When you have two oscilloscopes hooked up to a third oscilloscope to debug it, so you can make it work and debug your actual DUT, you know you're having fun.
Confirmed, the scope is lying to me about being armed. More data coming shortly.
Ok so, here's the basic setup.
Scope that's misbehaving is my Teledyne LeCroy WaveRunner 8404M-MS, in the right rack (see pic earlier in thread).
It's being triggered by the trigger-out signal of the Teledyne LeCroy SDA 816Zi-A on the left side rack.
Intention is that I arm the WaveRunner trigger, then trigger the SDA, then both should trigger in sequence. But sometimes the WaveRunner isn't triggering and I'm trying to debug.
So I slowed the Ethernet link to the WaveRunner down to 100baseTX mode (so I can easily protocol decode) and used my PicoScope 6824E and Ethernet test fixture to capture the TX/RX Ethernet differential pairs, the trigger pulse going into the WaveRunner, and the "armed" signal on the AUX output port of the WaveRunner.
And what I'm finding is exactly what I was afraid of: The WaveRunner is lying to me about its true arm state.
Here's the timeline:
* 23.801 ms: ask if trigger is armed (should not be)
* 25.201 ms: scope reports trigger stopped
* 27.448 ms: send arm command
* 27.458 ms: ask if trigger is armed (should be)
* 29.102 ms: scope reports trigger armed
* 30.755 ms: trigger event occurs, but is ignored
* 30.781 ms: Armed signal on AUX port goes high
* 31.069 ms: INR poll reports trigger armed