question grab bag #3 (leftovers from google hangout)

B. the Zynq has CANbus controllers, have you guys done anything with them and are those port pins brought out for a transceiver ?

C. how many of the XADC input pins are brought out, and what sort of analog supply/filtering is in place on the module ? Restated how noisy is the XADC on the Zynq compared to an external serial ADC ?

D. Can you talk about ESD risks with Zynq chip, on Zedboard they put 200-ohm resistors on some of the PMOD port pins from the PL, unsure if that is for EMI purposes or for ESD. Does a typical hobbyist need to be much more cautious with this board compared to run of the mill MCU / RasPi / Beagleboard ?

G. the onboard antenna is is ideally suited for both 2.4 and 5GHz ? any comments on typical range to a decent base station ?

K. Any particular low cost cameras you have up and running with snickerdoodle in the lab ? What interfaces are easiest to work with ?

Hi Rob, I will try to answer these as best as possible.

B) We haven't done anything with the CANbus controllers but you can certainly route the RX and TX pins out any of the FPGA pins for attachment to a transceiver. They are also multiplexed on the microprocessor pins if those pins aren't being used for anything else, however most of our add-on card are using all of the microprocessor pins for Ethernet or USB (well except for breakyBreaky of course... lol)

C) All of the 16 XADC channels on the Zynq come out on the FPGA pins so naturally they are going to be mixed with digital signals which is less than ideal. On the other hand all of the XADC channels are differential unlike your garden variety external serial ADC. Keep in mind though that the XADC input range is about 1V differential since it operates off of the 1.8V rail.

D) If you want the deep details see: http://www.xilinx.com/support/documentation/white_papers/wp433-Mitigating-ESD-EOS.pdf however just as one point of comparison according to table 1-15 in the Xilinx Device Reliability Guide http://www.xilinx.com/support/documentation/user_guides/ug116.pdf the Human Body Model (HBM) passing Voltage is +/-2000V on the FPGA pins. See: https://en.wikipedia.org/wiki/Human-body_model --- that's an identical HBM rating to the STM32F072 which I would consider to be a pretty garden variety MCU: http://www.st.com/st-web-ui/static/active/en/resource/technical/document/datasheet/DM00090510.pdf (see section 6.3.12).

G) The onboard antenna is a dual-band antenna: http://www.pulseelectronics.com/download/2963/w3006pdf/pdf I'm going to refrain from commenting on range because that's incredibly dependent on the RF environment and link budget/encoding, base station etc. and we have an absolutely hideous WiFi environment in San Francisco because everyone is so close together. If you want some deep tech specs here is the datasheet for the dual band WiFi module: http://www.ti.com/lit/ds/symlink/wl1837mod.pdf

K) No low-cost (RPi style) cameras up and running yet. Cameras are a tough one as the good/cool stuff turns out to not be cheap -- e.g global shutter, pixel sync etc. Our starting point is utilizing an HDMI camera because it is easy to see that it works (by plugging it into a TV etc.)
Oh and one more point on (D) I forgot to add is that we don't add series resistors like some of those PMOD ports on Zedboard et. al. do because that would slow down the I/O. I think they are mostly doing that because PMODs are generally pretty low speed interface devices and the series resistors make it possible to continuously short output pins to a power rail w/o damaging them.