locked VS: VS: [WSJTX] Q65 and non-linear RF power amplifiers #Q65

Reino Talarmo

Is the Q65 ramp amplitude shape and timing documented somewhere?  I couldn't find it in the User Guide or the Quick Start Guide.
Does it vary with sub-mode?

Why is it needed? Reduction of "key-clicks" at the start and end of transmissions?

Hi Steve,


I made some measurements with v2.5.0-rc1 and got results:

FST4 with environmental variable FST4_NOSHAPING=1 starts without a ramp, but ends with about 13 ms Gaussian shape ramp-down

FTS4 with environmental variable FST4_NOSHAPING=0 or no variable starts with about 13 ms ramp-up and ends with a bit less than 13 ms ramp-down

Q65 starts without a ramp-up, but ends with about 2.7 ms ramp-down


I all situations and *modes* Tune starts with a bit less than 10 ms exponential ramp-up and ends with a bit less than 10 ms exponential ramp-down. The same happens, when transmission is stopped with Halt Tx and started in the middle of a Tx period.


I did not spotted any change with sub-modes.


I like the slow ramp-up and ramp-donw as it keeps transmission cleaner. On the other hand having it selectable / optional is also a nice feature with some TX implementations and should be applied also for Tune and perhaps to ramp-down as well. The latter depending on amplifier implementation, the PTT is active hundreds of milliseconds after tone is gone. On the other hand new amplifier designs should take that into account.


A minor conclusion about FST4 with nonshaping is that is does not apply to Tune or to a late start of transmission. So in principle the prevention of the slow starting issue is not fully implemented.


73, Reino OH3mA



To be fair, it's poor transmitter design.   It shouldn't be the job of software to have to be set for this.
A properly designed transmitter, if it is going to be susceptible to low drive level, needs to have level detection circuitry on its input to stop it going into transmit if drive is too low.