[Amps] T-R switching

[kg7hf at comcast.net]

Greetings,

I preface this with I am not a EE (I write code for a living) ;)   

With the help of a few books, I have attempted to design and build a circuit, but I am having trouble making it work consistently without destroying various components.  I have at least two problems that I know of. 

http://home.comcast.net/~kg7hf/Project2/TR_schematic.JPG

The first is that either T17, or Z1, fail after some time, 30 minutes to an hour which results in the TR switch staying in the keyed position.  

The second is somewhat of a stranger issue, perhaps they are related.  This circuit is going into a small solid state amplifier circuit 5 - 10 watts in will produce 50 - 100+ watts output.  When the TR is "working", it seems if I excite the amp with 1 to 5 watts, the TR circuit relay RL17 switches on and off extremely rapidly, as if RF were getting into the relay and causing it to switch on/off, this produces almost a buzzer type effect.  As I increase the exciter power above 5 watts, the relay starts holding and behaves normally (until I get an all out failure).  I've tried what I think is obvious, ferrite beads, and 0.1uf cap from the vcc to ground at the relays, still no change in behavior .

 RL17 & 18 are RTE24012F, SPST2 in my schematic is the input from the key circuit of the exciter, in this case, a Kenwood ts680 on 6 meters.  SPST1 is just a delay circuit and not populated right now.  RL18 is populated, but not connected, and it does not buzz, only RL17.  

If anyone has any comments or suggestions on things to try, I've pretty much run out of ideas, the circuit works forever as long as it's not connected to the PA strip.

Thanks, 73s,
Paul kg7hf.