Triggering at lower levels - mushy trace and loss of triggering

[trys]

Hiya all,

I'm new to EEVblog forum and new to a Rigol DS1054Z. It's my first DSO, apart from a DSO Nano that I have.

I'm having difficulties understanding what is normal, and what could possibly be a fault.

Here is the scenario.

At low signal levels, in this example about 20mV pp, frequency 1 KHz I'm not getting triggering on the Rigol. I just get a mush of two different waves flashing about back and forth, with what appears to be lots of noise. In using edge detect triggering, and irrespective of the trigger level I set. I've attached the image below, but it's just a snapshot so doesn't show the image darting left and right.

I've also hooked up my DSO Nano handheld scope to the same signal at the same time, and I get a triggered waveform, and I can see distortion in the wave, and I can read it well. I've attached the image below.

So, please bear in mind my inexperience with this, why is it I can't get a decent plot on the Rigol?

My initial thinking is that the higher bandwidth of the Rigol is showing the noise that the DSO Nano just doesn't see, and as a result it doesn't know to trigger off the main signal?

On a similar issue, I'm getting loss of triggering if (say) I probe a square wave and I reduce the V/div scale without changing the trigger level. I don't see why the trigger is affected by the vertical scale if it not just looking for the trigger voltage set?

I'm thinking that I'm fundamentally not understanding triggering or something. 

Trys

Edited to add: Just found Dave's video on trigger holdoff - I'm thinking it along those lines?? https://www.youtube.com/watch?time_continue=1&v=ta096oBzSac

[_Wim_]

You could try and use the bandwith limit (20MHz) for your input channel. Because the Rigol has much larger bandwidth than the nano, I think it is triggering on the noise instead of the sine wave...

[trys]

Thank you!

I'm four minutes into EEVBlog Dave's Youtube vid on holdoff too now, and I'm impressed at how little I know. I've wizzed up the holdoff to 1ms on the Rigol (I din't even know what it was until a few minutes ago) and now I get a waveform with all the noise.

Does that image look about right?

Trys

[Keysight DanielBogdanoff]

You are almost certainly triggering on the noise of the signal, try adjusting the trigger holdoff or adding in a bandwidth limit like _Wim_ mentioned. You could also try a different acquisition mode.

We had a blog post a while back that shows how to handle this issue:
https://community.keysight.com/community/keysight-blogs/oscilloscopes/blog/2016/09/01/zombie-apocalypse-survival-guide-oscilloscope-edition

[Keysight DanielBogdanoff]

Yes, that signal looks right. If you want to learn more about how to use an oscilloscope you can check out this series I put together a while back.

https://www.youtube.com/playlist?list=PLzHyxysSubUkc5nurngzgkd2ZxJsHdJAb

[trys]

Daniel,

That's a very useful blog post, and I enjoyed the lighthearted zombie-esque slant on it too. I saw the ghosting too - so that explains a lot.

Adding the 20M bandwidth limit didn't take away the effect I was seeing, so I imagine that means that it's triggering on noise frequency lower than that too. It did alter the sweetspot for the trigger holdoff though (I had to lower the holdoff to a bit under 1ms), so it shows it has an effect which is good.

The oscilloscope youtube playlist looks great - thanks for that. I've watched the first one now so far, and I'll watch the rest now. Thank you for that, and the time put into making that.

Trys

[trys]

Daniel,

You've got a very good style of instructing. I found episode 4 got right to the point on the issue I was facing - yes indeed the acquisition mode turned out to be the real biggie in getting an useful plot for me.

I've attached roughly the same signal but with a bit of averaging added. I've kept it lower (just 4) so that I can still see that it's a noisy signal, but still see something useful.

Thanks again.
Trys

https://youtu.be/yBC97UUljKo

[David Hess]

The options are:

1. Set the trigger coupling to high frequency reject.
2. Limit the input bandwidth.
3. Trigger on another channel attached to a signal synchronous with the desired signal.

[trys]

David,

Thank you.

I had tried option (1) to reject high frequency (HFR) and had no effect as it's broadband noise I imagine. I had also tried (2) to limit the input bandwidth to 20MHz (but again, broadband noise I imagine is the culprit again).

In the posts above, I'd found that the trigger holdoff gave me a steady plot, then I set the acquisition mode to 'Average' to smooth out the plot to get something a bit more useable.

Having said that, as you're suggested an external trigger, I'm going to have a go at that too next so that I have an extra weapon in my arsenal of taming the signal. Thanks for that. I suppose the disadvantage of an external trigger generally is that one has to know the frequency of the signal under test in the first place, which may be the thing that is the reason for probing it - or in practice would you sweep the trigger frequency?

I've learnt a lot from those three point - thank you.

Trys

[technogeeky]

Thank you!

I'm four minutes into EEVBlog Dave's Youtube vid on holdoff too now, and I'm impressed at how little I know. I've wizzed up the holdoff to 1ms on the Rigol (I din't even know what it was until a few minutes ago) and now I get a waveform with all the noise.

Does that image look about right?

Trys

I get that feeling almost every day, too.

Also, I bet High Resolution acquisition mode (instead of average mode) will help you here.

[trys]

The Hires mode appears to make it worse for me on this signal. I'm guessing that it's because the higher the resolution, the more noise it can display. Yikes.

Trys

[David Hess]

I had tried option (1) to reject high frequency (HFR) and had no effect as it's broadband noise I imagine. I had also tried (2) to limit the input bandwidth to 20MHz (but again, broadband noise I imagine is the culprit again).

Rigol has had problems in the past implementing the trigger coupling options so it would not surprise me if high frequency reject was just broken.  AC coupling was broken for a long time and as far as I know is still broken on the Rigol DS2000A series.  They have a habit of not fixing things until the magnitude of bad publicity catches their attention.

[trys]

*cough* Thanks for that David 

Edited to add discreet cough.

[trys]

David,

I have good news.

On a different signal I was able to use HFR to help trigger (see shot below, with the HFR setting shown), so the High Frequency Reject (HFR) is not broken on Rigol as you claimed. It's fine.

AC coupling works too.

Great stuff.

So as I said:

I'm new to EEVblog forum and new to a Rigol DS1054Z. It's my first DSO, apart from a DSO Nano that I have.

So far so good.

The solution for the original signal source with broad spectrum noise, as suggested by Daniel of Keysight was to wizz up the holdoff, and all was good.

Regards,
Trys