Any ideals?

I am currently working on a system that uses licensed 451.8375Mhz Motorola radios with PLCs for a telemetry system. One of the sites experiences random communications failures; the site should TX every 40 seconds. We have verified the radio is outputting full wattage (25watts), no reflective power, cable connections are sealed, ect the basics. It will go into communication failure for up to 6 hours sometimes. In all the documenting on when the communications failures happen there has never been a pattern of always at night, early in the morning, foggy conditions, while raining ect. sometimes it will be during the middle of the day with bright blue clear sky. We are logging what is TX'ed and RX'ed at site; unusual thing is the site that goes into comm fail always hears the other sites talking and always responds when it is it's turn but none of the other sites hear it. Problem seems to be isolated to the Transmit side?

Has anyone ever experience something like this?

 

You checked the power of the errant transmitter but can you check its frequency?
Or its modulation deviation?

 

No we do not have the equipment to do so but I can say that without replacing or reprogramming the radio it will on it's own eventually start communicating again.

 

What are you using to control the radio? What software are you using? What is the RSSI when the radio is not properly transmitting? What is the RSSI when the radio is working properly?

I've seen stuff like this happen when one of your cables has a center pin that is too short for proper mating. It will make/break at random intervals. It can still receive when broken, but the RSSI will be down. However, xmitting is a different story when broken. Try logging the main radio current over a failure and see if it has large fluctuations.

 

The PLC controls when to send data. Software is Motorola STS version 19.50. We do not have RSSI readings from the radios.
We are going to replace the cable and connectors to see if this makes a difference.