larkin wrote: >> >>> On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin >>> <jjlarkinxyxy@highlandtechnology.com> wrote: >>> >>>>Has anybody used these? >>>> >>>>We have two FPGAs on a board, a Zynq and an Artix7. We want to use the >>>>internal ADCs to read chip temperatures. We have been advised to >>>>ground the temp sense diode pins DXP and DXN "if they are not being >>>>used". >>>> >>>>Unless the XADC has a separate temp sense diode, seems to me that >>>>shorting the external diode pins might kill our ability to acquire >>>>temperature internally. >>>> >>>>I don't see any big downside to not grounding those pins. >>> >>> After too much research, it seems that there are independent temp >>> sensors, the external-pin one and another inside the XADC. >> >>And they recommend grounding DXP and DXN to improve the ESD rating. >>IIRC there's no functional issue caused by leaving them open. >> >>Regards, >>Allan > > Right, the esd thing makes no sense at all.Why do you say it makes no sense? A lot of devices have an improved ESD rating when their more sensitive pins are connected to low impedance sources (as opposed to being open). Of course, if those pins are on a via-less pad under a BGA on a PCB with good planes, the difference may be hard to measure. Regards, Allan
Reply by ●March 9, 20172017-03-09
On 05 Mar 2017 10:41:05 GMT, Allan Herriman <allanherriman@hotmail.com> wrote:>On Sat, 04 Mar 2017 09:29:14 -0800, John Larkin wrote: > >> On 04 Mar 2017 10:52:06 GMT, Allan Herriman <allanherriman@hotmail.com> >> wrote: >> >>>On Fri, 03 Mar 2017 14:00:17 -0800, John Larkin wrote: >>> >>>> On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin >>>> <jjlarkinxyxy@highlandtechnology.com> wrote: >>>> >>>>>Has anybody used these? >>>>> >>>>>We have two FPGAs on a board, a Zynq and an Artix7. We want to use the >>>>>internal ADCs to read chip temperatures. We have been advised to >>>>>ground the temp sense diode pins DXP and DXN "if they are not being >>>>>used". >>>>> >>>>>Unless the XADC has a separate temp sense diode, seems to me that >>>>>shorting the external diode pins might kill our ability to acquire >>>>>temperature internally. >>>>> >>>>>I don't see any big downside to not grounding those pins. >>>> >>>> After too much research, it seems that there are independent temp >>>> sensors, the external-pin one and another inside the XADC. >>> >>>And they recommend grounding DXP and DXN to improve the ESD rating. >>>IIRC there's no functional issue caused by leaving them open. >>> >>>Regards, >>>Allan >> >> Right, the esd thing makes no sense at all. > >Why do you say it makes no sense? A lot of devices have an improved ESD >rating when their more sensitive pins are connected to low impedance >sources (as opposed to being open). > >Of course, if those pins are on a via-less pad under a BGA on a PCB with >good planes, the difference may be hard to measure. > >Regards, >AllanDo you ground every unused pin on every FPGA? -- John Larkin Highland Technology, Inc lunatic fringe electronics
Reply by ●March 9, 20172017-03-09
On Wed, 08 Mar 2017 22:00:21 -0800, John Larkin wrote:> On 05 Mar 2017 10:41:05 GMT, Allan Herriman <allanherriman@hotmail.com> > wrote: > >>On Sat, 04 Mar 2017 09:29:14 -0800, John Larkin wrote: >> >>> On 04 Mar 2017 10:52:06 GMT, Allan Herriman >>> <allanherriman@hotmail.com> >>> wrote: >>> >>>>On Fri, 03 Mar 2017 14:00:17 -0800, John Larkin wrote: >>>> >>>>> On Fri, 03 Mar 2017 13:34:58 -0800, John Larkin >>>>> <jjlarkinxyxy@highlandtechnology.com> wrote: >>>>> >>>>>>Has anybody used these? >>>>>> >>>>>>We have two FPGAs on a board, a Zynq and an Artix7. We want to use >>>>>>the internal ADCs to read chip temperatures. We have been advised to >>>>>>ground the temp sense diode pins DXP and DXN "if they are not being >>>>>>used". >>>>>> >>>>>>Unless the XADC has a separate temp sense diode, seems to me that >>>>>>shorting the external diode pins might kill our ability to acquire >>>>>>temperature internally. >>>>>> >>>>>>I don't see any big downside to not grounding those pins. >>>>> >>>>> After too much research, it seems that there are independent temp >>>>> sensors, the external-pin one and another inside the XADC. >>>> >>>>And they recommend grounding DXP and DXN to improve the ESD rating. >>>>IIRC there's no functional issue caused by leaving them open. >>>> >>>>Regards, >>>>Allan >>> >>> Right, the esd thing makes no sense at all. >> >>Why do you say it makes no sense? A lot of devices have an improved ESD >>rating when their more sensitive pins are connected to low impedance >>sources (as opposed to being open). >> >>Of course, if those pins are on a via-less pad under a BGA on a PCB with >>good planes, the difference may be hard to measure. >> >>Regards, >>Allan > > Do you ground every unused pin on every FPGA?Of course not, but I do ground pins that the manufacturer says to ground. Last century (using QFP packages) I would sometimes ground extra IO pins to help with ground bounce on IO banks that had more O than I. Allan
