Finally got a chance to have a bit of a play with these since they arrived last week.
Anyway, its a fresh install of 1.6.5 on a windows 10 machine.
Added the wifi101 library from the library manager, added the zero from the board manager.
Also have added the due and the ESP8266 stuff from adafruits board file, not that I expect those would affect it.
Example sketches for the wifi101 do nothing when uploaded. Even with the sheild unplugged, anything I put for serial output at the top of setup doesnt happen.
Same examples work fine on a clone uno, eg the network scan or the connect to a WPA2 network sketches - exactly as expected.
Also, the zero will run example sketches for printing analog reads to the serial port etc.
So, anyone else had the 2 new products work together ok?
The Arduino.cc Zero and the WiFi 101 shield work fine for me in California. Things to check:
1. The USB cable to the Zero. Those little connectors can be problematic, and some of the cables are substandard.
2. Make sure you are wired to the programming port and check the device drivers. When you are changing boards, sometimes Windows messes up, restart the computer before you plug in the Zero board. Windows USB support can be a problem.
3. Try some sample program on your Zero that outputs to the serial monitor, and make sure you get output from it. I've gotten in the habit of using CoolTerm, but the Arduino serial monitor should also work.
4. Now try the Zero with the WiFi 101 shield. Try some sample programs: CheckWifi101Firmware Version and ScanNetworks are a good starting point.
5. If these all are fine, try some of the other examples. Most of them require you to set in the SSID and Password of your LAN and your WiFi router will have these stored somewhere in its control panel.
This should all work, please confirm on this forum topic whether or not it does.
Had another chance to play. This is on a different PC since its too smelly in my workshop with paint at the moment to do anything.
Arduino zero on its own with Analoginoutserial example works, spews values to the serial monitor as expected. Am using the programming port. Listed as COM 15 (Arduino/genuino programming port in the list) - device manager has it as Atmel Corp EDGB CMSIS -DAP
Fit WIFI-101 to the zero, and now nothing. The debug light on the zero comes on when I have the serial monitor open, but no data comes back from the zero. Remove the wifi 101 and reconnect the zero and the serial monitor gets data.
I have also tried it with an external DC power supply in case the USB port was sagging powering both, but exactly the same behavior.
Put the wifi-101 ontop of a fake leonardo and it has no effect on it running that same example sketch.
So I thought, Ok, its possibly the serial thats stuffed in this config.
So uploaded blink to the zero. No wifi101 attached - L will flash as expected. Both on USB or DC power.
Stick the wifi101 on top - nothing happens from the LED.
Blink on the leonardo works with the shield ontop. scan networks with the leonardo - also working.
Most of the rest of my stuff is packed away at the moment for renovations so I cant easily grab any other shields to stick on the zero and see what happens.
Edit - Just realized that I was using 1.6.6 on the other computer that I tried it on today, not the 1.6.5 that was using the first time.
Edit again: found a cheap wiznet5100 shield, tried that on both the leonardo and zero I have at hand and worked on both.
Ok, It gets weirder.
I plugged in just the row of power and A0 to A5, leaving the ISP header and digital rows far enough out so they were not connected. And it would start up and run the blink sketch.
So I put the wifi firmware one on the zero, and that ran and reported not detected.
So I moved all the code except the serial into the loop, and ran it, then after it had not detected a few times I pushed the wifi101 all the way onto the zero and it detected and listed the firmware.
So something on the wifi101 is affecting the ability of the zero to boot when connected.
It then worked for many goes with changing between different sketches, network scan, connect to network, NTP (although that is only giving me zeros, it is at least connecting and getting an IP)
It will then work fine till the board loses power. Then I have to repeat the whole process most of the times, but sometimes it will come right back up ok.
It's almost 4am. Im off to sleep now.
I thought to mention that the WiFi 101 works for me on 64-bit Windows 7.1, on a USB 2.0 port with Arduino IDE 1.6.5r5. Now I see that 1.6.6 was just released yesterday I will get around to moving to that.
The Zero is picked up as "Atmel Corp. EDBG CMSIS-DAP (COM10)" by the device manager, and 1.6.5r5 calls it "COM10 Arduino Zero (Programing Port))" so I think the addition of Genuino for 1.6.6 bides well for the Zero outside the USA. Another unit of the Zero comes in as COM12 for me, but also works just fine.
I have not tried WiFi 101 on any AVR Arduino, as it was documented just for the Zero as it came out.
Again, "When you are changing boards, sometimes Windows messes up, restart the computer before you plug in the Zero board. Windows USB support can be a problem." So other than Windows flakey USB support I don't know why the 101 isn't working for you with the Zero. And since you are up to COM 15 USB is maybe the problem.
BTW, was that cheap Wiznet 5100 shield an Ethernet or is there now a WiFi version of it?
The wiznet is an Ethernet shield.
It's not USB drivers since blink will often not run when I'm using DC power as well.
I have the feeling I may have a faulty one so will order more to play with soon.
I have a Zero and WS101 and I'm seeing the same problem - sketches work on the Zero by itelf, but the Zero fails to boot with the WS101 installed.
I'm using IDE 1.6.
Did little debugging and yes it appears the Zero fails to boot with the WS101 installed - nothing comes out the serial port. Looking at the pins I see the the TX pin changes to a high state when the WS101 is installed which indicates that the Zero is doing some form of serial handshake but of coarse the pins on the WS101 are open by default.
Is this fixed by using a down rev of either the Zero library or the IDE? Or can it only be fixed by the funky half plug in method?
Any comments on this? I'm trying to get my daughter's WS101 running for a Science Fair project. Arduino sent a second WS101, but it has the exact same problem.
One minor thing I've noticed is that after doing the half plug, then full plug action so the WS101 works - it will continue to work through a "quick" power cycle of the Zero board. The power cycle is still long enough that the LEDs go out - so unplug Zero from USB port and then plug back in immediately. Only if the Zero board is un-powered for say 10 seconds does everything lock up with both the Zero and WS101 being non-functional. Resets and re-programming don't change whether the WS101 works or doesn't work. Very strange.
Not sure if it matters, but WS101 board reports firmware version: 19.3.0
I don't own a wifi shield, but I know that in some version of the bootloader, if the pin5 is tied low during boot, the Zero stay in the bootloader until you upload some code...
The schematic indicate that digital pin 5 of arduino is used as WINC RESET. The pin is pulled high on the shield, but there is a 10nF cap between D5 and ground on the shield, after a level shifter, and 100nF between IOREF and ground. With the 100k pullup it means a 10ms charging time before the level shifter and 1ms after.
Maybe those cap are not charged enough for the SAMD21 chip to see a HIGH level on pin 5 right after powering the board...
You should try to update the samd core with the 1.6.2 version. Then burn the bootloader using either Atmel ICE or EDBG.
Here is the link to the part of the code that is disabled in the last bootloader that may cause issue with your boards : github link to bootloader (https://github.com/arduino/ArduinoCore-samd/blob/master/bootloaders/zero/main.c#L112)
Burning the new bootloader fixed the problem. WiFi 101 now works without the half plug approach. Thanks.
I have a similar problem with a SAMD21 microcontroller.
I am running the Wifi101 sketch for the firmware version (without connecting the ATWINC1500 wifi module) but it prints "WiFi101 firmware check." and "WiFi101 shield:" and it sticks there .... it doesn't print "not present".
The SAMD21 works fine with a I2C sensor and a SPI Real-time clock.
So, should I buy a debugger to burn a new bootloader?
Any help will be really appreciated!