Trigger settings question (sometimes missing trigger events)

Having problems ? let us know the details here
Post Reply
jmy173
User
User
Posts: 2
Joined: Wed Dec 01, 2021 8:44 pm

Trigger settings question (sometimes missing trigger events)

Post by jmy173 »

I'm new to using this scope (PicoScope 3406D), and seem to be having issues with missing trigger events. I'm capturing serial data using an advanced edge trigger or a simple edge trigger.
TriggerSettings.png
Most of the time, this works great.
GoodCapture.png
However, I occasionally see the scope start capturing in the middle of a data stream, and I'm not sure why.
LateCapture.png
When this occurs, some serial data is lost (not captured in the buffer) at the beginning of the data stream - I'm not understanding why the scope is not actually triggering until a dozen or so events that match the trigger criteria. In the attached image "LateCapture", why aren't the earlier rising edges causing a trigger?

bennog
Advanced User
Advanced User
Posts: 105
Joined: Mon Nov 26, 2012 9:16 am
Location: Netherlands

Re: Trigger settings question (sometimes missing trigger events)

Post by bennog »

Change the pre trigger from 50% to 5% or so.
Use longer timebase so you catch the whole message
Make sure you have at least 5 samples per bet (preferred 10 samples per bit)

Benno

jmy173
User
User
Posts: 2
Joined: Wed Dec 01, 2021 8:44 pm

Re: Trigger settings question (sometimes missing trigger events)

Post by jmy173 »

Thanks - that is helpful.

The bit rate is 250kb/s, so 1/250,000 sec/bit, or 4 microseconds per bit. 10 samples/bit would be a sampling rate of 10 samples/4 microseconds = 2,500,000 samples/s or 2.5 MS (aka 10x 250k). I made it 4 MS just for make sure.

I can increase the timebase, but each data frame is only 600 microseconds (around 128 bits). The picture I uploaded in the first post contains multiple frames. Here is a single frame (I added a serial decoder, which is a nice feature):
One_Frame.png
I can reduce the pre-trigger to a much lower percent, but it still seems like I'm missing trigger events sometimes-the scope triggers halfway through the data frame, which I still don't understand. Even if I make the pre-trigger as close to 0% as possible, if the scope doesn't trigger on the first bit, that data gets lost. I also lose some data occasionally at the end of the buffer (if the buffer ends during data transmission). Not sure I can do much about that, but is there a way to make sure the trigger occurs every time on the "first" signal that meets the trigger criteria?

For example, in this picture there are three full frames, and the end of a previous frame at the beginning. I don't understand why the scope triggered in the middle of a frame, and not before?
3_and_a_half_Frames.png

bennog
Advanced User
Advanced User
Posts: 105
Joined: Mon Nov 26, 2012 9:16 am
Location: Netherlands

Re: Trigger settings question (sometimes missing trigger events)

Post by bennog »

The decoding and the transferring of data takes time.
If there is data on the bus during decoding and transferring to the PC you will start halfway a message.

What I usually do is disable decoding during capturing and enable it after I have all the capturing done.

Benno

Martyn
Site Admin
Site Admin
Posts: 4094
Joined: Fri Jun 10, 2011 8:15 am
Location: St. Neots

Re: Trigger settings question (sometimes missing trigger events)

Post by Martyn »

You may get better success using a pulse width trigger to ensure that you capture at the start of an active packet, not just on a rising edge.
Martyn
Technical Support Manager

Post Reply