Author Topic: Rigol DS1000Z series (ds1054z, ds1074z, ds1104z and -s models) Bugs/Wish List  (Read 120405 times)

0 Members and 2 Guests are viewing this topic.

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
Please follow this link to the discussion thread on version 00.04.04.03.02 forward:

  http://www.eevblog.com/forum/testgear/rigol-ds1000z-series-buglist-continued-(from-fw-00-04-04-03-02)/



Here's my attempt to catalog the bugs and wish list items for the Rigol DS1000Z oscilloscopes.   This should apply to the regular and -s versions, but may not apply to the MSO versions)  I'll try to modify this top post as people add their found bugs and wish list items.  Hopefully Rigol will see this thread and respond to it.

This list is for software version: 00.04.02 and 00.04.03  only, with the current hardware (possibly only board 0.1.1).


Rigol fixed the following bugs (according to their documentation between 4.02 and 4.03):

  • Modified the method of key board test and screen test and optimized processing speed.
  • The display of keyboard lights and the show of menu was wrong while switch the timebase mode between YT and ROLL in the Normal trigger mode.
  • The saved data of REF waveform was wrong while STOP the system then open the channel.
  • The output of AUX was wrong while the function of Pass/Fail was closed.
  • The display of LA signal was wrong while adjust the delay calibration of LA signal.
  • The position of LA wave and label of LA channel was wrong while load the file of LA waveform then RUN the system.
  • The wave of channel1 can not be displayed while open the scan mode then switch the channel2 over and over again.
  • The position of LA wave was wrong while load the file of LA waveform before change the position of the LA channel.
  • The memory data of LA channel was wrong when use the remote command to read.
  • Save the CSV file failed while the system stopped.
  • Fixed the bug of unlocked the key board failure when save the file.



Bugs:

1) The new Large and Extra Large measurement fonts are nice, but they are reset to Normal on each power cycle.  This should be retained like all the measurement settings are.

2) The Select Item sub-menu on the Measurement menu has 5 check boxes implying that it should be useable for Normal font too, but it's grayed out.  The only way to remove a measurement from the screen using Normal font is a power off/on cycle.

One member has already suggested that the left side menu as a series of on/off toggles as the means for including/removing the measurement.  This in conjunction with the Select Item sub-menu (which could turn on/off the display position of an item) might give maximum flexibility and ease of use.

3) The FFT function only works for channel 1.  However it will work for the other channels as long as channel 1 is turned on.  It's actually more complicated than that for more details see this thread: http://www.eevblog.com/forum/testgear/another-rigol-1000z-bug/msg568478/#msg568478

Also, the FFT cdisplay shows bad aliasing under most circumstances.  See this thread for more details: http://www.eevblog.com/forum/testgear/rigol-ds1054z-fft-bug%28s%29/

4) The Large and Extra Large fonts cover up items that display at the bottom of the screen.  For example the brightness indicator and the math equation. 

5) Some users have reported that the zero line of the trace is not perfectly centered after doing a calibration.  This is not an issue on all units, however, so might be more of a hardware variability item.  Users expect that all traces along with their channel number indicator should perfectly centered on the display (on top of each other) after a calibration.

(See this post for some post-calibration screen shots:  http://www.eevblog.com/forum/testgear/rigol-ds1000z-series-%28ds1054z-ds1074z-ds1104z-and-s-models%29-bugswish-list/msg584977/#msg584977)

If it's not possible for the calibration routine to automatically do this, maybe allow the user to manually set the channel offset at the exact center and save that value until the next calibration.

6) Writing full memory to a CSV file takes an extremely long time.  Write speeds seem to be about 0.15 MB/sec instead of the expected 10MB/sec or so.

7) When show measurements from the MATH trace the units don't match to the units selected (or computed) in the MATH function.  For example:  Ch 1= Volts, Ch 2=Amps, Math= Ch1 x Ch2 will show Watts on the trace but MATH measurements will show Volts, instead of correctly showing Watts.

8 ) The 20M Bandwidth Limit setting for each channel is not saved, it resets of OFF each power cycle.

9) If the following settings are used:

  Input signal: probe compensation 1kHz square wave from the scope
  Trigger: DC / CH1 / 1.5V
  Aquisition: High Res / 120k mem. depth
  Horizontal position: 0ps
  Horizontal scale: 50us/div

  The scope will display the waveform shifted by around 75us.    See details in this picture:



10) Some users have reported scope lock ups when setting memory depth to AUTO, timebase to 1us, zooming in and changing the persistence from MIN something else. Apparently not all users experience this problem but those that do can reproduce it every time.  See this thread for much more detail: http://www.eevblog.com/forum/testgear/rigol-ds1054z-freeze-up-bug/

11) Remote I/O menu will not allow selection of LAN if USB is ON.  You need to unplug the USB device first.

12) Inside the CSV file, the scope stores the "increment", in the memory dumps that increment is 25X longer than it should be to correctly represent the expected sampling period.  For example: if the increment is 10us, one has to set the time increment to 400ns (2.5MSPS) in order to get the correct plot of data.

13) RAW mode of data acquisition over USB returns garbage, rather than the waveform.


Wish List Items:

A) Full screen X-Y mode, the current display is tiny.  Also, allow screen persistence to be set in X-Y mode.

B) Decoders should allow DATASRC to be either Trace or Memory.  This is documented in the manual but apparently only Trace is implemented, so this might be a bug.  Even the SCREEN option could be improved so that decoding starts at a start-bit not at the left edge of the screen.

C) When displaying a math function allow the underlying channels to be hidden.

D) Compress and move the waveform to the bottom of the screen when Measure All is on as is done, optionally, for FFT display.  Also for measurement history.

E) Allow Measure All to be selected in the AUTO options menu.

F) Allow the user to hide the left and right side menus.  Also the left side menu is often irrelevant and that screen space could be used for other display items.

G) When selecting a channel, show some corresponding channel color around the settings instead of having everything in just white.  Maybe just change the color of the tiny CH1, CH2, CH3, CH4 label next to the Coupling setting.

H) Use a larger font for the popup keyboard.

I) Save the last few file names in the Storage menu for faster access.  Maybe use the left side menu for this?  Alternatively, just remember the last file name entered and allow the user to increment any trailing digits.  So TEST0001 would go to TEST0002 automatically (or will a key press).

J) Allow the trace Display Persistence Time to be turned off -- instead of just MIN.  The scope would then just show each waveform as it's captured and blank the display before showing the next.  Even the MIN persistence level can hide some detail.

K) Allow larger fonts for certain items.  For example:  timebase, trigger level and channel gain (volts/div).  Also use a larger icon showing the coupling (DC/AC/GND) next to the channel gain.

L) Use a different color for the math trace, something that stands out better, maybe red?

M) Allow the user to select a color for the A and B cursors, individually.  Currently they are distinguished by a solid and dotted line both in white.

N) Allow for selectable FFT size (number of points: 128, 256, 512, 1024) calculation -- ie, trade off speed for accuracy.

O) Show the trigger type and coupling method (DC/AC/LFR/HFR) next to the trigger type icon in the upper right corner.

P) Consider moving the selection of the trigger coupling method to the top-level trigger menu instead of the Setting sub-menu.  Same for the Holdoff.

Q) Allow for variable high and low pass filtering for each channel, as was done in the earlier 1000 series scopes.

R) Add boxes for Math and Ref to the bottom bar alongside the channel info.  Color coding them and added relevant info such as gain or offset offset voltage.

S) Allow for a simple screen title.  Maybe allow more text to be added via a PC connection, perhaps a query that could be responded to with a key press on the scope.

T) Hardware frequency counter shows "<15Hz" when the STOP button is pressed even though it had been showing a valid value.  It would be best to freeze this value like other measurements are.

U) Ability to save a Recorded data to USB drive.

V) Consider checking the way the menu system works, as follows:

Push a soft menu button to open a menu, push the same button again to close the menu.
When a menu is open you can use the light blue up/down arrows to select an item in the menu, or use the multifunction knob if you prefer that.  When you have the correct item highlighted in the menu, push the soft menu button again to select the item and close the menu (or push the multifunction knob if you prefer).

W) Change file file numbering scheme so that instead of using the first available file number simply use a sequential number (stored in persistent memory) until reset by the user.

X) Use a meaningful amount of decimals in the horizontal position display (e.g. '0.00000000 ps' should be something like 0.0 ns since the smallest step size is 0.1ns)

Y) Have the FFT function compute an average if the channel it's set to is set to "average" in Acquire menu.  Other single-argument math functions seem to do this.

Z) Allow a push of the trigger level control to set the trigger level at the 50% level or at the 0% level - as it currently does.

AA) Improve the response of the vertical and horizontal position movements, they have quite a bit of lag.

AB) It would be helpful to be able to see the on-screen ASCII table when setting up an RS-232 trigger.

AC) Add (wired/wireless) USB keyboard support for navigating menus and character input.

AD) Add a menu option to allow the user to set the offset voltage (or whatever the current units are).  This is in addition to the normal offset control knob.

AE) Allow MATH to be used in the X-Y mode.  This means that instead of 6 options of what to display there would be 20 options (so that MATH can be either X or Y). 





« Last Edit: April 12, 2017, 10:53:06 PM by kwass »
-katie
 

Offline Muttley Snickers

  • Supporter
  • ****
  • Posts: 1586
  • Country: au
  • Don't Laugh, I'm Serious.
I dont own one as yet but these would have been nice for me:

D) Real Time Clock

E) Mouse Support



Muttley
One smart cookie, better make that two for good measure.
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
Quote
D) Real Time Clock

The hardware doesn't allow for that, so I don't think we can reasonably add that to the wish list, but this would be nice to have.

Quote
E) Mouse Support

I suppose that this would be possible with the current hardware, but I don't think that Rigol offers mouse support on even their high-end scopes.  I'd like to keep the wish list limited to things that would appear to be not too hard to code -- but for all I know even items A, B and C might be hard to do depending on how they wrote their firmware.

-katie
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4125
  • Country: cl
  • Canadian po
External trigger on rear panel BNC...

I don't think asking for a hardware change in a firmware will get you far.  >:D

I think asking for some things here is interesting, but I wonder how much will result from it.

Rigol has made a really cheap scope with a bunch of features. IMHO they have failed with some things but for the price I am sure it can't be made much better. For example the FFT only on channel #1. No big deal, just use channel #1 when you need FFT!

One of the other things that bug me, the FFT is SLOOOOOWWWW. When compared to the DS1052E it simply is a snail and IMHO almost unusable. I am sure that with all the things they are trying to pull off with the single processor that it probably cannot get faster.

Another thing that is simply a mess is the measurement displays. If you "clear" them they don't go away. They still just sit there with a grey color and interfere with adding a different measurement. The only way to get rid of them and start from scratch is to clear all and reboot.

Another, make it so the right menus can be disabled and show more screen. As it is now the displayed waveform has basically no more screen than the DS1052E. Yes I know it is higher resolution.

Last one for now, make the persistence so it can be 100% disabled.

The problem with any of these requests is they might not have much room to make improvements because of memory and timing restraints in the hardware. This is what you get for a bottom feeder price with so many features.
« Last Edit: January 11, 2015, 01:56:04 PM by Lightages »
I am NOT a distributor for Brymen.
 

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 2131
  • Country: ca
From the poor eyesight guy.

- Keyboard 5 s S 8 all blend together, make the keyboard bigger, way bigger.
- A little color around the ch1-4 selections. I know there are other indications but you push these buttons so a slight colour change would be a good indicator that they are active. Don't change the text colour that will make it harder to read.

Annoying bits.

- When Measure All is pushed the waveform will not move to a more visible area. That's pretty standard.
- Under Auto Options can't there be an option to turn on the Measure All.?
- Under storage, save the last five items so you can either reload them or see what you did.
- Under the vertical and horizontal menu have a location to save the current "on" items. That way you could just hit one button to load your usual measurements.
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
For example the FFT only on channel #1. No big deal, just use channel #1 when you need FFT!

Agreed, it's just a small bug, not a wish list item.

Quote
Another thing that is simply a mess is the measurement displays. If you "clear" them they don't go away. They still just sit there with a grey color and interfere with adding a different measurement. The only way to get rid of them and start from scratch is to clear all and reboot.

This is fixed in the latest firmware, there's a Select Item sub-menu

Quote
Another, make it so the right menus can be disabled and show more screen. As it is now the displayed waveform has basically no more screen than the DS1052E. Yes I know it is higher resolution.

Yeah, that would be nice, I'll add that to the list.

Quote
Last one for now, make the persistence so it can be 100% disabled.

100% disabled to me would mean that you'd never see anything.  Are you looking for an analog-type display -- real time trace mode?  Is that even possible with this hardware?
-katie
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
From the poor eyesight guy.

- Keyboard 5 s S 8 all blend together, make the keyboard bigger, way bigger.
- A little color around the ch1-4 selections. I know there are other indications but you push these buttons so a slight colour change would be a good indicator that they are active. Don't change the text colour that will make it harder to read.


It would be good to have more readable fonts in some places.   I'll add these suggestions.

Quote
Annoying bits.

- When Measure All is pushed the waveform will not move to a more visible area. That's pretty standard.
- Under Auto Options can't there be an option to turn on the Measure All.?
- Under storage, save the last five items so you can either reload them or see what you did.

I agree, the measure all is pretty useless because of that -- adding to the list.....

Quote
- Under the vertical and horizontal menu have a location to save the current "on" items. That way you could just hit one button to load your usual measurements.

You can sort of do that now with the new Select Item sub-menu.  But it's not quite what you're asking for I think.
-katie
 

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 2131
  • Country: ca
- After a cal 0v does not line up dead center on the graticule, it's not off just not dead center. Just seems weird.
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us

OP did not specifically say "firmware changes for the next update", therefore I believe it is reasonable to ask for Rigol to modify the hardware in DS1000AZ or whatever their next hardware revision nomenclature may be, just as they included the 50 ohm terminator in the DS2000A series.

I was really thinking just about firmware changes, since a lot of people have these scope and would just like to have the bugs fixed and/or some nice little extras added.

Quote
It will have about as much chance of getting Rigol's consideration as any of the suggestions that may come up in this thread, unless someone like Teneyes or any of the other regular Rigol correspondents makes a deal out of it.

I disagree.  They picked up on the measurement font size discussion and added the Large and Extra Large options in the new firmware.  (They did this silently not mentioning it when they released the firmware, perhaps it's a work in progress.)  As I recall this was discussed in a one of the threads in this forum and didn't get the high profile coverage that the jitter and AC coupling problems did in the BLOG forum.
-katie
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
- After a cal 0v does not line up dead center on the graticule, it's not off just not dead center. Just seems weird.

I saw the discussion about this, but don't quite understand.  Is the trace dead center and just the channel number indicator off by a pixel or so?  That's what mine looks like after a cal.
-katie
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4125
  • Country: cl
  • Canadian po
The select sub menu item does nothing to fix the ghost measurement items still hanging around until a clear all and reboot. It only has effect on selecting which items to see when using larger fonts.
I am NOT a distributor for Brymen.
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4125
  • Country: cl
  • Canadian po
Oh yes, the persistence option. The display shows some amount of persistence all the time, meaning showing a blend of history with real time. Without persistence the display would show only the latest acquisition each screen refresh. The persistence as implemented hides some noise in the signal.
I am NOT a distributor for Brymen.
 

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 2131
  • Country: ca
- After a cal 0v does not line up dead center on the graticule, it's not off just not dead center. Just seems weird.

I saw the discussion about this, but don't quite understand.  Is the trace dead center and just the channel number indicator off by a pixel or so?  That's what mine looks like after a cal.

It's only off by a pixel or two (perhaps less). Basically after a cal you would expect all traces to line up perfectly, dead center. That may change an hour later but just after a cal it should be perfect. What makes it unusual to me is that I have never seen any digital scope do it. It's one of those things that is just expected and never stated.
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
It's only off by a pixel or two (perhaps less). Basically after a cal you would expect all traces to line up perfectly, dead center. That may change an hour later but just after a cal it should be perfect. What makes it unusual to me is that I have never seen any digital scope do it. It's one of those things that is just expected and never stated.

After a cal the traces on my 1054z are perfectly centered (of course this changes over time after a cal).  The color of the trace is the last channel turned on and I see no hint of other channel "under" that.  I'm guessing that there is some slight variability in the hardware and/or temperature gradients that make this differ from scope to scope.  I put a very low speed fan in my scope so perhaps this allows for a more even temperature across the channels.

If this is the case, I don't see how firmware can correct it.
-katie
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
The select sub menu item does nothing to fix the ghost measurement items still hanging around until a clear all and reboot. It only has effect on selecting which items to see when using larger fonts.

Yes, of course!  I hope that they're planning to fix this, why are there 5 check boxes in the Select sub-menu if the large and extra large font's can only show 3 and 2 measurements.  I'm putting this in the BUG section.
-katie
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
Oh yes, the persistence option. The display shows some amount of persistence all the time, meaning showing a blend of history with real time. Without persistence the display would show only the latest acquisition each screen refresh. The persistence as implemented hides some noise in the signal.

OK.  I've got it on the wish list.
-katie
 

Offline LaurentR

  • Frequent Contributor
  • **
  • Posts: 384
  • Country: us
3) The FFT function only works for channel 1.  However it will work for the other channels as long as channel 1 is turned on.

The FFT issue is a tad bit more complex than that and it's definitely a bug, not a missing feature. On my scope, I originally couldn't FFT on channel 1 unless channel 2 was on  |O
It has to do with the "SourceA" setting for dual-operand math operators. Try this:
* Turn Math ON
* Pick a 2-operand op (e.g. A+B)
* Select Ch2 as SourceA
* Set Operation to ON
* Now change the op to FFT.
* Set Source to Ch2
* Voila, it works, even if Ch1 is Off (and now FFT on Ch1 may not work anymore :palm:)

So it seems that regardless of the "Source" setting for the FFT, if the "SourceA" setting for any 2-operand op is set to a channel that's Off, you get the DataInvalid message. I did report it to Rigol, who managed to reproduce the issue.

See this thread for some extra reference:
http://www.eevblog.com/forum/testgear/another-rigol-1000z-bug/msg568478/#msg568478
« Last Edit: January 11, 2015, 05:10:05 PM by LaurentR »
 

Offline avvidclif

  • Regular Contributor
  • *
  • Posts: 56
  • Country: us
I asked and it was reported to Rigol as a change that the Measurement select buttons on the left side for Horizontal and Vertical measurements be set up as a toggle. IE: Push once for on and again for off. It didn't make this upgrade version.

Oh well, maybe later.
Clif Holland KA5IPF
www.avvid.com
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
I asked and it was reported to Rigol as a change that the Measurement select buttons on the left side for Horizontal and Vertical measurements be set up as a toggle. IE: Push once for on and again for off. It didn't make this upgrade version.

Oh well, maybe later.

I've incorporated this into the wish list.
-katie
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
3) The FFT function only works for channel 1.  However it will work for the other channels as long as channel 1 is turned on.


See this thread for some extra reference:
http://www.eevblog.com/forum/testgear/another-rigol-1000z-bug/msg568478/#msg568478

Thanks for reminding me of all the details on this.  I've link to it.
-katie
 

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 939
  • Country: us
Wish List Items:

...

F) Allow the left and right menus to be hidden and expand the waveform to fill the extra screen space.

A nice list of Wishes, overall.  But this one is highly unlikely to ever happen.  The 1000z is hard-coded to display 12 time divisions (600 pixels), and may even have some hardware elements locked to that.  This was done to increase the display speed.  I can't see them ever changing that. 

And I suspect if you asked them directly (and they answered), they'd say if you really want 14 divisions, buy their DS2000 instead.  Their goal was not to kill all sales of their high-end product lines, by making their lowest cost econo-scope fully feature-compatible with them.
 

Offline pickle9000

  • Super Contributor
  • ***
  • Posts: 2131
  • Country: ca
Wish List Items:

...

F) Allow the left and right menus to be hidden and expand the waveform to fill the extra screen space.

A nice list of Wishes, overall.  But this one is highly unlikely to ever happen.  The 1000z is hard-coded to display 12 time divisions (600 pixels), and may even have some hardware elements locked to that.  This was done to increase the display speed.  I can't see them ever changing that. 

And I suspect if you asked them directly (and they answered), they'd say if you really want 14 divisions, buy their DS2000 instead.  Their goal was not to kill all sales of their high-end product lines, by making their lowest cost econo-scope fully feature-compatible with them.

Nailed that one, never happen.
 

Offline Lightages

  • Supporter
  • ****
  • Posts: 4125
  • Country: cl
  • Canadian po
Yeah I don't think it will be changed. I did not know that the screen size was hard coded. How sis you determine that?
I am NOT a distributor for Brymen.
 

Offline Mark_O

  • Frequent Contributor
  • **
  • Posts: 939
  • Country: us
Another, make it so the right menus can be disabled and show more screen. As it is now the displayed waveform has basically no more screen than the DS1052E.  Yes I know it is higher resolution.

Yes, there is twice the rez (600 vs 300 pixels), so the equivalent of 24 divisions on the 1000E. 

But that extra horizontal space is put to good use.  The permanent controls and menus on the side bars eliminate one of my biggest gripes about the 1000E (and D, and C, and CD) series.  Having the menus pop up and down constantly.  To me, that was just a PITA, and I think Rigol made the right choice there.  We somehow managed to get by with 10 divisions for decades, the 1000Z has 12, and some are unhappy it doesn't have 14 (as the DS2000 does).  Why stop there?  Why not 16, 18, or 20?
 

Offline kwass

  • Frequent Contributor
  • **
  • Posts: 317
  • Country: us
Yeah I don't think it will be changed. I did not know that the screen size was hard coded. How sis you determine that?

I'm going to scratch this off the list.  I think that Rigol should see a list of feasible changes not a lot of pie in the sky items.
-katie
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf