Oscilloscope Zoom Out Quirk

[jemangedeslolos]

It doesn't let me go beyond the screen.

I´m not really sure ( as a former rigol owner), but the "bigger" MSO5000 couldn´t do that too.
Maybe a actual owner could test and report it here.

The MSO7000 can. It would be surprising if MSO5000 can't.

[Gandalf_Sr]

Give me a specific test to try and I'll run it on my MSO5074

[nctnico]

I vaguely recall that this whole subject started (and then got carved off to a new thread by Dave) because a poster said something to the effect - "this scope doesn't work the way I expect it to and therefore nobody should buy one" - and that was (rightfully IMHO) jumped upon as being unreasonable.

Just for clarity: I wrote that for me (personally) it is a hard fail.

[nctnico]

Since there doesn't seem to be many who specifically ask for that feature in a scope it would help if you could list some actual *models* that support it (which I'm sure you know it appears you wouldn't buy a scope which doesn't). Because my gut feeling is (can't say for sure because we don't usually test for that kind of thing, simply because no-one has ever asked for it) you will find that this is not widely supported.

Well, the majority of the major brands seem to bring out scopes which support recording beyond the screen. Since architectural choices are likely to be constant I don't quite follow how you still come to the conclusion it is not widely supported. All the evidence points in the other direction! Sure there will be some models which behave differently so it is wise to check a particular model during the test drive but in general the list with brands is a good starting point. We could go through the effort of listing specific models but that will likely need to start at listing platforms / series.

BTW: I edited my posting and added R&S to the list (marked with yes).

[nctnico]

Give me a specific test to try and I'll run it on my MSO5074

Set the oscilloscope to 100ns / division, set the memory depth to the maximum number, connect the probe to the calibrator output, set the trigger level to trigger on the signal, set the scope to run mode, disconnect the probe (so no more trigger events; do not press stop!), turn the time/div knob to 200ns/ division. If you get more signal then your scope supports recording beyond the screen.

[maginnovision]

I feel like you're saying there is never any advantage ever in your life because siglent scopes can't do it.

Oh FFS, do you actually think I've not used other DSO's ? Really ?

My first DSO had 2.5 Kpts memory/channel so to get the best from it you used it wisely.
Even the other 2 laterTeks I had were similarly pitiful.

What are we actually trying to do here, teach the reader some obscure scope usage method rather than proper well know methods that return universally predictable results from any DSO ?

No, I think you've used other scopes. Your argument seems to be not every scope can do it so people shouldn't do it or talk about it? I'm really confused about people being vehemently against this when it's just another way to use the tool. Plenty of scopes can do things my scopes can't but I'm not telling them they don't know how to use a DSO because they use those available features.

Some people don't see the point in doing it and that's fine but it's still another method to do a job regardless your feelings. It's literally capture and zoom out(or scroll to the point of interest) OR capture and zoom in, then zoom back out again(or scroll to the point of interest). One of these is the shortcut.

[Martin72]

Set the oscilloscope to 100ns / division, set the memory depth to the maximum number, connect the probe to the calibrator output, set the trigger level to trigger on the signal, set the scope to run mode, disconnect the probe (so no more trigger events; do not press stop!), turn the time/div knob to 200ns/ division. If you get more signal then your scope supports recording beyond the screen.

Shouldn´t be the trigger set to "manual" in this case ?

[nctnico]

Set the oscilloscope to 100ns / division, set the memory depth to the maximum number, connect the probe to the calibrator output, set the trigger level to trigger on the signal, set the scope to run mode, disconnect the probe (so no more trigger events; do not press stop!), turn the time/div knob to 200ns/ division. If you get more signal then your scope supports recording beyond the screen.

Shouldn´t be the trigger set to "manual" in this case ?

Good point. I forgot to mention the trigger mode should be set to 'normal'. Not single, not automatic.

[Martin72]

A 1khz Signal and then 100ns ?
Nevertheless, I´ve just tried it out on the sds2104x+ and the signal disappear.
Now it´s your turn, Gandalf_Sr….

[Gandalf_Sr]

Give me a specific test to try and I'll run it on my MSO5074

Set the oscilloscope to 100ns / division, set the memory depth to the maximum number, connect the probe to the calibrator output, set the trigger level to trigger on the signal, set the scope to run mode, disconnect the probe (so no more trigger events; do not press stop!), turn the time/div knob to 200ns/ division. If you get more signal then your scope supports recording beyond the screen.

I did this simple test.  To clarify, the scope has to be in 'normal' mode or it will retrigger itself after the probe is removed from the calibrator output.

Anyway, after the probe was disconnected, I could wind the horizontal axis all the way out to 10 mS/div and still see valid waveform so it looks like the MSO5000 does capture outside the visible screen area.

[Gandalf_Sr]

Actually, I need to clarify.  After the capture is done, turning the horizontal time control to zoom out shows the captured memory area in the little zoom window top center. At some point, 5 mS/div, it clearly runs out of displayable data and the picture doesn't change even though the indication of the x axis time does - which seems like a bug to me as you could end up looking at a display where the x axis time shown isn't correct for the displayed trace.

[2N3055]

Give me a specific test to try and I'll run it on my MSO5074

Set the oscilloscope to 100ns / division, set the memory depth to the maximum number, connect the probe to the calibrator output, set the trigger level to trigger on the signal, set the scope to run mode, disconnect the probe (so no more trigger events; do not press stop!), turn the time/div knob to 200ns/ division. If you get more signal then your scope supports recording beyond the screen.

In that case, if that is criteria (not pressing STOP) Keysight 3000T doesn't capture anything outside screen in normal mode.
Only when set to digitizer mode, in which case it becomes little wierd to setup, and is not really how you want it for normal interactive mode....
Set to 100ns / division, burst from siggen, trigger set to normal, waiting for next trigger in RUN mode, STOP not pressed:

Still waiting for next trigger, timebase expanded to 200ns / division


This shows what I'we been explaining. 3000T is capturing only screen full of data from trigger to trigger. It is only when you press stop that it will reassemble some data before begin of the screen, and it will keep capturing until it runs after memory. What I suspect is that it literally goes from RUN to STOP, by capturing one SINGLE capture on next trigger after you press STOP (Just with half of the buffer because of state of capture engine at the time, the ping-pong buffers).

I presume it behaves like that, because, if press stop now, while it's waiting for next trigger, it doesn't get any additional data:


So Keysight 3000T can't be used that way, unless it is in DIGITIZER mode, that makes using scope weird for interactive use, negating all user friendliness and ease of use Keysight is famous for.
I would presume all current generation Infiniivision scopes will have same behaviour (chipset related).

If anybody needs any other test, let me know.

Regards,

[Gandalf_Sr]

A further test...
1. Set up a decoder for RS232 at 230,400 Baud on Ch1 with correct polarity and threshold level.
2. Press the mode button until (Normal) appears on the screen briefly
3. Press [Acquire] and set memory depth to max (200M in my case)
3. Monitor RS232 signal with probe and then disconnect to see data frozen on screen with decode data shown below trace
4. start zooming out and notice that top center zoom window (if that's what it's called) is mostly white (that white represents captured data) and there's a small black area that moves as you scroll left and right. It gets bigger as you zoom out and it's indicating the area that's shown on the main screen.
5. In this test I was able to see oodles of data before and after the trigger point and it was all decoded as I looked at it on the main screen

That was fun 

[Gandalf_Sr]

..and I can do the same in Auto mode or if I press Stop, or if I do a single trigger.  There is a boat load of data captured and it's either side of the trigger point.

[EDIT] But it only does this if I manually set the memory depth to max; if the memory depth is set to 'Auto' then the behavior is just like the pictures posted by 2N3055

[egonotto]

Hello,

some pictures from RTA4004.

If you want samples outside of the windows you can use a fix memory size.

If you use auto it can be samples outside windows but not in all cases.

Best regards
egonotto

[maginnovision]

Yea, I've noticed(on the rtb) it's not uncommon to have data outside of the window even in auto memory.

[nctnico]

Yea, I've noticed(on the rtb) it's not uncommon to have data outside of the window even in auto memory.

I think R&S is using one of the fixed memory length selections and not an arbitrary length in auto memory mode.

[Gandalf_Sr]

So the rule seems to be that, in Auto setting, the scopes (usually) only capture what fits on the screen.

[egonotto]

Hello,

nctnico wrote:
"I think R&S is using one of the fixed memory length selections and not an arbitrary length in auto memory mode"

But the memory size in auto mode can differ from the selection of fix memory size. (for example 1.26KSa)

Best regards
egonotto

[maginnovision]

Yea, I've noticed(on the rtb) it's not uncommon to have data outside of the window even in auto memory.

I think R&S is using one of the fixed memory length selections and not an arbitrary length in auto memory mode.

It's possible they have steps in auto, I have seen odd numbers like 19.7M though. Also like egonotto most of the smaller memory sizes aren't user selectable. It seems like perhaps the auto selections are optimized for history/segmented memory modes.

[nctnico]

Yea, I've noticed(on the rtb) it's not uncommon to have data outside of the window even in auto memory.

I think R&S is using one of the fixed memory length selections and not an arbitrary length in auto memory mode.

It's possible they have steps in auto, I have seen odd numbers like 19.7M though. Also like egonotto most of the smaller memory sizes aren't user selectable. It seems like perhaps the auto selections are optimized for history/segmented memory modes.

That seems more likely indeed.

[SilverSolder]

I didn't realize this was so contentious?  -  on my old Agilent 54622D megazoom,  I take it completely for granted that you can zoom both in and out (provided the horizontal speed is high enough) and I use it all the time when exploring new waveforms where it is hard to say what is going to be the best timebase - I thought this is just one of the many awesome convenience advantages of a digital scope! 

I no longer have a 54622D but I don't think it worked any different than all the InfiniVision scopes - they only use the max available memory on short timebases if you're either in SINGLE or on the last acquisition after pressing STOP.

It's *not* about just changing the timebase to change the time period displayed on screen.

Agreed, you have to press STOP to catch the last triggered event of a running signal, or do a single shot.  Then, zooming out works.   (It doesn't work with signals triggered in NORMAL mode.)

The timebase has to be <=1ms on this particular scope or there is no extra data outside the screen area.  I guess the reason for that is that at 2ms, the entire memory is filled just by what is displayed on the screen.  (20ms * 200mSamples/sec = 4 Mbytes...  surprisingly, exactly the amount of RAM the scope has for analog acquisition!)

With two channels turned on...  the amount of memory is halved, effectively, so now there is no "unseen data" unless the timebase is <=2ms, instead of <=1ms.

[rsjsouza]

I have used this in the past, and IIRC the old TDS3054 couldn't do it due to the puny memory it had. However I vaguely recall my old DS1102E doing a scrolling/zooming only on slower timebases, but my current DS4014 with its vast 140MS memory does that easily.

[nctnico]

I didn't realize this was so contentious?  -  on my old Agilent 54622D megazoom,  I take it completely for granted that you can zoom both in and out (provided the horizontal speed is high enough) and I use it all the time when exploring new waveforms where it is hard to say what is going to be the best timebase - I thought this is just one of the many awesome convenience advantages of a digital scope! 

As a side note: it could be I have picked up this trick when using the 54622D or the (I think earlier) 54645D at one of my employers.

[Someone]

DS1000z behaves differently if you set it to AUTO or if you manually fix acquisition length. 

I tried both, didn't made a difference. I could not zoom out beyond the screen data.

One more thing to add to confusion.

But if you set to 24 MS length (1ch) it will get 24 MS every time, even if you set timebase to 5 ns/div.  But your retrigger rate will be the same as if you set timebase to 2ms/div (12divs, 24ms full screen).
Detail will be the same too, and it will do both at 1GS/s.  And with DS1000z , you don't even have to go into zoom mode. Timebase knob works even when stopped. So you capture at any timebase from 5ns/div to 2ms/div and get the same memory content and look at it with timebase knob and horizontal pos knob. Or you can use zoom if you like that.

Maybe that's it, I used the zoom function.

I'll try later with just going to STOP and just twiddling the timebase knob.

Update: I tried just changing the timebase, and with the memory set to manual it did indeed show the full 24Mpts. I did note some artifacts which aren't in the signal, though, so I'm not sure if this behavior is intentional

Yep, another of the corners for nctnico's atypical use case, rejection of zoom mode. You went with the obvious/logical use case!

Zoom mode adds one very important control that is missing when letting the scope expand around the visible window: you gain control of where the trigger is located within the full capture depth.