Author Topic: New Rigol DS1054Z oscilloscope  (Read 1434556 times)

0 Members and 2 Guests are viewing this topic.

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 1760
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4250 on: September 11, 2018, 05:19:56 am »
Why are these numbers subtly different from the .csv numbers?

How subtle?

I've converted the differences between the two datasets into raw ADC values and they're all less than one ADC step (the majority are less than half an ADC step).

nb. One ADC step is exactly two pixels on screen so in physical terms the majority of differences between RAW and .csv are less than a pixel.

Seems like some sort of a rounding error to me. Maybe csv is done with fixed-point math.


Conclusion: RAW data is slightly better but using CSV isn't a real problem.

You mean if less than 3 channels are active?
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4251 on: September 11, 2018, 05:53:09 am »
Maybe you've accidentally hit on the reason why Rigol decided to enable the "sinc on/off" button when you get to the extremes, ie. to see how much Gibbs is on screen.  :-+

To find out on what it triggers on you might isolate exact "virtual" part of wfm that could be attributed to mr Gibbs and try to trigger on that.
 

Offline David Hess

  • Super Contributor
  • ***
  • Posts: 9495
  • Country: us
  • DavidH
Re: New Rigol DS1054Z oscilloscope
« Reply #4252 on: September 11, 2018, 06:31:07 am »
Triggering is completely digital however it is probably done in two stages:

1. First the FPGA detects a trigger with no or a minimum of interpolation and transfers the acquisition to the processor without any sin(x)/x interpolation.

2. The processors then resolves the actual horizontal trigger position after interpolation and aligns the acquisition in the display record.  DSOs with analog triggering skip the first part of this because they already know the horizontal trigger position to high resolution but they still need to align the acquisition record with the display record.

DPO style DSOs have to resolve the interpolated trigger to generate a histogram which is then sent to the processor.  The DS1000Z and most DSOs do this in the processor instead of the FPGA or other logic leading to a marked different in acquisition rate.

Didn't the DS1000Z series AC trigger coupling bug reveal something about how the trigger interpolation was done?
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4253 on: September 11, 2018, 05:50:57 pm »
Conclusion: RAW data is slightly better but using CSV isn't a real problem.

You mean if less than 3 channels are active?

I mean that using the CSV format is just as good as grabbing the data over LAN with DSRemote (or whatever).

The real problem here is the assumption that you'll get the "best" data by zooming in, switching to dots mode then saving those 15 points of data.

That simply isn't true.

If you don't understand why that is, watch the video again until you do:


My advice for recording time-aligned pulse inputs on 4 channels simultaneously would be:
Leave the DS1054Z in normal mode (not dots), zoom in, turn sin(x)/x off, then save the 'screen' data. By saving 1200 values you'll avoid the temptation to draw straight lines between sample points in Excel and trying to draw conclusions from that (take note, Wolfie!)

OTOH needing to record 4 time-aligned pulses simultaneously seems an unlikely scenario to me. Much better to use 2 channels (500MSamp./sec., Nyquist at 250Mhz) or 1 channel (Nyquist at 500Mhz, probably perfect reconstruction of everything that gets through the front end).

Why should you turn sin(x)/x off when 4 channels are enabled?
Sampling theory tells us that 250MSamples/sec can only reproduce signals up to 125MHz. If there's anything above frequency that in the input signal then we start to see aliasing, Gibbs phenomenon, etc.

A Rigol DS1104Z doesn't have a brickwall filter at 125MHz, it has a -3dB point at about 130MHz (measured) and there's still visible signal well above 300MHz.

These high frequency components will distort the signal, as shown here:


Rigol has been a bit "naughty". When you push the button to turn off sin(x)/x it does stop doing sin(x)/x, yes, but a different reconstruction filter is used instead. This filter does a reasonable job of removing Gibbs artifacts from what you see on screen, as seen here:


What it does NOT do is show the raw sample points. (Why not? Watch the video again....)
« Last Edit: September 11, 2018, 06:55:24 pm by Fungus »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4254 on: September 11, 2018, 05:59:08 pm »
...but can you trigger on that supposed sin(x)/x distortion (specifically over- or undershoot)? If no then Sinc=OFF shows true analog bw and is not a filter of sorts (like some say). If yes plot thickens :P
Possible you need little more sharper signal or just sine at 100MHz or so.
« Last Edit: September 11, 2018, 06:00:44 pm by MrW0lf »
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4255 on: September 11, 2018, 06:33:05 pm »
...but can you trigger on that supposed sin(x)/x distortion (specifically over- or undershoot)?

No.

If yes plot thickens

Sorry to disappoint you.

Triggering is completely digital...

Maybe they use the other filter (the one that appears when you turn sin(x)/x off).

« Last Edit: September 11, 2018, 06:38:55 pm by Fungus »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4256 on: September 11, 2018, 06:47:25 pm »
...but can you trigger on that supposed sin(x)/x distortion (specifically over- or undershoot)?
No.

Very interesting, but can you trigger on something that over or under (eg not visible) supposed Gibbs-suppressor filter, or it trigges just fine exactly on Sinc=OFF trace?
If it trigger exactly on trace then my original test might have shown approximation of true analog bw because no way one can rise amplitude so much with just regular sin(x)/x, in fact sin(x)/x should not rise amplitude at all (surely <0.5dB):



So the general idea would be finding out true analog amplitude that scope sees using trigger.
We already know it is below Sinc=ON trace, so it cannot be trusted for true analog bw measurement.
Question remaining, is it at Sinc=OFF trace or somewhere between.
« Last Edit: September 11, 2018, 06:56:25 pm by MrW0lf »
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4257 on: September 11, 2018, 07:10:09 pm »
Very interesting, but can you trigger on something that over or under (eg not visible) supposed Gibbs-suppressor filter

I can do much better than that:

I set the thing up so that the peak goes either side of a horizontal grid line when you turn sin(x)/x on/off, eg.:



Now if I move the trigger point up to to that grid line I lose trigger when I turn sin(x)/x off:


« Last Edit: September 11, 2018, 07:13:20 pm by Fungus »
 
The following users thanked this post: MrW0lf

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4258 on: September 11, 2018, 07:21:17 pm »
I can do much better than that:

I set the thing up so that the peak goes either side of a horizontal grid line when you turn sin(x)/x on/off, eg.:

Very good. So it seems truth is somewhere in between. Of course would be correct to check with good signal gen to plot actual response. Currently I would conclude that there is something more going with Sinc=ON than just regular Sinc. Sinc=OFF is filter and it needs to be because they cannot suppress effects of possible ADC post corrector otherwise. Interestingly enough my experimental derivative-based bandwidth boosting post correction looks much like Sinc.

Edit: I had a brain short circuit, what your test did show that it does not trigger in between. :palm: Guess too little coffee in the morning.
« Last Edit: September 11, 2018, 07:31:24 pm by MrW0lf »
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4259 on: September 11, 2018, 07:33:16 pm »
Even better...

If you switch to dots mode and push the trigger upwards towards the peak, it only triggers when one of the physical sample points is above the trigger line.

This limits the number of horizontal positions where it can trigger and you get this:



I can't believe nobody ever posted that before. Am I finding out more about DSOs in ten minutes of playing than you did with two years of drawing graphs in Excel?  :popcorn:
« Last Edit: September 11, 2018, 07:50:30 pm by Fungus »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4260 on: September 11, 2018, 07:40:34 pm »
If you switch to dots mode and push the trigger upwards towards the peak, it only triggers when one of the physical sample points is above the trigger line.

Just to be sure, so this is with Sinc=OFF, yes?
Or does it trigger off virtual points with Sinc=ON as well (I do seem to remember that it moved points with Sinc=ON).
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4261 on: September 11, 2018, 07:48:01 pm »
Just to be sure, so this is with Sinc=OFF, yes?

I can do the same thing in both modes.

(but I need a different trigger level for each one because the peak moves up/down when I switch between them)
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4262 on: September 11, 2018, 08:01:47 pm »
(but I need a different trigger level for each one because the peak moves up/down when I switch between them)

Ok so it triggers off virtual sample points, since analog signal = const. ADC post corrector confirmed IMHO. It's probably deep down in secret core lair and cannot be turned off. To make it look more like regular scope they use Sinc=OFF feature, which is indeed just a filter then to suppress suspicious wfm features. But they overdid it all a little. Both post corrector and suppressor.
Or you have any other explanation?
« Last Edit: September 11, 2018, 08:03:29 pm by MrW0lf »
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4263 on: September 11, 2018, 08:26:50 pm »
To make it look more like regular scope they use Sinc=OFF feature, which is indeed just a filter then to suppress suspicious wfm features.

What's a "regular scope"?

I've already posted examples of other 'scopes that don't really turn off sin(x)/x.

But they overdid it all a little. Both post corrector and suppressor.

Why?
a) It looks like it works to me.
b) You can only see it on in extreme settings where Gibbs might be a real factor. Everywhere else you see sin(x)/x.

The only problem I see is that it frustrates people who make assumptions about what "regular scopes" are then try to characterize based on those assumptions. They might get incorrect results, yes.  :popcorn:

For everybody else? No problems.
« Last Edit: September 11, 2018, 08:49:57 pm by Fungus »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4264 on: September 11, 2018, 08:49:16 pm »
What's a "regular scope"?

One with dots staying put and good for further (custom) DSP.

For everybody else? No problems.

No need to get all defensive. Originally I did come to this forum for sound advice, even described my early day hardships:

Quote
Afterwards started building some experimental analog contraption involving heavily non-linear components. When started testing almost went mad. It did not produce designed signals no matter what I did, non-linearities were all wrong. For days I debugged and tested and calculated until finally I found the culprit deep in the PS software menus... Sin(x)/x on by default. After switching it OFF discovered that contraption was working as designed from the day 1. So maybe one can get away with 2.5 samples per wfm (PS actually had 4 at max freq, all ch in use) for very well known situations... but for heavily experimental stuff only thing that counts is raw data, period. When you start replacing raw data with math fantasy you usually get string theory or something, not maglev trains :)

So I got owned even by classical sin(x)/x. When I bought DS1054Z had no idea that there is no raw data accessible and even sin(x)/x is with a twist. This seemed so weird I got interested how contraption works and did run some tests. Sold it after 2 months so no wonder it took a little to finally find out all the little details w/o having scope at hand :)

But... In my early-early days when I had 25MHz scope all I did was tinker some electrical motors and never needed anything over 100kHz so even that scope was overkill. Rigol would have been entirely ok for that job also. So it all depends what you do, what you want to learn about etc. So it is helpful when there is objective info around not just Like vs Dislike more suitable for knitting club type activity.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4265 on: September 11, 2018, 09:03:01 pm »
What's a "regular scope"?
One with dots staying put and good for further (custom) DSP.

Watch the video again, those dots will be aliased, therefore lies.

Dots are only true when the input signal is bandwidth limited (brick wall) to the Nyquist frequency. That isn't the case here.

(nor on a Siglent SDS1204X-E with all four channels on).
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4266 on: September 11, 2018, 09:17:25 pm »
Watch the video again, those dots will be aliased, therefore lies.

Not really. Look wonders of DSP here:
https://www.eevblog.com/forum/testgear/picoscope-2000/msg1299041/#msg1299041

I'm was able to extract pretty sound normalized amplitude and fuzzy unsigned phase (!) data on 25MHz scope running at 781.3kS/s eg most severely aliased. Plot is from 0 to 100MHz done on some notch filter. It is jolly good fun to learn DSP hands on directly in scope software doing wild stuff.



Of course one could do this on any scope having access to raw data, just on most scopes not directly in GUI. Might try how high new 100MHz scope goes. 400MHz sounds about right.

For reference, here is abstract notch filter, so my phase plot is sound just unsigned.
« Last Edit: September 11, 2018, 09:33:16 pm by MrW0lf »
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3043
  • Country: fi
  • Starting with DLL21
Re: New Rigol DS1054Z oscilloscope
« Reply #4267 on: September 11, 2018, 09:23:49 pm »
Very interesting, but can you trigger on something that over or under (eg not visible) supposed Gibbs-suppressor filter

I can do much better than that:

I set the thing up so that the peak goes either side of a horizontal grid line when you turn sin(x)/x on/off, eg.:



Now if I move the trigger point up to to that grid line I lose trigger when I turn sin(x)/x off:



Why you show all images using  16 waveform average. It is nice to see this  first image with example 1s persistence on and without averaging.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4268 on: September 11, 2018, 09:45:36 pm »
Why you show all images using  16 waveform average.

Because that's the optimal setting? :-//

(and it's what it was set to, yesterday). 

It is nice to see this  first image with example 1s persistence on and without averaging.

If you insist:

(images removed because they mostly show problems in my crappy probing, not the characteristics of the oscilloscope)
« Last Edit: September 12, 2018, 01:52:42 am by Fungus »
 
The following users thanked this post: rf-loop

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3043
  • Country: fi
  • Starting with DLL21
Re: New Rigol DS1054Z oscilloscope
« Reply #4269 on: September 11, 2018, 10:42:04 pm »


Edit: Here's the same thing with sin(x)/x off. Triggering is much tighter:


I'm not sure what conclusions can be drawn with my really crappy probing though.

What you mean triggering is much tighter. I can not see any difference.

Of course it show much much more than average. Average may hide example even very high corners wobbling (if exist) due to aliasing and or Sinc filter produced.
Without average we can see roughly this amount and it can see there but looks like it is around ok range.
Also can see that without Sinc there is not so much wobbling in corners and edges upside and downside from trig level. Also this is normal ok. This is one reason why it is good that oscilloscope can always turn to true sample dots without any interpolation (poormans partially emulated RIS mode what is good for some kind of waveforms where Sinc is not best  and vice versa.) (of course separate trigger engine interpolation is working if or when it exists independent of displayed wfm interpolation) endependent of what example Keysight say or advertise or produce.
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4270 on: September 11, 2018, 11:41:07 pm »
What you mean triggering is much tighter. I can not see any difference.

(image removed)
« Last Edit: September 12, 2018, 01:53:22 am by Fungus »
 

Offline rf-loop

  • Super Contributor
  • ***
  • Posts: 3043
  • Country: fi
  • Starting with DLL21
Re: New Rigol DS1054Z oscilloscope
« Reply #4271 on: September 12, 2018, 01:10:01 am »
What you mean triggering is much tighter. I can not see any difference.



This reason is of course not trigger.

These images are details from previous images where trigger level is around screen center line (in these original full images trigger level is 3.48V). Trigger is there and there can not see differences in trigger jitter in trigger level position. This what you see in image (red circles) is normal "corner wobbling".
« Last Edit: September 12, 2018, 01:58:27 am by rf-loop »
If practice and theory is not equal it tells that used application of theory  is wrong or the theory itself is wrong.
It is much easier to think an apple fall to the ground than to think that the earth and the apple will begin to move toward each other and collide.
 

Online Fungus

  • Super Contributor
  • ***
  • Posts: 9646
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4272 on: September 12, 2018, 01:49:01 am »
This reason is of course not trigger.

OK, a big part of that signal seems to be my horrible probing. I poked the bottom of the Arduino with the probe instead of clipping onto a the Dupont wire and it all got better.

Images have been removed to avoid wrong conclusions. If nobody steps up with a proper pulse generator+BNC then I might try again later with the little probe spring directly onto the AVR chip.
« Last Edit: September 12, 2018, 01:56:00 am by Fungus »
 

Offline MrW0lf

  • Frequent Contributor
  • **
  • Posts: 921
  • Country: ee
    • lab!fyi
Re: New Rigol DS1054Z oscilloscope
« Reply #4273 on: September 12, 2018, 02:17:38 am »
Back in the day I got pretty ok results probing it like this, two silvered wires stuck into header so they would lock with probe for hands free operation:



With ground lead:


Without ground lead, attachment with wires:


Interesting detail, wobbly wfm feature steepens rise!
« Last Edit: September 12, 2018, 02:23:44 am by MrW0lf »
 

Offline metrologist

  • Super Contributor
  • ***
  • Posts: 1760
  • Country: 00
Re: New Rigol DS1054Z oscilloscope
« Reply #4274 on: September 12, 2018, 02:38:46 am »
Conclusion: RAW data is slightly better but using CSV isn't a real problem.

You mean if less than 3 channels are active?

I mean that using the CSV format is just as good as grabbing the data over LAN with DSRemote (or whatever).


Well, I have not used those tools, but you're saying the result is the same? Data is shifted half a screen late and you missed half the pulse? That's really all I am saying. I'm not using dots (which does not matter) or zoom (whatever that really means, it's in full screen default mode). I already understand your other points, but that's not mine. If I use a wider pulse and decrease sweep speed, it seems to work fine, but that was just a glance to see that the pulse was close enough to where it should be in the record set.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf