Dual Timebase Oscilloscopes

[simon christo]

Hi there! Can somebody please explain the advantages of a Dual Timebase Oscilloscope? I know that 2 Channel Digital scopes have an advantage
over Analog, as the trace on a Analog on Alternate isn't "Real-Time", because the two sweeps are done one after the other.But on a Digital - from what I've been told, doesn't have a slow chop mode (1 Khz or so) , but alternates between the pixels of the dual channels, basically in my view, would be beter for comparing 'High Speed Signals' between channels than an Analog.Am I right? After that we can now talk about Dual Timebases!
Thanks Simon.

[Moshly]

Have a look at this ->

[nctnico]

On a DSO zoom mode is dual timebase model. You also have to realise that on an analog oscilloscope you will be looking at different sweeps (and thus different signals) where on a DSO you'll be looking at the same 'sweep'.

[shakalnokturn]

Analog (CRO) is as real-time as you'll get, it's true that alternate sweeping does not allow you to display two different inputs simultaneously, chopped sweeping either.
It should not be much of a problem as long as you're looking at repetitive waveforms.

On any scope the ability to picture an event correctly is knowing the how to use the trigger.

Digital are less real-time in that there is more delay between what you're probing and what you're viewing.
They are the tool of choice for catching non-repetitive events but bring other limitations due to the sample rate.
Another limitation is the number of ADC's used, with one ADC/channel a digital scope will sample simultaneously on each channel, however some (usually 4 channel models) may share one ADC over two inputs.

[nctnico]

Analog (CRO) is as real-time as you'll get, it's true that alternate sweeping does not allow you to display two different inputs simultaneously, chopped sweeping either.
It should not be much of a problem as long as you're looking at repetitive waveforms.

On any scope the ability to picture an event correctly is knowing the how to use the trigger.

Digital are less real-time in that there is more delay between what you're probing and what you're viewing.
They are the tool of choice for catching non-repetitive events but bring other limitations due to the sample rate.
Another limitation is the number of ADC's used, with one ADC/channel a digital scope will sample simultaneously on each channel, however some (usually 4 channel models) may share one ADC over two inputs.

Let's not turn this into another analog versus digital discussion but your knowledge on DSOs seems to be seriously lacking. First of all the delay on a DSO is unnoticable. The trace update is much faster than your eyes can register. Secondly on a DSO each channel has it's own ADC so all channels are acquired (sweeped) at the same time (*). However on some DSOs you can group the ADCs together and interleave them to achieve higher samplerates / bandwidths on less channels.

* there are some very rare Tektronix DSOs which use alternate acquisition but these where a mistake to put on the market.

[David Hess]

On a DSO zoom mode is dual timebase model. You also have to realise that on an analog oscilloscope you will be looking at different sweeps (and thus different signals) where on a DSO you'll be looking at the same 'sweep'.

That is not correct as there are many DSOs which have both dual timebases *and* zoom mode; these are distinct features despite cheap manufacturers like Rigol presenting the later as the former.  One way to identify true dual timebase DSOs is by the presence of separate trigger settings to support the second timebase; usually each timebase has an associated trigger.

A dual timebase DSO can use different sample rates for each acquisition record.  Zoom displays an existing acquisition with a different time/div and therefor also works on stored waveforms.  Zoom has largely replaced dual timebases because long record lengths allow for an extended operating range of zoom and it is less expensive to implement.

It is not universal that displaying dual timebases requires separate sweeps or acquisitions even on single beam instruments.  Some analog oscilloscopes, like the 4 bay Tektronix 7000 series, support horizontal chop mode operation instead of only alternate mode operation between the timebases.  There may be some DSOs which support the same thing although offhand I do not know of any; there is nothing which precludes it except for cost and complexity and it is a trivial change to support it.

Note that in the above, dual timebase really means dual delayed timebase.  True dual independent timebase oscilloscopes whether analog or digital storage are very rare.

Hi there! Can somebody please explain the advantages of a Dual Timebase Oscilloscope?

I think you really mean dual delayed timebase oscilloscopes so I will only discuss them.  Dual independent timebase oscilloscopes are very rare and essentially come down to having two independent oscilloscopes sharing the same display.

The obvious advantage of a dual delayed timebase is being able to magnify a feature for display however this is the least of it.  For video work, it allows triggering on a video frame, and then displaying only a selected video line.  For measurement work, the delayed timebase can be used to make precision slideback time measurements by aligning two different features on the CRT with the difference in the calibrated delay time control indicating the measured time.  Since this uses the CRT as a null indicator, CRT accuracy is not a factor in the measurement.  An extension of this is the dual delta delayed timebase oscilloscope which can display the delayed timebase with two different delay settings.

DSOs can make measurements directly without slideback measurements.  Zoom supported by large record lengths allows for good magnification capability.  On the other hand, selecting a waveform feature to display still requires a dual delayed timebase for the triggering capability, but this is not a common requirement anymore although I like to use it with switching power supplies.

I know that 2 Channel Digital scopes have an advantage over Analog, as the trace on a Analog on Alternate isn't "Real-Time", because the two sweeps are done one after the other.

Some analog oscilloscopes can capture more than one sweep at a time.  This can be done with dual independent beam instruments, which are effectively two oscilloscopes sharing the same display, or by chopping the horizontal signals between two timebases like with the Tektronix 4 bay 7000 series.

But on a Digital - from what I've been told, doesn't have a slow chop mode (1 Khz or so) , but alternates between the pixels of the dual channels, basically in my view, would be beter for comparing 'High Speed Signals' between channels than an Analog.Am I right?

Some early DSOs also supported user selectable alternate and chop mode but now this is done automatically.  DSOs which interleave their ADCs between channels effectively chop between active channels or groups of channels between samples but as a practical matter, acquisitions can be considered simultaneous.  "Real time" as applied to DSOs used to mean simultaneous acquisitions on multiple channels with independent digitizers.

[David Hess]

First of all the delay on a DSO is unnoticable.

I notice it.  It is especially a problem with control updates.

Secondly on a DSO each channel has it's own ADC so all channels are acquired (sweeped) at the same time (*).

* there are some very rare Tektronix DSOs which use alternate acquisition but these where a mistake to put on the market.

Most DSOs interleave separate ADCs for multiple channels for the reason you give below and because it is cheaper than having separate dedicated digitizers and sample memories.  I am not sure which Tektronix DSOs you are referring to unless you mean some really early ones, and even those supported user selected chopping.  Tektronix specifically released "real time" DSOs which simultaneously sampled; these can be distinguished by having constant sample rates no matter how many channels were active.

However on some DSOs you can group the ADCs together and interleave them to achieve higher samplerates / bandwidths on less channels.

Those statements conflict.  Support for interleaved ADCs usually means separate channels are *not* simultaneously sampled but for most practical purposes, chopping between channels at high sample rates is good enough.

[nctnico]

On a DSO zoom mode is dual timebase model. You also have to realise that on an analog oscilloscope you will be looking at different sweeps (and thus different signals) where on a DSO you'll be looking at the same 'sweep'.

That is not correct as there are many DSOs which have both dual timebases *and* zoom mode; these are distinct features despite cheap manufacturers like Rigol presenting the later as the former.  One way to identify true dual timebase DSOs is by the presence of separate trigger settings to support the second timebase; usually each timebase has an associated trigger.

These must have been very early DSOs then because I've never seen a DSO which had a seperate trigger for zoom and non-zoom mode. Maybe this was in the era that DSOs had a few hundred samples worth of memory. But those days are long gone. Once you have a decent amount of memory it makes much more sense to use zoom mode because you are looking at the same acquisition record.

However on some DSOs you can group the ADCs together and interleave them to achieve higher samplerates / bandwidths on less channels.

Those statements conflict.  Support for interleaved ADCs usually means separate channels are *not* simultaneously sampled but for most practical purposes, chopping between channels at high sample rates is good enough.

No. The channels are not chopped. The signal from each channel is fed into it's own ADC in normal mode. In interleave mode the signal from one channel is fed into two (or more) ADCs where each ADC has a 360/number_of_adcs degree phase shift applied to the clock.  Almost all of the low end oscilloscopes use integrated DSO front-end + ADC chips from Analog Devices nowadays. The higher end ones use dedicated ASICs.

[David Hess]

On a DSO zoom mode is dual timebase model. You also have to realise that on an analog oscilloscope you will be looking at different sweeps (and thus different signals) where on a DSO you'll be looking at the same 'sweep'.

That is not correct as there are many DSOs which have both dual timebases *and* zoom mode; these are distinct features despite cheap manufacturers like Rigol presenting the later as the former.  One way to identify true dual timebase DSOs is by the presence of separate trigger settings to support the second timebase; usually each timebase has an associated trigger.

These must have been very early DSOs then because I've never seen a DSO which had a seperate trigger for zoom and non-zoom mode. Maybe this was in the era that DSOs had a few hundred samples worth of memory. But those days are long gone. Once you have a decent amount of memory it makes much more sense to use zoom mode because you are looking at the same acquisition record.

I never said anything about different triggers for zoom and non-zoom mode.  I said each timebase has a separate trigger so a dual timebase oscilloscope will have separate A and B triggers for timebases A and B whether they are called that now or not. (1)

As for those days being long gone, using a single timebase and zoom, how do you magnify a waveform feature and display it without jitter in real time?  (2) With a dual delayed timebase, you set the delayed trigger to trigger on the waveform feature of interest.  In video applications this displayed a selected line of a frame without jitter for easy viewing and measurement.

(1) There were some weird exception like oscilloscopes with *single* delayed timebases which internally have two triggers for a single timebase; they trigger, delay, and then start the sweep.  The delay in this case takes the place of the main timebase but is considerably simplified and cannot produce a sweep.

(2) High end DSOs can do this now with smart triggers or whatever they want to call it but it does not operate the same way, and how many decades did it take for them to get to this point?  Older DSOs used dual delayed timebases instead because it was cheaper until increasing integration made it economical to do it during decimation.  And budget modern DSOs have neither capability.

[nctnico]

On a DSO zoom mode is dual timebase model. You also have to realise that on an analog oscilloscope you will be looking at different sweeps (and thus different signals) where on a DSO you'll be looking at the same 'sweep'.

That is not correct as there are many DSOs which have both dual timebases *and* zoom mode; these are distinct features despite cheap manufacturers like Rigol presenting the later as the former.  One way to identify true dual timebase DSOs is by the presence of separate trigger settings to support the second timebase; usually each timebase has an associated trigger.

These must have been very early DSOs then because I've never seen a DSO which had a seperate trigger for zoom and non-zoom mode. Maybe this was in the era that DSOs had a few hundred samples worth of memory. But those days are long gone. Once you have a decent amount of memory it makes much more sense to use zoom mode because you are looking at the same acquisition record.

I never said anything about different triggers for zoom and non-zoom mode.  I said each timebase has a separate trigger so a dual timebase oscilloscope will have separate A and B triggers for timebases A and B whether they are called that now or not. (1)

As for those days being long gone, using a single timebase and zoom, how do you magnify a waveform feature and display it without jitter in real time?  (2) With a dual delayed timebase, you set the delayed trigger to trigger on the waveform feature of interest.  In video applications this displayed a selected line of a frame without jitter for easy viewing and measurement.

Jitter in zoom mode isn't much of a problem on a DSO as long as the signal you look at is generated using a stable clock (which is typically the case for video and many other signals). And even if jitter is a problem you can always stop the signal on a DSO and look at it.

[David Hess]

Jitter in zoom mode isn't much of a problem on a DSO as long as the signal you look at is generated using a stable clock (which is typically the case for video and many other signals). And even if jitter is a problem you can always stop the signal on a DSO and look at it.

So the alternative when your signal is not obliging enough to be stable is to discard real time operation?  That is a relief.

First of all the delay on a DSO is unnoticable. The trace update is much faster than your eyes can register.

Maybe you should have said, "The trace update is as fast as you can push the run/stop button."

[nctnico]

Jitter in zoom mode isn't much of a problem on a DSO as long as the signal you look at is generated using a stable clock (which is typically the case for video and many other signals). And even if jitter is a problem you can always stop the signal on a DSO and look at it.

So the alternative when your signal is not obliging enough to be stable is to discard real time operation?  That is a relief.

In the end an oscilloscope's purpose is to show signals which have a relation to eachother so you can see an effect in signal a caused by signal b. What you want is to combine two oscilloscopes in one and show unrelated signals on the same screen. There are some DSOs which offer alternate trigger so you can show unrelated signals stable on the same screen. Zoom mode adds the dual time base here. But in the end it really depens on what you are doing. Alternate triggering is something which is useful to me about once a decade.

[AVGresponding]

Note that in the above, dual timebase really means dual delayed timebase.  True dual independent timebase oscilloscopes whether analog or digital storage are very rare.

Maybe in the US, but here in europe there are tons of old Philips dual timebase scopes on evilbay/craigslist etc.   

[Tom45]

Note that in the above, dual timebase really means dual delayed timebase.  True dual independent timebase oscilloscopes whether analog or digital storage are very rare.

Maybe in the US, but here in europe there are tons of old Philips dual timebase scopes on evilbay/craigslist etc.   

And any of the 4 slot Tek 7000 scopes can be dual time base.

[David Hess]

Jitter in zoom mode isn't much of a problem on a DSO as long as the signal you look at is generated using a stable clock (which is typically the case for video and many other signals). And even if jitter is a problem you can always stop the signal on a DSO and look at it.

So the alternative when your signal is not obliging enough to be stable is to discard real time operation?  That is a relief.

In the end an oscilloscope's purpose is to show signals which have a relation to eachother so you can see an effect in signal a caused by signal b. What you want is to combine two oscilloscopes in one and show unrelated signals on the same screen. There are some DSOs which offer alternate trigger so you can show unrelated signals stable on the same screen. Zoom mode adds the dual time base here. But in the end it really depens on what you are doing. Alternate triggering is something which is useful to me about once a decade.

Alternate triggering and dual independent timebases are not substitutes for dual delayed timebases.  And dual independent timebases do not automatically support delayed operation as anybody with a 4 bay Tektronix 7000 series mainframe and a pair of single timebases can say.

And as my example earlier shows, zoom is not a complete substitute for dual delayed timebases.  All of the dual delayed timebase DSOs I have used also supported zoom and would happily apply it to the delayed acquisition.  But applying it to the delaying acquisition, which is what a single timebase DSO would do, does not produce the same result as using the delayed timebase.

Simon Christo asked about advantages of a dual delayed timebase and the display of part of a waveform without jitter in real time is one of them.  I have often taken advantage of this to display reverse recovery and other events in jittery switching power supply controllers.