Pi TX RX to Moteino RX TX

[odiephd]

I am trying to eliminate the USB cable between the moteino and Pi.   I connected the Pi TX/RX to the Moteino RX/TX (swapped RX-TX, TX-RX), and powered the moteino with 5V and GND from Pi as well.  I am getting very strange behavior.  See attachments.

I used the standard Gateway testing program without the Node transmitting, and all I saw on the Pi serial port was the void setup() routine repeated over and over (See attached).    I remove power from the moteino and it stops.  I turn it back on and it comes up with the standard "Listening at 915 MHz...".   I turn on the node moteino and I see the transmission progression (ABC....etc) on the Pi serial port as expected.  After part or whole sequence (A...Z), I start seeing "Listening at 915 MHz..." again.  I attached the serialoutput so you can see the behavior.

Why is it running the void setup() routine again and again??

Thanks,
Jim

[Felix]

Where does your 5V wire from the PI connect to the Moteino?
Something is resetting your Moteino. What else is attached to it?

[odiephd]

The 5V is from pin2 on Pi to the VIN on Moteino.  There is nothing connected to the Moteino or Pi.  When I run this Moteino connected to my computer through the FTDI connector, it behaves normally. 

I agree something is resetting it.  When I first power the Moteino, it continuously resets.  I unplug the power, wait, and power it back up, it behaves normally for a moment, then starts resetting again. 

I have a more advanced config with sensors connected other Nodes and a Pi gateway working well, but that gateway is using the Pi USB port.   

Seems so straight forward.   Any other things I can try?

[perky]

Can you post pictures showing the connectivity? How are you powering the Pi?

First thoughts:

1) Possible high impedance on the power supply from the Pi to the Moteino. Are you sure the GND and 5V (especially the GND) is connected to the right pins? Have you read the pinout diagrams correctly?

2) Have you completely removed the FTDI connector from the Moteino and not just unplugged it from the other end? The VIN is on the FTDI connector, leaving that in might parasitcally power the FTDI chip in the cable, and the RST pin is also connected to that chip so could in theory reset the Moteino.

Mark.

[odiephd]

The Pi is powered by 5V/2A plug from wall outlet.

1) Yes, I believe so.  The Mote GND is connected to Pin 6 on Pi.  The Mote VIN is connected to Pin 2 (+5V) on Pi. 
2) The FTDI connector is not connected whatsoever. 

See connectivity pic.

[odiephd]

I assembled a new Mote, connected to the Pi and it is not resetting.  Works great.

Not sure why the other Mote is not working properly.  I am out of ideas to try, but I can investigate if you have a hunch.

Thanks
Jim

[perky]

Thanks for the picture. It's a little grainy, but it looks like you might have connected GND on the Pi to pin 5, not pin 6. At least it appears to have a light patch at the base of the header, the TX and RX pins don't have that light patch. Can you check again.

Mark.

[odiephd]

Hey Mark,

Did you see my other post?  I assembled another Mote and everything works fine.  It is something to do with that Moteino.

Sorry about the compressed pic, the 500kb limit made it a bit grainy but it is connected properly.  I'm using only one of the Pi GPIO since it is pins 2,6,8, & 10.

Jim

[RoSchmi]

Perhaps it's worth a try to add a capacitor of about 100 nF near the moteino from GND to Vin.

[perky]

Did you see my other post?  I assembled another Mote and everything works fine.  It is something to do with that Moteino.

I did, but just because it appears to work on another Moteino does not mean that everything is OK! It may be working by the skin of its teeth because it has slightly different thresholds, or takes a slightly different amount of current. If it was using an I/O pin for ground it could be down to the impedance of the buffer which might be different, and that would have the effect of adding impedance to the power supply which in turn could cause voltage drops on the supply when it takes current.   

Sorry about the compressed pic, the 500kb limit made it a bit grainy but it is connected properly.  I'm using only one of the Pi GPIO since it is pins 2,6,8, & 10.

OK, I was just wondering why the photo appeared to show a light patch at the base of the header around pin 6 and it was difficult to see the alignment of the connectors in the header. It might be just that the connector wasn't pressed fully down so I'll take your word for it

Perhaps it's worth a try to add a capacitor of about 100 nF near the moteino from GND to Vin.

Alternatively link pin 1 to pin 6 on the FTDI header. That will do the same thing as adding a 100nF cap to RST.

Mark.

[odiephd]

I added 100 pf cap between VIN and GND, and it solved the problem.   In the recent past, I've done the pin 1/6 wire trick on the FTDI connector but that was just to get the Mote to boot up.  I suppose this is standard issue to add the 100 pf cap.

Thanks for your support! 
Jim