Replies to This Discussion

Permalink Reply by Vernon Barry on April 21, 2012 at 3:39pm

I made this same mistake. There are two solder jumper spots on the bottom side of the APM 2.0 for the "autoMUX"

What that means is that reading the top silkscreen on the top side of the board, the serial UART output can either be port 0 or port 2. By default, my board did not have solder connecting to port to either "jumper" on the bottom side of the APM 2.0.

You need to connect it to port 0 so on both sets of pads, solder a short between the middle pad and the inner pad towards the middle of the board that has silkscreened UART 0. Again, both the TX and RX side need connected. Basically, the UART 0/2 is not connected to anything so you Xbee only has power by default (that's the way both of my boards shipped). While at it, I also had to solder jumper the PPM<- to the PPM -> pads near the inputs as well.

Alternatively, you can also solder in pins on the long edge of the board to the port UART 0 as well and just use those. That UART 0 connection is next to the ICSP for the mega (the larger 6 pin 2x3). 

• ▶ Reply

Permalink Reply by Dwgsparky on April 21, 2012 at 3:53pm

Hi Vernon

It looks like that may be the problem, I can see that there is no solder link fitted to either of the UART0 or UART 2

Did I miss this in the instructions ? I must have read them a million times while I was waiting for the board?

I will try it tonight. Many thanks.

• ▶ Reply

Permalink Reply by Vernon Barry on April 21, 2012 at 3:59pm

OK, at least that is what I "thought" you were supposed to do. I back to my original workaround of using the UART0 pins near the ICSP pins on the long side of the board, not the row near the ouputs that the directions say to use?

Anyway, I know for a fact using the hardwired UART 0 on the side works, I have two of these setup this way.

 

I just checked again, and the instructions say to use the port near the outputs, but I know for a fact it did not work on either of my brand new boards. http://code.google.com/p/arducopter/wiki/APM2telemetry

I have gone to the trouble of probing the actual pins from the APM, but today i came across the "jumpers" only because the 2nd APM kit was a non-soldered version, thus I was more inclined to poke around. Neither board came to me with any solder jumpers on the "autoMUX". I initially called DIY drones with the first board and explained the problem I too couldn't connect to the UART shown in the directions (This was before I found the unsoldered autoMUX on the bottom). They said UART0 is shared with the USB and that there was some special switching that happens when USB is plugged in, the port 0 likely will not work. SO lets say for bench testing purpses, you cannot power the board via USB, and then attempt to connect via Xbee (something that does work on APM1).

I would like to discover the right setup of jumpers to get it to work like the directions, but like I've said, I have flown with the other uart0 being used just fine.

• ▶ Reply

Permalink Reply by Dwgsparky on April 21, 2012 at 4:05pm

So If I use the pins at the side do I still need to do the solder links ? I prefer not to as my bionic eye is out for repair!

 

• ▶ Reply

Permalink Reply by Vernon Barry on April 21, 2012 at 4:22pm

No, no solder required on the jumpers to use the hardwired uart0.

And sorry I don't have better info, I'm literally testing this and typing. (My workaround has been working for a month)

The use of the hardwired one on the side is proven, just plug it in and it works. Pinouts is the same as the one at the outputs so basically plug and play.

• ▶ Reply

Permalink Reply by Vernon Barry on April 21, 2012 at 4:29pm

I'm trying different combos and about to dig into the circuit schematics and figure out "why" the pins they say to use seem to be connected to nothing from a tx, rx perspective. I get nothing out of them with a logic probe. I only just noticed the jumpers today and trying that path but can't seem to find the info. I'm sure it's in some dark corner of the wiki. Again, we are in the same boat, I thought maybe I had a bad first board (it was one of the very first APM 2.0s) and did not want to send it in for repair and was happy when I discovered the hardwired UART0 on the side and it worked). I actually bought this second APM 2.0 just to see if it was a bad board or something else. Bad news is it's something else, but good news is, my orignal APM isn't damaged!!!!

Since this is a multilayer board, it's really hard to follow the traces. The TX and RX pins (from the UART they say to use) lead to those jumpers on the bottom (at least that's what I think). Not having the jumpers solder seemed a simple explanation why it didn't work, but now having done that, I am still purplexed??????

• ▶ Reply

Permalink Reply by Dwgsparky on April 21, 2012 at 4:40pm

I think we are both in the same boat but you have more knowledge than me, I will check it out and let you know what I find.

My first APM ever

 

• ▶ Reply

Permalink Reply by Dwgsparky on April 21, 2012 at 5:12pm

Hi, I tried UART0 at the side without the links and it still will not work. I will try with links tomorrow.

• ▶ Reply

Permalink Reply by Vernon Barry on April 22, 2012 at 3:57am

The links are not required for using UART0 on the side so follow the basics and try a loopback test of you xbees first with x-ctu or just any plain serial terminal (arduino has one too). To test loopback, simply power the remote xbee normally from any 5 volt source but loopback the TX and RX pins to each other on the Xbee shield. Even a PC jumper will work, they just need shorted together. X-ctu even has the cool "range test" which gives even more info but all we need to check is a basic link.

Manual for range test http://ftp1.digi.com/support/documentation/90001067_a.pdf

This one is a little better. http://creativeelectron.net/blog/2009/09/xbee-making-your-first-app...

The ONLY difference is the the remote Xbee carrier we use has a FTDI style pins rather than DB9 so we can easily see the TX and RX pins which need shorted togther to make loopback.

 

Follow the basics, get the radio link working in loopback at the correct baud rate 57600. Once you have that working, it's pretty much garanteed the mission planner will connect.

• ▶ Reply

Permalink Reply by Dwgsparky on April 22, 2012 at 1:14pm

Hi Vernon

I followed the manual when I installed the X-CTU and it was fine.

I tried the loopback test ..no good so I installed each xbee in turn into the USB adapter, each xbee commumicated OK with X-CTU and I could read all the values and update the firmware etc so both Xbees are working.

I made a USB to indiviual header cable and connected what would normallly be the aircraft xbee adapter to the PC, all the Leds light as before so I unplugged the cable and installed the xbee to the adapter and plugged it back into the PC again. Tried the x-CTU to read the parameters .......nothing. I tried the master slave switch ....nothing....tried reversing the Rx Tx cables ... nothing .

Replaced the xbees in the USB adapter .. all works OK.

Looks like my Xbee adapter with for the plane is not working.

Thanks for your help. we will talk again I'm sure

Dwgsparky

 

 

Permalink Reply by Vernon Barry on April 23, 2012 at 4:05am

I too had trouble with the 3DR Xbee adapter. I also happened to buy my Xbees and this adapter from Sparkfun http://www.sparkfun.com/products/9132

which you make one minor modification to bypass the diode. This got me to do some checking at what ws really going on because the Sparkfun explorer adaper has far less components and then the mod has you short out a diode for even less active parts. The bottom line is that all that is done is the 3.3 volt regulator reduces the 5 volt power and then by removing the diode, the TX and RX pins are directly connected to the DOUT and NIC pins of the xbee module itself.

On the blue 3DR adapter, they have buffer ICs and probably level shifters along with the switch doing all kinds of shifting. Like I said, mine didn't work either and I followed the directions too. So, I fixed it by removing the 2 buffer ICs and took 2 wires and directly connected the digital in and out pins (RX and TX at the APM) and connected them to the socket pins for the Digitl in and out of the Xbee so that wiring wise, it matched the Sparkfun adapter. It now works fine, the TX and RX work fine.

Permalink Reply by Vernon Barry on April 23, 2012 at 4:55am

From a posting at sparkfun:

How exactly is this 5v-3v3 level shifting performed? It maybe because I’m not that versed in electronics, but the schematic and pcb doesn’t seem to do anything in that regard. Atleast not for data-out. For data-in, if the 5v line goes low it pulls the data-in pin behind the ordinary diode also to low. And the xbee is protected from the 5v by that same diode. But how do you get data-out from 3.3 upto 5V?

The response (and why my mod worked)

I’m seconding this. After playing with this board and the 900MHz XBee Modules, you do need to short the diode to get it to work right. I looked at the datasheet for the XBee Pro 900 and it says that the module has 5V tolerant IO inputs. So you don’t need to worry about hurting the module. I’m not sure why the diode is dropping the levels below a logic 1 though. Thanks for the original comment AdamB, would never have thought of this if you hadn’t posted

• ‹ Previous
• 1
• 2
• Next ›
• Page