EEVblog #683 - Rigol DS1000Z & DS2000 Oscilloscope Jitter Problems

[edavid]

So Rigol have released a final version of the firmware to some people and it's still not fixed?

That is correct.
Version reported is 00.04.02.03.00 for model DS1000Z when I insert the thumb drive.
It does not update successfully. Using 2 different thumb drives proven to have worked when I updated using the SP3 beta.

Since you haven't been able to install the new firmware, how can you say whether it does or doesn't fix the bugs?

[orin]

So Rigol have released a final version of the firmware to some people and it's still not fixed?

That is correct.
Version reported is 00.04.02.03.00 for model DS1000Z when I insert the thumb drive.
It does not update successfully. Using 2 different thumb drives proven to have worked when I updated using the SP3 beta.

It's looking to me that I will have to
1- sniff the SPI data going to the ADF4360-7 PLL chip, to determine the needed parameters so as to
2 - design a PLL loop LP filter using the software simulator tool
3 - stick that in the DSO

I don't believe 00.04.02.03.00 is a later version which would explain your inability to install it.  00.04.02.03.09 was mentioned way back in this thread and one would assume ...09 was a later version than ...00.

[thmjpr]

I've gone back to my waveform reader for these clock analyses, and I've been able to develop some heuristics to identify the raw data bytes instead of manually looking at a dump of hex bytes.  It works for all the files I have, which in the grand scheme of things, isn't many.

Thought I'd post it if it can be of any use to others.  Save it as "rig_read.m".
...

Thanks, thats much faster than dealing with csv files.. After a few edits it works in matlab, but I think I need to work on my FFT code. DS2072.
edit: updated code and added marks FFT code.

[poida_pie]

So Rigol have released a final version of the firmware to some people and it's still not fixed?

That is correct.
Version reported is 00.04.02.03.00 for model DS1000Z when I insert the thumb drive.
It does not update successfully. Using 2 different thumb drives proven to have worked when I updated using the SP3 beta.

It's looking to me that I will have to
1- sniff the SPI data going to the ADF4360-7 PLL chip, to determine the needed parameters so as to
2 - design a PLL loop LP filter using the software simulator tool
3 - stick that in the DSO

I don't believe 00.04.02.03.00 is a later version which would explain your inability to install it.  00.04.02.03.09 was mentioned way back in this thread and one would assume ...09 was a later version than ...00.

So what version was installed in peoples' DSO prior to the beta f/w that broke things?
Do you have a DS1000Z series device? And it does not have the beta f/w installed?
Love to know what version you have.
 

[orin]

So Rigol have released a final version of the firmware to some people and it's still not fixed?

That is correct.
Version reported is 00.04.02.03.00 for model DS1000Z when I insert the thumb drive.
It does not update successfully. Using 2 different thumb drives proven to have worked when I updated using the SP3 beta.

It's looking to me that I will have to
1- sniff the SPI data going to the ADF4360-7 PLL chip, to determine the needed parameters so as to
2 - design a PLL loop LP filter using the software simulator tool
3 - stick that in the DSO

I don't believe 00.04.02.03.00 is a later version which would explain your inability to install it.  00.04.02.03.09 was mentioned way back in this thread and one would assume ...09 was a later version than ...00.

So what version was installed in peoples' DSO prior to the beta f/w that broke things?
Do you have a DS1000Z series device? And it does not have the beta f/w installed?
Love to know what version you have.

Unfortunately, I did install the beta and don't remember the original version.  I got the keyboard problems with the beta and since I'm so close, my scope is on vacation (loan with Chris in Beaverton... hopefully it gets some new firmware/fix soon and comes back.

[Optisat]

So what version was installed in peoples' DSO prior to the beta f/w that broke things?
Do you have a DS1000Z series device? And it does not have the beta f/w installed?
Love to know what version you have.

I received my DS1054Z 2 weeks ago, no beta installed.
Software: 00.04.01.SP2
Board version: 0.1.1
I was advised by my distributor in sweden to wait for the official release.
Got the beta firmware from Rigol germany with a letter stated that 00.04.01 was the latest official release
and that "the final releases will be accompanied by complete release notes".

Unless I see a release note from Rigol with any distributed SW I will not trust it to be official SW.

//Optisat

[leppie]

I received my DS1054Z 2 weeks ago, no beta installed.
Software: 00.04.01.SP2

IIRC, that is the exact version I had.

[RoTTe]

So what version was installed in peoples' DSO prior to the beta f/w that broke things?
Do you have a DS1000Z series device? And it does not have the beta f/w installed?
Love to know what version you have.

I received my DS1054Z 2 weeks ago, no beta installed.
Software: 00.04.01.SP2
Board version: 0.1.1
I was advised by my distributor in sweden to wait for the official release.
Got the beta firmware from Rigol germany with a letter stated that 00.04.01 was the latest official release
and that "the final releases will be accompanied by complete release notes".

Unless I see a release note from Rigol with any distributed SW I will not trust it to be official SW.

//Optisat

Got mine two weeks ago, and my version is -out of the box-: 00.04.02.SP3

[MarkL]

I've gone back to my waveform reader for these clock analyses, and I've been able to develop some heuristics to identify the raw data bytes instead of manually looking at a dump of hex bytes.  It works for all the files I have, which in the grand scheme of things, isn't many.

Thought I'd post it if it can be of any use to others.  Save it as "rig_read.m".
...

Thanks, thats much faster than dealing with csv files.. After a few edits it works in matlab, but I think I need to work on my FFT code. DS2072.

That looks strange.  Here's the FFT routine I've been using (make_fft.m):

Code: [Select]

function [f, out] = make_fft(t, v, window="hanning", linear_out=0)
  #
  # v0.1 - MarkL @ eevblog,  Dec 23, 2014
  #
  # Compute the FFT of a signal given time (t) and voltage (v)
  # arrays.  Each are in row order (N rows x 1 column).
  #
  # Returns two row order arrays of of freqeuncy and dBV.
  #
  # If linear_out==1 then output units are in Vrms.  This is useful
  # for computing the linear average before converting to dB.
  #
  # The default window is hanning, but can be overriden with any
  # other window function supported by the libraries.  If "none"
  # is given, no window is applied.
  #
  # Example usage:
  #
  #   [f, dbv] = make_fft(t, v);
  #   plot(f, dbv)
  #
  #   [f, vrms] = make_fft(t, v, "blackman", 1);
  #   plot(f, vrms)
  #

  n = numel(v);

  if (!strcmp(window, "none"))
    eval(sprintf("w = %s(n);", window), "");

    # The reduction in signal level that occurs when applying the
    # window is reversed by multiplying by the inverse of the mean of
    # the window coefficients.
    #
    scale = 2/n/mean(w);

    if (linear_out)
      out = abs(resize(fft(scale*(w .* v)), n/2, 1))*sqrt(0.5);
    else
      out = 20*log10(abs(resize(fft(scale*(w .* v)), n/2, 1))*sqrt(0.5));
    endif

  else
    # No window.
    #
    scale = 2/n;
    if (linear_out)
      out = abs(resize(fft(scale*v), n/2, 1))*sqrt(0.5);
    else
      out = 20*log10(abs(resize(fft(scale*v), n/2, 1))*sqrt(0.5));
    endif
  endif

  # Compute the frequency array.
  #
  f_inc = 1/(t(numel(t)) - t(1));
  f = ([0 : (n/2)-1] * f_inc)';

endfunction


# emacs
# Local Variables:
# mode: text
# End:

Easy to use:

  [t, v] = rig_read("whatever.wfm");
  [f, dbv] = make_fft(t, v);
  plot(f, dbv)

Do you get the same results with this FFT?

[orin]

RE: Firmware version numbers

There seems to be much confusion.

I just looked at the beta in a hex editor and it is 00.04.02.03.09

So no surprise 00.04.02.03.00 would not install.

[jkrichards]

RE: Firmware version numbers

There seems to be much confusion.

I just looked at the beta in a hex editor and it is 00.04.02.03.09

So no surprise 00.04.02.03.00 would not install.

I think it would be easy to change the 00.04.02.03.00 to 00.04.02.03.09 in the firmware file.  Maybe then it would load for the ones who tried the Beta.

[orin]

RE: Firmware version numbers

There seems to be much confusion.

I just looked at the beta in a hex editor and it is 00.04.02.03.09

So no surprise 00.04.02.03.00 would not install.

I think it would be easy to change the 00.04.02.03.00 to 00.04.02.03.09 in the firmware file.  Maybe then it would load for the ones who tried the Beta.

Very easy, but is there a checksum?  Anyone know?

[poida_pie]

RE: Firmware version numbers

There seems to be much confusion.

I just looked at the beta in a hex editor and it is 00.04.02.03.09

So no surprise 00.04.02.03.00 would not install.

I think it would be easy to change the 00.04.02.03.00 to 00.04.02.03.09 in the firmware file.  Maybe then it would load for the ones who tried the Beta.

Very easy, but is there a checksum?  Anyone know?

I checked the contents of the f/w supplied to me by Rigol on the 22nd Dec and it contains the plain text
f/w version "00.04.02.03.00"
I can confirm the beta contains "00.04.02.03.09"

Let's see if when I change that one byte from an ascii "0" to "9" I can load it.
When inserting the thumbdrive I get "The same software version detected. Update?"
And it does not load successfully.
This compares with the  "An older version detected. Update?" message I get with the unchanged f/w I received two days ago.

[thmjpr]

Easy to use:

  [t, v] = rig_read("whatever.wfm");
  [f, dbv] = make_fft(t, v);
  plot(f, dbv)

Do you get the same results with this FFT?

Ended up using octave with your code and of course the result is good. Still working on matlab to determine what its doing differently for FFT.

DG1022 function generator 50 ohm terminated. DS2072. 100MHz BW limit:

[kerrsmith]

I have now received my new scope and have spent the last day or so getting to know how it works (well the basics anyway - it has a lot of options compared to my old analogue one!).

The details from the system menu are:

Software Version: 00.04.02.SP3
Board Version 0.1.1

I will try to get some data from it to be analysed but this may not be for another day or two.

I have tested it with a signal from my Arduino's crystal and also a signal from my DDS 9850 signal generator board. Currently these are the only two signal sources I have available to test it with.

I will attempt to do a data capture as suggested by @MarkL using the crystal as soon as I get a chance.

[kerrsmith]

I think the AC trigger coupling error is still present as when I have clean wave on the screen in DC coupling mode and then change it to AC coupling it gets a little fuzzy.

If I can figure out how to do a screen shot I will post it up - just need to read this up in the manual.

[jkrichards]

I ordered a DS1054Z last month and have been waiting for news about its delivery and bug fixes (I have been closely following this thread and other relevant ones for ages).

I just received an email saying my scope was due to be posted out today so I called up Rigol to ask about the AC coupled trigger mode bug and the jitter issue as I had not seen any mention of a fix on this thread.

I have been told by a very helpful member of staff that they have just received a batch of scopes that have been flown over directly to the UK and they all have a new firmware on them which fixes these issues.

Looks like you have been told wrong. That doesn't surprise me.  Now we need to see if you have these two issues on your new scope.  If the issues are fixed then we know they made hardware changes.

[jkrichards]

I think the AC trigger coupling error is still present as when I have clean wave on the screen in DC coupling mode and then change it to AC coupling it gets a little fuzzy.

If I can figure out how to do a screen shot I will post it up - just need to read this up in the manual.

You need to check AC triggering NOT AC coupling.  Switch from DC to AC in the triggering section.  You can leave coupling on DC.

[kerrsmith]

I think the AC trigger coupling error is still present as when I have clean wave on the screen in DC coupling mode and then change it to AC coupling it gets a little fuzzy.

If I can figure out how to do a screen shot I will post it up - just need to read this up in the manual.

You need to check AC triggering NOT AC coupling.  Switch from DC to AC in the triggering section.  You can leave coupling on DC.

Yes, this is what I meant I just forgot to write the word triggering - sorry for the confusion.

Screen shots on the way...

[kerrsmith]

Here are my two screen shots - they are at 9MHz:

[jkrichards]

Here are my two screen shots - they are at 9MHz:

You are only changing from DC to AC coupling NOT triggering.  Leave it on DC coupling.  Go to the triggering menu and change it to AC.  I can clearly see in your captures that you are in the coupling menu.

Also your AC coupling screenshot is fuzzy because you don't have the trigger set correctly.

[mamalala]

Very easy, but is there a checksum?  Anyone know?

Well, i can only speculate ... but: There have been tools to mangle the .gel file for other Rigol gear. Maybe that can give a hint as to how any checksum is present and/or calculated? http://gotroot.ca/rigol/

Greetings,

Chris

[kerrsmith]

Here are my two screen shots - they are at 9MHz:

You are only changing from DC to AC coupling NOT triggering.  Leave it on DC coupling.  Go to the triggering menu and change it to AC.  I can clearly see in your captures that you are in the coupling menu.

Also your AC coupling screenshot is fuzzy because you don't have the trigger set correctly.

This is the trigger menu - I have attached three screen shots:

The first is the main channel input coupling menu
The second is the trigger menu with DC trigger coupling
The third is the trigger menu with AC trigger coupling

These two menus are different, you need to press the trigger menu button to get to the second and third images, the first comes from the channel selection menu button.

Here is a link to a video of Dave showing this:

[jkrichards]

Here are my two screen shots - they are at 9MHz:

You are only changing from DC to AC coupling NOT triggering.  Leave it on DC coupling.  Go to the triggering menu and change it to AC.  I can clearly see in your captures that you are in the coupling menu.

Also your AC coupling screenshot is fuzzy because you don't have the trigger set correctly.

This is the trigger menu - I have attached three screen shots:

The first is the main channel input coupling menu
The second is the trigger menu with DC trigger coupling
The third is the trigger menu with AC trigger coupling

These two menus are different, you need to press the trigger menu button to get to the second and third images, the first comes from the channel selection menu button.

Here is a link to a video of Dave showing this:

OK... I see.

Looks like they have not fixed this AC triggering issue.

[edavid]

I have now received my new scope and have spent the last day or so getting to know how it works (well the basics anyway - it has a lot of options compared to my old analogue one!).

The details from the system menu are:

Software Version: 00.04.02.SP3
Board Version 0.1.1

That is not a new firmware version