Intermittent "No Response" response

Question

I've been working with S1 XBees for some time now, investigating firmware settings and understanding the API mode packet structure. I am a software developer and new to electronics.

I have 9 XBees set up as endpoints and one as a coordinator. I have a program running on the computer hooked up to the coordinator that sends a IO sample request to each XBee endpoint in turn, pausing between each request and pausing for 10 minutes when all 9 requests have been sent.

Unfortunately, for every 9 requests sent, 2-4 of them get a NO_RESPONSE response from the coordinator. It's not always the same endpoint that can't be reached, and an endpoint that didn't respond in one set of 9 requests may respond in the next set of 9.

Does anyone know why this might be? All endpoints are set to never sleep - and sleeping is the only logical reason I can think of that an endpoint would only respond sometimes. I've been chewing through this problem for three months without any real successes.

Answer 1

My first thought is a bad connection or several bad connections somewhere in the network. But you say it is a different endpoint point from time to time. However, I would not rule out a bad connection. If you have this option, I would try and disconnect an endpoint one at a time and see if the problem continues. If it continues, there could be more than one problem connection. Therefore, I would divide and conquer by dividing the network in half or thirds.

I have not worked with xbees at all, but I have alot of experience with networking control equipment. In my experience, 9 times out of 10, it's hardware - especially using serial and coaxial cabling.

Hope this helps.

Comments

Thanks for the input. The traces on the circuit boards seem to be fine and all components are seated well. I've just concluded a four-week test wherein the error rate was 0.15%. The difference is that in this most recent test, the coordinator and all endpoint XBees were all in an area of about a square foot. Previously there was about three feet of space between the coordinator and the endpoints. Changing the power level made no difference.

It appears the XBees have a problem consistently communicating across any distance. I am still trying to determine why 0.15% of the time these units are unable to communicate when they're right next to each other.