Measured the P2 CW break-in behaviour and adjusted the CW "rf delay"

accordingly.

diff --git a/new_protocol.c b/new_protocol.c
index dfe61b3221969db99ff0f51e36df5475c5fc6dea..499e6cdb7cee892f4d850a87f820f47a0cb3ddea 100644 (file)
--- a/new_protocol.c
+++ b/new_protocol.c
@@ -1282,7 +1282,13 @@ static void new_protocol_transmit_specific() {
     transmit_specific_buffer[10]=cw_keyer_weight; // cw weight
     transmit_specific_buffer[11]=cw_keyer_hang_time>>8;
     transmit_specific_buffer[12]=cw_keyer_hang_time; // cw hang delay
-    transmit_specific_buffer[13]=0; // rf delay
+    //
+    // Measurements with my Anan-7000 show that RF delays up to 20 msec
+    // tend to chop the first dot, and the "rising" envelope at the
+    // beginning of the dot is also not good. This was tested with the
+    // "straight key" option (external keyer attached to CW input jack).
+    //
+    transmit_specific_buffer[13]=30; // rf delay
 
     transmit_specific_buffer[50]=0;
     if(mic_linein) {