Long timebase on O-Scope = long time before seeing trace?

[SkipMorrow]

I built a monostable 555 circuit with about a 12 sec delay. I wanted to see it operate on my Rigol 1054Z DSO, so I put the timebase at 5 sec/div. I was surprised to see that after hitting the clear button, it would take about 30 seconds until the trace would show up on the screen. In fact, the trace seems to start just when it hits the vertical line at the center of the screen. After that point, you could watch the trace in real time. I can see the channel that I have attached the my pushbutton respond when I press it, and I can see the capacitor charge slowly. So why can't I see it "right away" after pressing the clear button? And even more important, is there a setting I can change so I can see the trace in real time all the time? I have seen that if I decrease the timebase to 2 sec/div, it does show up a lot faster (about 10 seconds), but then I don't have much time left to play with the trigger again before the trace goes off the screen.

Skip

[mikeselectricstuff]

This is a common issue with some digital scopes. Some do update continuously, and some will sometimes update continuously ( e.g. as long as decodes are turned off - this happens with Keysight).

Roll mode is one way to get round it.

[SkipMorrow]

Thanks! Roll mode is PERFECT!

[David Hess]

Good DSOs including most ancient ones support a roll or scan modes or both so at slow sweep speeds, the display is continuously updated.

[rs20]

Roll mode is good, but take the time to understand the "problem" as well: you have your horizontal position set at the center of the screen, because this is the default for digital scopes. This means that the 'scope must wait for half a span's worth of data to buffer before it can even consider responding to a trigger. Once it sees a trigger, it can dump the data in the buffer onto the screen, correctly positioned so that the triggered edge is coincident with the center of the screen.

So the other solution is simple: set your horizontal position to the very left of the screen. This should give you the near-immediate display that you want, but also maintains triggering and a nice way to compare successive waveforms, and is fully compatible with all the features you're used to (measurements, etc).

[David Hess]

So the other solution is simple: set your horizontal position to the very left of the screen. This should give you the near-immediate display that you want, but also maintains triggering and a nice way to compare successive waveforms, and is fully compatible with all the features you're used to (measurements, etc).

This does not always work because some cheap DSOs wait until the acquisition record is full before updating the display.

[rs20]

So the other solution is simple: set your horizontal position to the very left of the screen. This should give you the near-immediate display that you want, but also maintains triggering and a nice way to compare successive waveforms, and is fully compatible with all the features you're used to (measurements, etc).

This does not always work because some cheap DSOs wait until the acquisition record is full before updating the display.

True, but the OP mentioned that his scope does start updating in realtime once the trace passes his chosen horizontal position marker.

[mikeselectricstuff]

Roll mode is good, but take the time to understand the "problem" as well: you have your horizontal position set at the center of the screen, because this is the default for digital scopes. This means that the 'scope must wait for half a span's worth of data to buffer before it can even consider responding to a trigger. Once it sees a trigger, it can dump the data in the buffer onto the screen, correctly positioned so that the triggered edge is coincident with the center of the screen.

So the other solution is simple: set your horizontal position to the very left of the screen. This should give you the near-immediate display that you want, but also maintains triggering and a nice way to compare successive waveforms, and is fully compatible with all the features you're used to (measurements, etc).

I've never understood why only a few scopes (KS & R&S are the only ones I know of) allow you to set the zero-time reference to the left or right instead of the centre. 99% of the time you want to know what happened before or after the trigger, not both sides. The problem with doing by ofsetting is the position moves when you change timebase setting.

[EEVblog]

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.
Roll mode is very intuitive to a beginner IMO.

[T3sl4co1l]

Some classic DSOs did real time trace updates, e.g. HP 546xx.  (I don't recall roll mode on the '00B, but maybe others in the family did.)

I take the opposing issue against Tek's contemporary offerings (TDSxx0): on long timebases, they default to roll mode, which makes it impossible to measure signals visually!  You must change the trigger setting to get a stationary trace; and even then, it only updates after a complete acquisition, not in real time.  (Surely they could've done both, but, alas, these scopes are smack dab in the middle of Tek's terrible interface years.  Whereas the HP 546xx's are a pleasure to use, to this day.)

Tim

[mikeselectricstuff]

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.

Not always - roll mode is typically not triggered, so not useful for capturing slow, triggered events.

[David Hess]

I've never understood why only a few scopes (KS & R&S are the only ones I know of) allow you to set the zero-time reference to the left or right instead of the center. 99% of the time you want to know what happened before or after the trigger, not both sides. The problem with doing by ofsetting is the position moves when you change timebase setting.

Until now, I took it for granted that all DSOs allowed setting the trigger position.  The ones I regularly use either have a dedicated button on the front which instantly changes it or I can manually set it through a menu selection to any position.

I take the opposing issue against Tek's contemporary offerings (TDSxx0): on long timebases, they default to roll mode, which makes it impossible to measure signals visually!  You must change the trigger setting to get a stationary trace; and even then, it only updates after a complete acquisition, not in real time.  (Surely they could've done both, but, alas, these scopes are smack dab in the middle of Tek's terrible interface years.  Whereas the HP 546xx's are a pleasure to use, to this day.)

I only have older Tektronix DSOs and they either have a decided button or the roll/scan configuration takes 2 button presses, Aquisition Mode - Roll/Scan.  It has been so long since I have used a TDS series DSO that I do not remember how they handle it.

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.

Not always - roll mode is typically not triggered, so not useful for capturing slow, triggered events.

Yep, that is what scan mode is for.

On my DSOs when scan mode is used on a triggered signal, the display starts updating exactly when the trigger occurs and then the scan proceeds across the display.  In automatic mode if no trigger is available, then the scan starts at the left.  In single shot mode, the scan will only occur once.

So the only difference between this and waiting for the acquisition record to fill before updating the display is that the display starts updating as soon as the trigger occurs and you do not have to wait to see something.

[free_electron]

set your triggerpoint to the left side of the screen as opposed to the middle.
if the triggerpoint is in the middle and timebase is 1 second per division then it takes 5 seconds for a trace to form. why ? because the left 5 seconds is the pre-trigger buffer that must be filled first. the pretrigger buffer behaves like a fifo. once trigger comes it is latched and the scope fills the other half with 5 more seconds of what happened after trigger.

if you set the triggerpoint to the left of the screen this pretrigger buffer is gone.

[mikeselectricstuff]

set your triggerpoint to the left side of the screen as opposed to the middle.

Can the OP's 1054Z do this ( apart from via the X position control ? )
Inability to set the reference to somewhere other than centre is a  common complaint I've had about pretty much all the lower cost scopes I've used

[David Hess]

set your triggerpoint to the left side of the screen as opposed to the middle.

Can the OP's 1054Z do this ( apart from via the X position control ? )
Inability to set the reference to somewhere other than centre is a  common complaint I've had about pretty much all the lower cost scopes I've used

If it can, I do not see it in the DS1000Z user manual.  The trigger position function is part of an advanced trigger.  The delayed sweep function which has nothing to do with delayed sweep can be set for a negative delay displaying the acquisition record prior to the trigger but that is not the same thing.

I do not see it in the DS2000A series user manual either.

[JPortici]

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.

Not always - roll mode is typically not triggered, so not useful for capturing slow, triggered events.

i like what picoscopes do, with slow timebases it's like a continous roll at left of the trigger point. after trigger it fills the rest of the screen.

other scopes that work like this?

[David Hess]

i like what picoscopes do, with slow timebases it's like a continous roll at left of the trigger point. after trigger it fills the rest of the screen.

other scopes that work like this?

The scan mode I described is similar except that nothing to the left of the trigger point is displayed until the trigger.  Then the rest of the display to the right of the trigger point is filled.

What you describe is an improvement over this.

[vk6zgo]

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.
Roll mode is very intuitive to a beginner IMO.

And was  both counterintuitive & annoying in the extreme, to us greybeards trying to find an equivalent in early DSOs to "free running" (auto) on an analog 'scope.
The latter function is very useful for quickly checking dc supply lines, or for watching for a dc value change.

"Roll" would  show these as a transition at random position on the screen,making them hard to see.

[bson]

Some scopes go into auto-roll mode when at slow timebase settings. If it's slow enough to see changes by eye then you want roll mode.
Roll mode is very intuitive to a beginner IMO.

On the other hand, it would be really nice if it could update the trace live as it goes, but I guess in many cases the record can't both be captured and processed at the same time.  Edit: but then again, if it can do a roll...

LeCroy scopes have a neat feature where they can do a rolling trace of measurements...