Products > Test Equipment
Issue with burst mode on Rigol DG1022Z dropping cycles between bursts.
(1/1)
ions:
I am encountering an annoying issue with Rigol DG1022Z Function Generators. There seems to be a bug when operating in burst mode and triggering externally. My goal here is to generate a 60Hz sine that maintains a relative phase shift to the 60Hz signal that comes out of the wall. Unfortunately its not as simple as setting the AWG to output 60Hz with my desired start phase and call it a day, there is small phase drift that builds up over time. My solution is to use a Line Trigger (which converts the 60Hz sinusoid from the wall into a square wave) to externally trigger a burst of a specified the number of cycles N (a cycle is the same as a period) of the sine wave to output. So N = 5 will output a sine wave for 5 periods, then idle at the specified idle level. It is meant to idle until the next trigger event. At which point the next burst of N cycles is produced. See attached image of very small idle period between bursts (used N = ~50, freq=60.00001Hz, slightly higher freq means burst should complete just before next trigger event). (IMG_1559)

The issue I'm seeing is intermittent, where occasionally there are long idle periods (longer than several cycles) between bursts. This means somehow the AWG is missing a trigger event. See attached image with long idle periods (which has N=30cycles, f=60.00001Hz). (IMG_1558)

If anyone has encountered this and figured out why sometimes cycles get dropped between bursts, please let me know what you know. I'm a grad student, and we are thinking might just need to upgrade to a better AWG, but saw this forum and wanted to see if anyone knows. Thanks
 
Grandchuck:
http://www.leapsecond.com/pages/mains/

Phase/frequency deviations 60 Hz line
Hexley:

--- Quote from: ions on January 20, 2024, 12:11:23 am ---If anyone has encountered this and figured out why sometimes cycles get dropped between bursts, please let me know what you know. I'm a grad student, and we are thinking might just need to upgrade to a better AWG, but saw this forum and wanted to see if anyone knows. Thanks

--- End quote ---
How about capturing the line trigger signal on a second trace so you can verify that the trigger did (or did not!) happen before the dropped burst?
ions:
I think this ended up being the issue. After looking closely at the line trigger signal on the scope I could see fluctuations large enough to create a situation where the next trigger event comes just before the end of the burst. My solution is to characterize the size of these fluctuations and increase the frequency of the AWG such that bursts will always (or 99% of the time) finish before the next trigger event. Or I need to stabilize the line trigger signal somehow, but I'm not really sure how to do that.
2N3055:
Really proper way of doing this would be to create 10MHz oscillator PLL locked to mains frequency, and then use that locked 10MHz oscillator as a reference for AWG.

Just saying...
Navigation
Message Index
There was an error while thanking
Thanking...

Go to full version
Powered by SMFPacks Advanced Attachments Uploader Mod