Not ideal (waiting for the rain to stop before I dash inside)... but I had my PC create two tones one at 700Hz the other at 1900Hz pretty much as a two tone generator might. I could feed this directly into the rear of the rig through the data socket but in this case I'll just key the mic so the radio picks it up and using my little RF sniffer/sampler/demodulation/detector rig I can view the SSB signal on the oscilloscope. Video below.
This is with my FT897 and I started at 5w and wound it up to 100W just to be sure. All into a dummy load as I am sure live on air two tone test not a wise idea.
I'll do a proper test shortly but even though the results are not conclusive I am reasonably satisfied there isn't a problem with the rig. I suspect it was more a case of someone being off frequency or using a beam in the wrong direction.
The videos of the tests. (The ipod recorded them sideways sorry !) below.
The eagle eyed will realise the rf sniffer is on the left and demod is on the right.
The RF pass through I use is a length of solid earth wire from typical household ring main cable.
The circuit was not hard to build, it is just 4 BNC in a line on a strip of PCB. RF pass through between middle two, and the RF snif, I use a resistor divider to allow be to snif the RF 'live' is on one end and a simple diode demodulator on the other. I can feed either or both into the oscilloscope. The proper one I have is in a nice long dicast box along with an antenna current output too.
My CW signal was not commented on but this is what it looks like from the rig.
I always operate from home especially during contests with a 7 pole low pass filter with about 80dB rejection above 30MHz as neighbours have commented that they can sometimes hear the signals through their FM radios... not for last two years now though.
This link has a lot of interesting info.
The little gizmo I use basically works similar to this one http://www.nu9n.com/images/scope_chain.gif
The demodulator is as I say in the video a simple dual diode IN34 detector.
The RF sampler is a resistor divider chain. It could be a torroid transformer around the RF line, which is simply a bit of wire connecting the RF In to RF OUT.
I have the two signals going into two channels not going in as X and Y as he used for the trapezoidal test.
You adjust the timebase until such times as you can 'see' the signal.