[Moon-net] IC746 transient output spikes

[Leif Asbrink]

Hi Lance,

> A bit off topic, perhaps, but I know others here also have these rigs.
> 
> I have an IC746, and have been pleased with it so far.  However, I often see RF 
> power output spikes when I switch to XMIT. This is more of a concern now, because 
> I am planning to use this with a solid state amp which cannot tolerate any such 
> excessive power input spikes.  Is there some work-around for this problem that I 
> can implement here?  I look forward to any suggestions you might be able to provide.

I would think this is the common problem of mis-use of ALC that nearly
all modern rigs suffer from. In a properly designed transmitter the ALC
should not be allowed to change the transmitter gain by more than about
2 dB and the purpose is to compensate for "repeaking" There is a book 
from Collins where you can read about the details. The ALC is a wideband
signal that AM modulates the transmitter and modulation has to be kept 
very low to avoid excessive wideband splatter.

The compensation for ageing, different gain on different bands, component
tolerances, temperatures etc must be made with a narrowband signal because
this modulation is large. The design error that so many transmitters suffer 
from is that they use a fast attack, slow release detector to generate this
signal (the TGC, transmitter gain control.) In the short time between tx on
and full TGC voltage, the transmitter runs at maximum gain with its saturated
power output. Awful to neighbouring amateurs and deadly to semiconductor 
power amplifiers. (The proper design would be to start with a TGC woltage
that gives a gain well below what is needed and then allow it to ramp up
at a modest speed. Straight forward to implement in todays transmitters having 
computer control. The problem is that manufacturers do not feel any pressure
from their customers to do these trivial software changes.) 

The recommended cure is to apply a DC voltage to the ALC input to bring the 
gain down to a level below where any ALC or TGC action can happen. You can then
manually adjust the transmitter gain to fit the power amplifier. There will
no longer be any ALC so you will loose about 2 dB on the average power in SSB 
mode. In CW (or JT65) there will be no loss in average power. If you want
to recover the 2 dB in SSB mode you can use an RF speech processor (or a digital
one in Hilbert space) between the microphone and the transmitter.

As an alternative you may mistune the tx amplifiers or load them with 
resistors to make the gain between the ssb filter and the last stage 
low enough to never produce any ALC/TGC signal.

73

Leif / SM5BSZ