Out of curiosity, are there DSOs with alternate trigger?

[Howardlong]

I've checked two scopes here in the lab - both Agilent DCO1014A and Tektronix MDO3104 have it.

Cheers

Alex

can you show or explain how you did it on the MDO3104? I needed it some weeks ago and didn`t find a way to trigger on signal at Ch1 and an more or less independent signal on Ch3 (which is no problem on my analog Iwatsu)

Michael

I agree, ISTBC, but I don't think it's possible. I suspect it's the confusion of that scope's dual A/B trigger functionality with what an alt trigger is.

[David Hess]

can you show or explain how you did it on the MDO3104? I needed it some weeks ago and didn`t find a way to trigger on signal at Ch1 and an more or less independent signal on Ch3 (which is no problem on my analog Iwatsu)

Michael

I agree, ISTBC, but I don't think it's possible. I suspect it's the confusion of that scope's dual A/B trigger functionality with what an alt trigger is.

The confusion might be well deserved.  On some oscilloscopes with A and B sweeps and triggers, it is possible to use alternating A and B sweep with separate A and B sweep triggers to do alternate triggering.

[guenthert]

I must admit, I don't quite understand the use case.  Isn't this solved with any storage oscilloscope (or sufficiently fast data acquisition system) using single shot, triggered on either signal?  Isn't this (one of) the reasons why a great number of stored samples is desirable?

[JPortici]

can you show or explain how you did it on the MDO3104? I needed it some weeks ago and didn`t find a way to trigger on signal at Ch1 and an more or less independent signal on Ch3 (which is no problem on my analog Iwatsu)

Michael

I agree, ISTBC, but I don't think it's possible. I suspect it's the confusion of that scope's dual A/B trigger functionality with what an alt trigger is.

The confusion might be well deserved.  On some oscilloscopes with A and B sweeps and triggers, it is possible to use alternating A and B sweep with separate A and B sweep triggers to do alternate triggering.

well, alternate is pretty straightforward: alternates between two [...]
betweens two what? there lais the confusion. in my case i think at alt trigger as it's implemented in my tek 7000 mainframe, which is one of the following:
- Trigger from and show only channel a or b, alternatively. trigger is the same, source is different
- Trigger from and show only channel a or b, alternatively. different trigger settings.
- Trigger from the same channel, different trigger settings.
Option 1 is what i'd need.. and i think it's perfectly doable with a DSO but i didn't know many that implemented this function

[David Hess]

I must admit, I don't quite understand the use case.  Isn't this solved with any storage oscilloscope (or sufficiently fast data acquisition system) using single shot, triggered on either signal?  Isn't this (one of) the reasons why a great number of stored samples is desirable?

A digital storage oscilloscope can do that but alternate triggering allows both waveforms to be stably displayed in real time with the trigger in phase with each waveform separately even if the waveforms are asynchronous with each other.

Examples include viewing the switching or ripple of two separate unsynchronized switching regulators at the same time.  Or viewing a switching regulator waveform and a separately clocked logic signal.

well, alternate is pretty straightforward: alternates between two [...]
between two what?  There is the confusion.

Alternate trigger means that when alternate instead of chop is used in dual trace mode and the trigger source is set to vertical, the trigger source follows which trace is currently being displayed.  This was an possibly unintended but useful consequence of early oscilloscope designs where the vertical trigger source selection meant literally that; the trigger source was taken from the vertical signal *after* the vertical channel switch.  Even earlier designs may have *only* had vertical triggering so always operated in alternate trigger mode.  Later designs like the 7000 series emulated this behavior because it was a useful feature.

Alternate sweep means that in dual trace mode, the A sweep and A trigger display one trace and the B sweep and B trigger display the second trace.  The oldest oscilloscope that I know of which can do this is the Tektronix 547.  This mode emulates a true dual beam oscilloscope with separate sweeps and triggers but requires only one beam.

Alternate delayed sweep is something else which displays the same set of signals with two different sweeps.  This is the famous mode where one sweep shows an intensified zone while the second sweep shows the contents of the intensified zone effectively magnifying that part of the slower sweep.  Early delayed sweep oscilloscopes could display one sweep or the other but not both alternately.  Sometimes alternate delayed sweep can be used to emulate alternate sweep.

In my case i think at alt trigger as it's implemented in my tek 7000 mainframe, which is one of the following:
- Trigger from and show only channel a or b, alternatively. trigger is the same, source is different
- Trigger from and show only channel a or b, alternatively. different trigger settings.
- Trigger from the same channel, different trigger settings.
Option 1 is what i'd need.. and i think it's perfectly doable with a DSO but i didn't know many that implemented this function

The Tektronix 7000 series has unusually flexible triggering and alternate triggering in the traditional sense was deliberately included on multiple levels.  The mainframe can alternately trigger between vertical plug-ins and dual input vertical plug-ins can also alternately trigger between inputs.

The 7000 series also support alternate sweep with the same signals displayed with different sweeps and triggers and alternate pairs with different signals displayed with different sweeps and triggers.  The dual beam 7844 (and the older dual beam 555 and dual beam 556?) took these capabilities to an extreme.

As pointed out by other posters, some modern DSOs support alternate triggering.  My old 22xx series DSOs which use analog triggering support alternate triggering simply because the original analog trigger design did also.

[JPortici]

The 7000 series also support alternate sweep with the same signals displayed with different sweeps and triggers and alternate pairs with different signals displayed with different sweeps and triggers.  The dual beam 7844 (and the older dual beam 555 and dual beam 556?) took these capabilities to an extreme.

and i LOVE my 7k.

As pointed out by other posters, some modern DSOs support alternate triggering.

and i wanted to know which one, or what fancy name they used for it... and i may also found out an interesting alternative way to do it with a picoscope, which i have, that i'll check tomorrow

[guenthert]

I must admit, I don't quite understand the use case.  Isn't this solved with any storage oscilloscope (or sufficiently fast data acquisition system) using single shot, triggered on either signal?  Isn't this (one of) the reasons why a great number of stored samples is desirable?

A digital storage oscilloscope can do that but alternate triggering allows both waveforms to be stably displayed in real time with the trigger in phase with each waveform separately even if the waveforms are asynchronous with each other.

Examples include viewing the switching or ripple of two separate unsynchronized switching regulators at the same time.  Or viewing a switching regulator waveform and a separately clocked logic signal.

And you can't do that when you have acquired the data in a single shot?  You can take than all the time you like to see those waveforms 'stable'.  I don't quite see the value of seeing two asynchronous waveforms 'in real time' (and typically much, much faster than the eye could follow).
I get the impression that you (and op) try to apply procedures used to get around limitations of analog oscilloscopes to storage 'scopes.  For the former single shot is only marginally useful (w/o camera virtually pointless), for the latter it is very useful.  New possibilities, new procedures.

I do see the value of triggering on a combination of both signals, as demonstrated earlier in this thread using the picoscope.  That's a feature, I'd like to see in my 'scope.

well, alternate is pretty straightforward: alternates between two [...]
between two what?  There is the confusion.

That wasn't me, your post mixed up the quotes (sadly easy enough to do using this forum's software).

[MrWolf]

And you can't do that when you have acquired the data in a single shot?  You can take than all the time you like to see those waveforms 'stable'.  I don't quite see the value of seeing two asynchronous waveforms 'in real time' (and typically much, much faster than the eye could follow).

Perhaps when doing digital data signal analysis single shot is ok, because carrier stays at same freq... but when looking at PWM, motor drivers etc it is all varying and sometimes pretty fast. I have done a lot of fiddling with custom made motors and pretty typical case is messing with Hall sensor position (outputs square of independently varying period+Duty and frequency), while keeping an eye on coil voltages/currents.

[guenthert]

Perhaps when doing digital data signal analysis single shot is ok, because carrier stays at same freq... but when looking at PWM, motor drivers etc it is all varying and sometimes pretty fast. I have done a lot of fiddling with custom made motors and pretty typical case is messing with Hall sensor position (outputs square of independently varying period+Duty and frequency), while keeping an eye on coil voltages/currents.

But aren't fast changing signals more a reason to evaluate them off-line?  Hmmh, sorry, I don't have done enough of any of this to fully grasp the problems occurring in application.

[JPortici]

this case is square wave at arbitrary frequency in -> square wave at frequency + delta out. both frequency and delta may vary dinamically.
to be honest, i don't actually need alt trigger for this, infact the job was done.
but if i had my tek mainframe at work i would have used alt trigger, stable display of in and stable display of out

[David Hess]

I must admit, I don't quite understand the use case.  Isn't this solved with any storage oscilloscope (or sufficiently fast data acquisition system) using single shot, triggered on either signal?  Isn't this (one of) the reasons why a great number of stored samples is desirable?

A digital storage oscilloscope can do that but alternate triggering allows both waveforms to be stably displayed in real time with the trigger in phase with each waveform separately even if the waveforms are asynchronous with each other.

Examples include viewing the switching or ripple of two separate unsynchronized switching regulators at the same time.  Or viewing a switching regulator waveform and a separately clocked logic signal.

And you can't do that when you have acquired the data in a single shot?  You can take than all the time you like to see those waveforms 'stable'.  I don't quite see the value of seeing two asynchronous waveforms 'in real time' (and typically much, much faster than the eye could follow).

Real time does not have to mean fast.  Alternate triggering allows stable continuous observation of two or more separate and unrelated signals.  This is especially convenient when adjustments or changes in conditions affect both signals.  It would be tiring to have to hit run/stop or single sequence all the time and even then, the untriggered waveform is going to jump around.

I get the impression that you (and op) try to apply procedures used to get around limitations of analog oscilloscopes to storage 'scopes.  For the former single shot is only marginally useful (w/o camera virtually pointless), for the latter it is very useful.  New possibilities, new procedures.

This is a limitation of most digital storage oscilloscopes and few analog storage oscilloscopes.  Think of it as an alternative to requiring two separate oscilloscopes, even digital storage oscilloscopes, in certain applications.  It is a more difficult feature to include in a modern digital storage oscilloscope than an oscilloscope which uses analog triggering where it is practically free and this may explain its rarity in modern instruments.

The only procedure which replaces alternate triggering is the possibility of using a second oscilloscope which while satisfying, may be difficult to justify economically.  Sometimes that is the only way however.  When I decided that I needed a 4 channel DSO, the cheapest good option was to buy two separate 2 channel DSOs.

[JPortici]

tried the logic trigger thingy on the picoscope, it's no substitute for alternate trigger (no stable display of the two waveforms, it's a bit jittery) but the situation has really improved
my setting was
A rising OR B rising

[MrWolf]

A rising OR B rising

Hm, I did try only AND, with two different frequs there will be correlation point always, only question is how often

(no stable display of the two waveforms, it's a bit jittery) but the situation has really improved

Sometimes tweaking the trigger levels and especially hysteresis helps. With PWM sine applying 12bit smoothing really helped. So indeed its not a Tek mainframe but tasks can be still accomplished.
Currently Im playing with derivatives on math channels which greatly amplifies any jitter/noise present. There seems to be small noise related to sampling rate. Probably this is one culprit with logic triggers. On math channel 25MHz low pass gets rid of it.

[KlausF]

Strange. Can anyone figure out a reason why there is no alternate trigger any longer? For me, that´s crazy.   
All the crappy cheap scopes have it.

MrWolf wrote, picoscope has it, but that is actually not true. Picoscopes do NOT have alternate trigger.

[David Hess]

There is less need for it if you can simultaneously sample and a single sweep is acceptable which all DSOs have been able to do for a long time.

A real time alternate display requires more hardware like dual triggers and timebases adding to the expense although these already exist if the DSO supports a delayed timebase with trigger after delay.