Triggering on AM modulated signal

[aneevuser]

I have a DS1054Z and an FY6800 sig. gen.

I'm trying and failing to get the scope to trigger reliably on an AM modulated sine wave, in such a way as to let me see the modulation envelope. It seems that the problem is that neither an edge nor a slope trigger has fine enough resolution to distinguish between an edge that occurs when the envelope is large and one when it is small.

Occasionally I've been able to get a more-or-less stable display, by playing around with various combinations of carrier and modulation frequencies, but this is definitely more luck than judgement (I think it may be most favourable when the carrier freq. is some nice multiple of the modulation freq. or something like that). The most useful approach so far has been to simply pause the scope and look at what's in the memory, but I'd really prefer a real time display.

So, two questions:

1) Does anyone have any tips as to how to do this, in particular with a DS1054Z?

2) Is this a problem specific to digital scopes? I vaguely seem to recall looking at such signals without problem on analogue scopes but that was many years ago so my memory may be faulty - maybe it was a hassle then too.

[capt bullshot]

As you said, it's a hassle in general. With an analog scope, one tries to set the trigger to a as high as possible level so the scope triggers on a modulation signal maximum level.

Way more simple, if you have access to the baseband signal, is to trigger on the demodulated signal using another channel or ext. trigger input.

[Ian.M]

.... and if you don't have the baseband signal available its time to build an AM demodulating RF probe.   Use the classic diode charge pump circuit with a BAT54S for the pair of diodes, 22pF input capacitor, around 200pF for the smoothing capacitor (including capacitance of the coax to the scope) and a 100K load resistor.   Keep the connections to the common pin of the BAT54S and the 22pF capacitor as short as possible.  It may be worth connecting the tip to the 22pF capacitor va an individual pin of a turned pin socket so you can insert a piece of stiff wire for low level signals or a resistor to attenuate high level signals and reduce loading on the D.U.T.  The ground lead at the D.U.T. should be very short - its probably best to use a springy sharp solid wire rather than a flexible lead+ clip.   

[aneevuser]

Way more simple, if you have access to the baseband signal, is to trigger on the demodulated signal using another channel or ext. trigger input.

Right. This is blindingly obvious in retrospect, and works perfectly, within the slightly clunky limits of the UI of the FY6800.

Thanks.

For those who are interested:

The FY6800 has two o/p channels, and allows modulation of CH1 by CH2 (or by an external signal). So you can use CH1 as carrier and CH2 as baseband.

However, you can only adjust the modulation index in real-time while seeing the changes on the scope; if you want to change either CH1 or CH2 freq., the device stops generating the modulated signal, but reverts to the control page for CH1 or CH2 - this means that you have to change those frequencies blindly, and go back to the modulation page to see the effect.

[edit: oops! I got this wrong - the modulation page in fact allows you to modify the freq. or amplitude of the CH1 signal - you can't however modify the CH2 signal on that page, AFAICS)]

I assume the fancy Rigol and Siglent generators are more flexible that way, but for the price of the FY6800, I wouldn't want to complain (too much...)

[aneevuser]

.... and if you don't have the baseband signal available its time to build an AM demodulating RF probe. 

Interesting suggestion, which I hadn't considered at all.

In fact, one of my reasons for playing around with modulated signals is to try to overcome my previous negative experiences of RF electronics (in brief: nothing I made ever worked properly, AFAIR), so this is something that I will definitely be trying at some point, regardless of the fact that the "trigger on CH2" technique works fine.

[dzseki]

In the old days oscilloscopes had LF and HF rejection for triggering, which is useful exactly in such case.

[macboy]

The best way to do this is to split the modulating signal and send it two the external trigger (or another channel) of the scope and trigger from that. In fact, even better, if the function generator has a "sync" output, then trigger from that. Then there is no worry over waveform shape or amplitude, you can even get a stable trigger on a complex arbitrary waveform, since the sync signal is a nice clean pulse at the beginning of each period.

Other good ideas are to look at high and low frequency reject on the trigger, and trigger holdoff.  Holdoff was a big thing missing from my previous digital scope, and I really noticed its absence frequently. It is very useful. Holdoff will wait a certain period of time after the last trigger to re-arm the trigger circuit for the next one.

[aneevuser]

In the old days oscilloscopes had LF and HF rejection for triggering, which is useful exactly in such case.

Presumably that is only useful for carrier and baseband signals that differ sufficiently in frequency? The "trigger on CH2" method works fine even when the signal frequencies are very close (as long as they are nice multiples of each other).

[aneevuser]

The best way to do this is to split the modulating signal and send it two the external trigger (or another channel) of the scope and trigger from that.

Yes, this works fine for AM.

In fact, even better, if the function generator has a "sync" output, then trigger from that. Then there is no worry over waveform shape or amplitude, you can even get a stable trigger on a complex arbitrary waveform, since the sync signal is a nice clean pulse at the beginning of each period.

The FY6800 has no sync output AFAIK, so this is a non starter for that sig. gen.

Other good ideas are to look at high and low frequency reject on the trigger, and trigger holdoff.  Holdoff was a big thing missing from my previous digital scope, and I really noticed its absence frequently. It is very useful. Holdoff will wait a certain period of time after the last trigger to re-arm the trigger circuit for the next one.

I don't think that the DS1054Z has h/k frequency reject. In fact, I was unaware until 2 minutes ago that it had a holdoff control - it does, but it is on the 2nd page of the trigger menu. And... playing with holdoff works brilliantly!!! Thanks. You're a star.

In fact, by playing around with the holdoff time, I've been able to trigger reliably not only on AM signals, but also FM, PSK, ASK, FSK signals, which hadn't been the case with the "trigger on CH2" approach.

ISTR that the upper limit of holdoff on analogue scopes was pretty small; the holdoff on the DS1054Z can be set in units of seconds - it goes at least up to 10 s - I think that's way beyond the capability of an analogue device.

[alsetalokin4017]

It's in the Trigger Coupling menu. I haven't tested it to see if it will work in this particular case; I'll leave that to the OP.

(Doesn't the FY6800 have a TTL output on the back...? My MHS5200 does...)

[aneevuser]

It's in the Trigger Coupling menu. I haven't tested it to see if it will work in this particular case; I'll leave that to the OP.

I see. That's another menu that I've yet to investigate. This thing is swarming with hidden features.

(Doesn't the FY6800 have a TTL output on the back...? My MHS5200 does...)

Yes, but how does that help?

It also has sync in/sync out connectors but the manual seems to suggest that they are used to synchronise multiple devices - I have no idea how that works though, or what signals it supplies on the sync out. I'll have to plug it into the scope to see if i can see anything useful there.

[tggzzz]

It's in the Trigger Coupling menu. I haven't tested it to see if it will work in this particular case; I'll leave that to the OP.

I see. That's another menu that I've yet to investigate. This thing is swarming with hidden features.

That's a classic problem with modern scopes. The older analogue scopes had the advantage that all controls were visible on the control panel.

In fact, one of my reasons for playing around with modulated signals is to try to overcome my previous negative experiences of RF electronics (in brief: nothing I made ever worked properly, AFAIR), so this is something that I will definitely be trying at some point, regardless of the fact that the "trigger on CH2" technique works fine.

You might like to keep in mind that scope are often the wrong tool for looking at RF signals. They are insensitive, have lots of non-linearities (think intermodulation products), and work in the time domain whereas many RF operations are better considered in the frequency domain.

Consider the usefulness of:

• spectrum analyser: sensitive, linear, show amplitude vs frequency
• modulation domain analysers: show frequency (or time-interval) vs time - very good for FM, chirps, VCO settling times and similar
• network analysers: show transfer functions and impedances vs frequency

[aneevuser]

That's a classic problem with modern scopes. The older analogue scopes had the advantage that all controls were visible on the control panel.

Right. I think that I may have to break the habit of a lifetime and read the manual.

Consider the usefulness of:

• spectrum analyser: sensitive, linear, show amplitude vs frequency
• modulation domain analysers: show frequency (or time-interval) vs time - very good for FM, chirps, VCO settling times and similar
• network analysers: show transfer functions and impedances vs frequency

All good points and true.

I have indeed considered the usefulness of such devices, and then my wife considered the prices on my behalf. She's so helpful like that!

[RJSV]

Perhaps more to the point of exploring RF concepts and modulation,  I have a project designing a hardware circuit for FFT (audio spectrum detector).   Exploring the demodulation concepts, and simpler detection methods,  I came up with, at last,  an explanation of the software FFT
routines starting with a synchronous detector.  The circuit must align with both frequency and phase, by trying each possible combination.  That is essentially what a FFT routine does,  using classic 'butterfly' shaped diagrams...
   I possibly will post more details.  Yes, I know all this can be performed on a 1 dollar PIC chip, or other.