Rigol DS1052E Cannot Trigger Properly at Sub 1Hz, Why?

[Mechatrommer]

Rigol DS1052E Cannot Trigger Properly at Sub 1Hz frequency. Why?
sometime for a while, sometime forever.

[saturation]

Not sure, but for low frequencies you can use the roll mode, as a temporary fix.

[RayJones]

What exactly is the problem?

If you have a timebase of 1 sec/div it should not be able to re-trigger for 10 seconds - or maybe even 12 given you can expand the display to 12 divisions.
Perhaps there is some weird internal re-trigger holdoff preventing proper operation.

I also find it frustrating with slow sweeps like that, but compared to the old CRO, it is bliss as the waveform stays on the screen!

[Mechatrommer]

demo'ed in the following picture, sine wave signal. the movement is very nice up and down horizontal line at the frequency rate. i set to 5 second persistence display so we can see the decaying trails.
or maybe the dso memory is not long enough to see any meaningfull slope/gradient for trigger to be accepted. how about other dso model? higher end? do they have this kind of problem too? or not?

[RayJones]

Have you tried Normal trigger, not Auto?

[Mechatrommer]

yes, normal trigger seems to work ok. i dont really need it to trigger in auto at this sub hz. there is always workaround, like rollmode as saturation suggested, and this normal trigger. i just wonder why? there should be explananation.

[RayJones]

 CRO is working exactly as it should. Normal trigger only starts the trace upon a valid trigger. Auto trigger will start the trace after some period of time, which is why you see what you do, especially with 200ns timebase.

[scrat]

Even in the Tek TDS3032 I have by hand the triggered modes don't work if the time base is too slow (400ms and above).
I'm not sure about the higher end Tek, but IIRC it does the same.

[RayJones]

Yeah, but the traces presented by Shafri show 20ns/div and 200ns/div - hardly slow timebases.

I'm getting confused by muddled information...

[Wim_L]

It's really quite simple, and common for most scopes. Two modes are of interest here:

- Normal triggering: the scope will trigger on a valid trigger event, as long as it's not currently drawing the screen already, in retrace (for analog scopes) or in the holdoff time.

- Auto triggering: just like normal, but if it has to wait too long for a trigger event, it will spontaneously trigger after some predefined time.

For a well-known signal with a properly set up trigger, normal works nicely. Auto is handy when you don't know what the signal is nice or how to set the trigger, because you get to see at least some flashes of signal to get an idea where you can start to set the trigger.

The waiting time befaore automatic triggering occurs depends on the model. In the manual of my Hameg scope, they specify that a repetition frequency of less than 20 Hz will cause an automatic start of the sweep, so the time would be 50ms for that particular scope. Note that the timebase setting doesn't matter. Only the frequency of trigger events. You could set the timebase at 1 ns/div, but if you're not getting a trigger event within the specified time, the scope will start the sweep.

[RayJones]

Yeah I know all that.

On one hand we have people saying use roll mode for your slow timebase.
But then we have the OP (Shafri) posting images showing a 20ns/div screenshot.

Of course in AUTO mode the damn thing will show a load of crap with a sub hertz signal as it self "triggers" at whatever interval it chooses.

Perhaps Shafri needs to clarify what exactly he is trying to achieve.
Is he trying to capture a glitch on a real slow signal hence the fast timebase?
Is he trying to trigger on a slow edge and the CRO does not like this?
Can he use single shot and manually trigger so he can capture his slow signal and view without overwrite?

[Mechatrommer]

- Auto triggering: just like normal, but if it has to wait too long for a trigger event, it will spontaneously trigger after some predefined time.

maybe this is the 100% mark answer i'm looking for. as mentioned above, i have no need to find event at this low hz, just curious. if i'm serious to find something, i'll use as suggested rollmode and normal or single trigering, pulse etc. but it will be annoying if someone try to trigger the <1Hz signal in auto mode, and only see up and down line without any explanantion.

note: a proper trigger for this low freq (at small ns/div) is only a horizontal line at trigger voltage level. so if trigger at 0V, we should see a stable horizontal line at origin, not up and down.