Evan, I think you’re right about the spike spacing. It does vary tremendously depending on the transmit frequency. Whatever it’s mixing with, the spikes contract to collapse onto the carrier right about 3.740 MHz. It’s very dramatic. These plots were made at about full power, around 22 watts with the IRF510s in the V3, and attenuated before the TinySA Ultra so that the fundamental was about -27 to -29 dBm. The ones near the zero beat are a 50kHz sweep, with others a wider sweep (up to 5 MHz). Here’s 3.535MHz:

and 3.700MHz:

next, 3.800MHz:

watch the spacing contract at 3.752MHz:

next, 3.737MHz:

here’s higher resolution near the zero beat, 3.738MHz:

and 3.739MHz:

?
here’s a higher resolution sweep, 3.740MHz, pretty much smack dab zero beat:

And widening out again at 3.741MHz:

So what is getting mixed with the carrier to give these spikes? How would the BFO fit into this picture? And are the magnitudes of these spikes low enough to disregard them?
Jack, N6LN