TE0720-02 inputs problem (Must B34_VREF be connected for inputs to work?)

[danwwright]

I have my own board and everything is working fine- *except* the only input to the TE0720-02 module on P20.  The input name is "PhotoInterrupt" and is shown in the attached very simple schematic.  This is on a bank being driven at 2.5V and a voltmeter test confirms that either 0V or 2.5V IS being correctly presented to the pin (depending upon whether the attached photo interrupter circuit is open or closed).  However if I route that input to any one of my outputs I see that it's always high.  All the outputs work fine- I can set them high or low and I'll see either 2.5V or 0V respectively.

I have checked the pinout of the TE0720-02 many times and the signal is going to the right place.  And I have verified that there is no damage to any connector pins or traces.

This is no doubt some stupid oversight on my part- but I cannot identify what it could be.  Could someone take a look at the attached simple schematic and xdc file to see if I'm doing this correctly?

The attachment is really a zip archive (despite the txt extension).  Please rename to "trenz_post.zip" and decompress to see the schematic and xdc file.

[danwwright]

The input "PhotoInterrupt" is on port 34 but (as you'll see from the schematic) B34_VREF is not connected.  I still should be able to tell the zynq to use internal vref but adding this line to the xdc:

set_property INTERNAL_VREF 0.90 [get_iobanks 34]

still leaves that input pinned high as far as the synq is concerned- despite the actual voltages being presented to it as confirmed with a volt meter.

[danwwright]

The answer appears to be that NO vref's are not the problem.  I have NONE of the vref inputs connected but if I configure W8 (called "TEST_3" in the attached schematic) as an input in the same way I had done for "PhotoInterrupt" then the RTL logic in the zynq can see whether it's high or low just fine.  TEST_3 is on B13 whereas PhotoInterrupt is on B34 but both are just banks being powered at 2.5V (again just check the attached schematic).  I don't know if the TE0720 has problems on all pins on B34 but please do check that P20 on the zynq is indeed connected to where the TE0720  connector pinout claims it is.

Thanks,
Dan

[Antti Lukats]

Hi Dan,

1) vref are not relevant for the IO standards you use
2) we are sure the pinout is correct

your schematic does not show any issues either

you can of course not have condition when large back-powering happens, eg if you connect P20 to hard 2.5V while the module is powered then input would burn.
Or if you have any external connection directly to ZYNQ IO pin and plug in cable in random order, then this could cause ESD damage, etc



br
Antti Lukats

[danwwright]

Hi Antti-

The photo-interrupter DOES switch a hard 2.5V source and I suppose that it's possible that the 2.5V rail is coming up before the 3.3V rail powering the TE0720.  What you meant was that the danger case was when 2.5V is applied to a zynq input when the zynq is NOT powered- correct?

Anyway I'll add some rework to direct the photo-interrupter to another zynq input (this time through a resistor).

Thanks,
Dan

[Antti Lukats]

Yes

you can not apply voltages to IO when io banks are not powered.

This is very common with most IC (unless they are specially designed for hot-swap)