Partial waveforms being generated

Another 31022 riddle...<br> <br> In our test application we generate a waveform that has a burst of 10 pulses. The pulses begin at the beginning of the waveform and are followed by 0s. The whole waveform is repeated at some frequency (let's say 60Hz).&#160; We load the waveform into CH1 of the AFG and when we want output we enable the output of CH1.&#160; Sometimes we get the 10 pulses repeated in groups as desired, but sometimes we get fewer than 10 in the first grouping.&#160; (image attached)<br> <br> Before we enable the output it's clear the internal timebase is still counting off the 60Hz. The trigger output is neatly sending out a square wave every 16.67mS.&#160; What I *think* is happening is that when I enable the output, the AFG doesn't wait until the next rising edge of trigger out to start outputting, but rather it simply starts at whatever point it would have been in during that cycle. So if the bursts take 10 of the 16mS and it's 5 mS into the 16mS cycle when I enable, I'll see the last 5 of the 10 pulses in the first group.&#160; Does that make sense? Am I analyzing this correctly?<br> <br> What I need to happen is that when I enable CH1 (press the button physically or via software), I want the AFG to begin sending the waveform in its entirety and not lop off the beginning to keep up with ongoing pulse train from the Trigger out. I would actually prefer that Trigger out simply stop when CH1 is disabled and start a new 16.67 time cycle when it's enabled.&#160; I can't use an external trigger but I believe I have the trigger source to be internal.&#160; I dont' see a setting that says &quot;clock internally but wait to start until I send you a start command&quot;.<br> <br> Is there a command sequence to do what I need to do? Am I properly interpreting the evidence or should it be working correctly and the AFG is just confused?&#160; I've tried it on two units and they behave similarly, but differently. Firmware versions 1.6.1 on each and they both have the SEQ option.<br> <br> Thanks!<br> <br> Howard<br> <br> <br> &#160;

Hi Howard,<br> <br> I've attempted to duplicate, but not seeing any issue yet.<br> I've used the Burst mode and sourcing an ARB defined series of pulses.&#160; Also used Burst mode and Pulse from the Basic mode.<br> For me, on the scope capture, the first burst of pulses after turning the output on has the expected number of pulses.<br> <br> If just a single burst is desired upon a software command, you can config the AFG for External trigger source and send *TRG command to satisfy the external trigger.<br> <br> Looks like we will need a little more information in attempt to duplicate the issue.<br> <br> Andrea

Thanks Andrea.&#160; I'm not looking for a single output. I want continuous output when I enable CH1, but I need it to start at the beginning of the waveform. I guess what I *want* is in addition to an internal and external clock mode, I want one that says &quot;be internal, but stop when CH1 is output is disabled and start a fresh waveform when CH1 is enabled again&quot;.&#160;<br> <br> Do you think it might work if I set the clock to external when I disable CH1, then enable CH1 first before setting the clock to internal?&#160; That's the experiment I was going to try in the lab today.<br> <br> Howard

Hi Howard,<br> <br> I think the problem here is how you're enabling the output.<br> Whenever the instrument is in &quot;Run&quot; mode, the waveform is playing and the clock is advancing. Enabling/disabling channel 1 or 2 does not stop or start the waveform, it simply disconnects the ADC from the output (you can hear the relay clicking).<br> <br> So the solution is to press Run when you want to start or stop the waveform.<br> In SCPI commands this is <code><span class=markedContent id=page220R_mcid0><span dir=ltr role=presentation style="font-family:sans-serif;">SEQControl:<span class="highlight appended">RUN</span></span></span></code> and <code><span class=markedContent id=page220R_mcid16><span dir=ltr role=presentation style="font-family:sans-serif;">SEQControl:STOP.</span></span></code>

Afonso, that makes perfect sense. I'm puzzled about your referring to a 'Run' key. On the panel there are buttons to set the run mode to continuous, burst, etc.&#160; I don't see any way to then effectively start and stop generation properly from the panel but the SCPI commands make a lot of sense. I'll go try that.<br> <br> For other aplications here where we won't have the SEQ option, does that command still work on a basic 31022?<br> <br> Thanks!

Hi Howard,<br> <br> The &quot;Run&quot; button is a physical push button in the &quot;Advanced Mode&quot; cluster on the front panel.<br> <br> For Basic mode (non sequential mode), there's no Run/Stop command, the waveform is always playing (Sweep and Burst are an exception, those require an internal or external sweep to start).

Thanks for the clarification. It's funny that the old 3022 we have is far more likely to start at the beginning of the waveform than the 31022. I think I should be able to make this do what we need now.<br> <br> &#160;

Adding additional information: &#160;If the SEQ option license is enabled, then the start of the waveform can be controlled by a WAIT event.<br> For SCPI wait, when your software sends *TRG, the SCPI wait event is satisfied and the defined waveform will commence.<br> <br> <a href="https://imgur.com/a/Eyq2lQA"><img alt="" src="https://imgur.com/a/Eyq2lQA" style="float:left;"></a><a href="https://imgur.com/a/Eyq2lQA"><img alt="" src="https://imgur.com/a/Eyq2lQA"></a>

Thanks Andrea. I did some tests manually and saw that using the sequence feature did work, and I see that the wait and a manual trigger would also work. The company has decided I have more important fish to fry so this is tabled for now - until it becomes a crisis in a few weeks. You know how that goes.&#160; :(