Complex triggering on modern DSO

[TK]

Is there any scope in the market that can do complex triggering like NORMAL trigger on a specific event and then SINGLE trigger on another event?

Example: I am trying to measure the triggering and decoding capability of SPI signals on some DSOs, I am sending 280,000 SPI messages, ranging from 1 to 8 bytes long.  The message with 8 bytes ends with the byte 0x74.  So there is 35,000 messages containing 0x74.  I want to trigger in NORMAL mode while searching for 0x74, but once in a while, at the press of a button, I am sending a single byte packet with value 0x75 and I want the scope to capture it and STOP.

The objective is to verify if the DSO can trigger on byte value 0x75 that occurs at random, while it is at the same time triggering and decoding on byte value 0x74

Is it possible?

[nctnico]

Not that I know off. This kind of trigger complexity is logic analyser territory.

[2N3055]

What nctnico said.

On 3000T you could set trigger for 0x74 and set last bit to don't care, so it would trigger on both 0x74 and 0x75. Then you set mask for 0x74 and set it to stop on violation.

[TK]

What nctnico said.

On 3000T you could set trigger for 0x74 and set last bit to don't care, so it would trigger on both 0x74 and 0x75. Then you set mask for 0x74 and set it to stop on violation.

Excellent suggestion... I am trying right now on the Siglent SDS1104X-E

[TK]

It worked!!

I had to set the mask with enough clearance to avoid the jitter... but it works perfectly 

[0xdeadbeef]

For the record. Some midrange LeCroys feature multi stage triggers. I think the options for each stage are limited. E.g. I'm not sure if you can use protocol triggers there.

[Someone]

For the record. Some midrange LeCroys feature multi stage triggers. I think the options for each stage are limited. E.g. I'm not sure if you can use protocol triggers there.

Across all brands protocol triggers are often a special class and not integrated with multiple trigger systems, same goes for some of the advanced trigger modes which are analysed offline in software. Multiple trigger modes are the basis for things like rise time triggers, setting a window comparator on the timing between two edge triggers with 10% and 90% thresholds. Running in hardware makes the capture of those events much more likely.

[0xdeadbeef]

Yeah, I checked today at work and the serial trigger is a separate menu item that can't be used in complex trigger modes (i.e. neither in cascaded nor qualified trigger modes). I suspected so but wasn't sure anymore.
Anyway, a cascaded four stage trigger is still pretty complex. Not sure if any Tek, Keysight or R&S scope in a somewhat discussible price range has anything that comes close.

[bson]

You can probably do it with a mask test in some limited fashion.  Trigger on the SPI with 0x74 and apply a pass/fail mask to a gated portion of the waveform, that which contains 0x74.  The mask will fail if that shifts around timewise though, or the masked waveform timing changes substantially.  But I agree that debugging bus traffic as opposed to the integrity of signals themselves is best done with a logic analyzer.

[tv84]

But I agree that debugging bus traffic as opposed to the integrity of signals themselves is best done with a logic analyzer.

Also agree. Nonetheless seems to be a nice academic exercise on the architecture of the scope.