Author Topic: REVIEW - Rigol MSO5000. Tests, bugs, questions  (Read 24757 times)

0 Members and 1 Guest are viewing this topic.

Offline AngusBeef

  • Regular Contributor
  • *
  • Posts: 78
  • Country: us
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #25 on: September 10, 2019, 03:18:46 am »
This is a much better topic to post this than the hacking thread.

While some users reported 04.08 correct their overshoot/undershoot problems, the official readme in 04.08 does not specifically state this particular fix.

I presume you adjusted your probes with the reference signal, and did a self-cal after the upgrade?  As the upgrade alone will not fix the problem.

This issue seem to be hardware related, as it only affects certain users.  I don't have this problem even with 04.04 on all four channels, so I cannot test whether 04.08 makes a difference.

The word is Rigol is aware of this issues a few months back, and they are working on a firmware fix for it.  The question is whether it is fully implemented in 04.08, or you have to wait for a future release.  If it persists, you may want to let Rigol know so they can help you to fix it.

Thanks - removed from the other post - I've only got it on one channel, the other 3 look a hair undercompensated when it's as good as I can get it, so my fear is that it's a hardware issue...it's just disappointing that the new toy isn't perfect - but it's a Rigol so it happens. I'll get ahold of them and see what I can do. Thanks!
 

Offline AngusBeef

  • Regular Contributor
  • *
  • Posts: 78
  • Country: us
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #26 on: September 26, 2019, 08:50:36 am »
Used the lfcal.hex from @mabl, backed up both the lfcal.hex from /default and /data then overwrote both of them and my issue with overcompensation went away.  :-+

Self-cal completed, I can get still get all 4 probes compensated correctly - no ringing, no overshooting, etc.
« Last Edit: September 26, 2019, 08:52:39 am by AngusBeef »
 
The following users thanked this post: thm_w

Offline matlipinski

  • Contributor
  • Posts: 13
  • Country: pl
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #27 on: October 29, 2019, 12:35:26 pm »
Hi.

I am writing once again (but in this thread this time).

I have bought MSO5074 (hw. version 1.00) some time ago, and from the beginning I have overshoot undershoot issues, as well as square signal from arbitral generator has also 20% over and undershoot. After applying newest firmware  01.01.04.08 and lfcal.hex from mable and running self calibration they are still there. After applying mable's lfcal.hex they vanished from 1st channel, almost vanished from second channel, but they are still on 3rd 4th channel. Do You think i should send it to Rigol? What can I do else? Is it normal?

Please look at the screenshots.

There is also screen shot from Rigol 1054Z - square signal from 1KHz test signal of MSO5074 seems to be OK on 1054Z. Under/oveshoot is also visible on 1054Z.
 

Offline TK

  • Super Contributor
  • ***
  • Posts: 1604
  • Country: us
  • I am a Systems Analyst who plays with Electronics
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #28 on: October 29, 2019, 12:41:23 pm »
If you have 01.01.04.08 you should not apply other ifcal.hex file.  I think you have a faulty scope.  When was it purchased?  Have you contacted Rigol EU to ask if your unit has the latest HW fixes (LCD, fan noise)?
 

Offline matlipinski

  • Contributor
  • Posts: 13
  • Country: pl
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #29 on: October 29, 2019, 01:19:07 pm »
OK I will revert lfcal.hex to original state and test it again.
Yes I have asked Rigol, if I have all fixes applied, they confirmed.
I have asked Rigol if this over/usndeshoot is normal, they said no, and sent me beta firmware 03.01.04.04. On 03.01.04.04 it was like on 01.01.04.08.
After sending them screenshots from 03.01.04.04 they said that this time it is OK.
I think it is not, it is my third oscilloscope, I had never such issues before.
I couldn't test arbitral generator, I had no licence earlier.
 

Offline matlipinski

  • Contributor
  • Posts: 13
  • Country: pl
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #30 on: October 29, 2019, 09:51:50 pm »
I have posted in "New Low Cost Rigol MSO5000 Oscilloscope" and "REVIEW - Rigol MSO5000. Tests, bugs, questions" threads.

Could You please show me Your 1KHz signal from MSO5000 as well as 1 or 10KHz signal from arb. signal gen. on Your MSO5000?

I will ask Rigol for help.

After self cal. on factory lfcal.hex it looks like this:
 

Offline TK

  • Super Contributor
  • ***
  • Posts: 1604
  • Country: us
  • I am a Systems Analyst who plays with Electronics
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #31 on: October 29, 2019, 10:33:42 pm »
CH1 and CH3
[attach=2]

1KHz square wave from Wavegen, direct COAX cable with 50ohm passthrough, 1X
[attach=1]
 
The following users thanked this post: matlipinski

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #32 on: November 07, 2019, 09:24:05 pm »
Got two "new" bugs/missing features:

Power on state:
Choosable are "Last" or "Default" - None of them works, in every case after power on my scope has the optinon menu open.

Cursors :

The horizontal and vertical cursor-lines are always together dispaying, you can´t choose between only vertical, only horizontal or both.


Offline thm_w

  • Super Contributor
  • ***
  • Posts: 2350
  • Country: ca
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #33 on: November 07, 2019, 11:19:36 pm »
Got two "new" bugs/missing features:

Power on state:
Choosable are "Last" or "Default" - None of them works, in every case after power on my scope has the option menu open.

The "Last" or "default" is more to save all of the configuration values: time base, probe attenuation, etc. I wouldn't expect it to save the exact state of the UI (although it could be nice if it did).

Seems like its storing a reasonable amount of data:
Quote
Syntax :SYSTem:SETup <setup_data>
contains 2506 bytes of effective data.
 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #34 on: November 07, 2019, 11:28:28 pm »
Quote
The "Last" or "default" is more to save all of the configuration values: time base, probe attenuation, etc. I wouldn't expect it to save the exact state of the UI (although it could be nice if it did).

Yes and no…

"Default" means to me, that the scope starts "like it was" on delivery state.
"Last" means to me that it will start with the last configurations I´ve taken before shut-down.

Seems logical....only to me ?

Offline SimonH

  • Contributor
  • Posts: 14
  • Country: ch
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #35 on: November 23, 2019, 09:33:19 am »
No. 17 is fine. You have to switch from 'screen' to 'memory'. Then you can save wfm. A bit strange, though, that by default, it saves only the screen.

FW FW v00.01.01.04.08


No. 1 is fine with HW version 1.01


One more thing: The protocol analyzer often just stops working. I think it has to do do with a faulty threshold. I've once also seen that it 'sees' some data, although the threshold was way beyond the waveform. And putting it to the right place made the decoding disappear. Not reproducible though.

 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #36 on: November 23, 2019, 11:42:01 am »
Quote
18a. No frequency and magnitude axes in FFT mode
[FW v00.01.01.04.04] / [FW v00.01.01.04.04]

18b. FFT does not have MAX and Average
[FW v00.01.01.04.04] / [FW v00.01.01.04.04]

18c. FFT size is not shown (and could not been set)
[FW v00.01.01.04.04] / [FW v00.01.01.04.04]

To 18b, it got peakhold, what MAX concerns.
To 18c, showing memorysize is indeed missing, sample-size is displayed instead.


Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #37 on: November 23, 2019, 09:44:14 pm »
Cursors :

The horizontal and vertical cursor-lines are always together dispaying, you can´t choose between only vertical, only horizontal or both.

Forgot to post the answer from rigol ( always rigol support in EU meant) :

For them it´s not a bug or missing feature, they suggest me to put the vertical or horizontal lines "away" when vertical or horizontal cursor value only are interesting... ;)

Offline SimonH

  • Contributor
  • Posts: 14
  • Country: ch
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #38 on: November 23, 2019, 11:11:16 pm »
One more thing about protocol analysis:

Apparently, the decoder only works if the context it needs to decode, i.e. a CAN frame, is visible on the screen. It seems to 'see' just what we see on the screen. Fair enough. BUT: If I have, say, a very long CAN frame, then I have to squeeze it so much that I cannot distinguish the data bits any more. And also cannot read what's the data any more. Of course, then there is still the table view. But it would be nice if I could zoom into the frame and actually see the data. This is not possible because as soon as I zoom in, it loses the context.

Am I missing something? Can I somehow "freeze" the decoding, so that it stays visible, keeps track on the waveform, but does not try to re-decode anymore? This would solve this issue.
 

Offline SimonH

  • Contributor
  • Posts: 14
  • Country: ch
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #39 on: November 24, 2019, 07:55:45 am »
Concerning my post above about protocol analyzer:

Of course I don't mean real-time encoding while a signal is being repeatedly captured, but inspecting a previously captured signal. Actually, I don't see any reason why the decoder should try to re-analyze the waveform every time I move or zoom it. Shouldn't it just decode (to get good speeds just the visible section) when it actually captures a signal, and as soon as the signal is stable, start a complete decoding of the whole memory content. And then just show that decoding, which is moved and zoomed simultaneously with the waveform? This would be simple, and one could zoom in into every detail.

One more thing: If I save a waveform (memory content, not screen) after having captured many bursts of communication into segmented memory ('recording'), I would assume the scope to store all these captures (the memory content, that is). Instead, only the actually seen segment is stored.
 

Offline SimonH

  • Contributor
  • Posts: 14
  • Country: ch
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #40 on: November 25, 2019, 10:49:03 pm »
I just found that if you enable the zoom function, you actually can zoom into details. As long as the context is visible on the "original" waveform, you can see the decoded data also in the zoom window. I though I have once seen this not working. But apparently it works.
 
The following users thanked this post: thm_w

Offline thm_w

  • Super Contributor
  • ***
  • Posts: 2350
  • Country: ca
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #41 on: November 25, 2019, 11:47:20 pm »
One more thing: If I save a waveform (memory content, not screen) after having captured many bursts of communication into segmented memory ('recording'), I would assume the scope to store all these captures (the memory content, that is). Instead, only the actually seen segment is stored.

May be possible to do it in a script:

Code: [Select]
save:format csv
save:csv:length maximum

for x in range(5):
   :record:current x
   :save:csv D:\x.csv

Depending on if the "save" functionality works the same in record mode as it does in other modes.
 

Offline matlipinski

  • Contributor
  • Posts: 13
  • Country: pl
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #42 on: November 26, 2019, 10:17:21 am »
For all whose MSO5000 has over/undershoot. I have bought MSO5074, and from the beginning i had over/undershoot problem. I contacted Rigol support. They sent me all possible firmware versions and told me: make self calibration procedure for each firmware version we have sent to you. But nothing changed.

I decided to send my MSO to them. They have fixed over/undershoot problem, and calibrated the scope. They have also applied newest hardware fixes (also to built in signal generator).
They also changed HW version number to 1.01. I had 1.00 earlier.

Now everything is fine.

Over/undershoots looked like attached bellow.
 

Offline irw

  • Contributor
  • Posts: 5
  • Country: pt
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #43 on: December 03, 2019, 12:39:48 pm »
Hi all,

How is the current status of the firmware? Still too buggy?

Thanks
« Last Edit: December 03, 2019, 01:03:16 pm by irw »
 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #44 on: December 03, 2019, 06:36:03 pm »
Hi,

It wasn´t to buggy even from the beginning, depending on what things you work on, you wouldn´t recognize them.

Offline irw

  • Contributor
  • Posts: 5
  • Country: pt
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #45 on: December 03, 2019, 06:46:34 pm »
My main concern is the vertical accuracy, as I read that it was about 4mV/div... at least the "True" value... As I need to measure low currents, I wonder how low can I get, if the accuracy is that "high"...
 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #46 on: December 03, 2019, 09:41:06 pm »
Hi,

This is a point which doesn´t have much to do with bugs, it´s a hardware thing.
When i remember it right, in the first FW update a 500µV Resolution was implemented.
"Of course" bandwith-limited and not really a true one.

Quote
As I need to measure low currents

Generally, scopes aren´t precisely as dedicated volt-/amperemeters.
If you need to measure dc-currents, you should do it better with an suitable amperemeter.
For example, we must record loadsteps and static values of our converters/inverters.
First one we do of course with a scope, but last we do with a high precision voltage/current/power device.
Because everyone knows that the vertical accuracy of a scope couldn´t reach the precision of a dedicated measure device.
For recording loadsteps the accuracy ist not the most important thing.
Finally we talk about a scope for 1000 bucks….
 
The following users thanked this post: thm_w, luma

Offline matlipinski

  • Contributor
  • Posts: 13
  • Country: pl
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #47 on: December 03, 2019, 10:22:33 pm »
Hi all,

How is the current status of the firmware? Still too buggy?

Thanks

I think FW 01.01.04.08 and HV 1.01 is usable, there is still a lot to do, but if are aware of limitations of the device, you are able generally measure what you want to.  :-+
 

Online Martin72

  • Super Contributor
  • ***
  • Posts: 1543
  • Country: de
  • Testfield Technician
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #48 on: December 03, 2019, 10:32:36 pm »
Nothing else I said before.. ;)

And what the "limitations" concerns, we got two scopes for 26000 bucks together in october.
Generally, they couldn´t do "more" as a scope for 1000....they could do some things better, but more...no.


Offline SimonH

  • Contributor
  • Posts: 14
  • Country: ch
Re: REVIEW - Rigol MSO5000. Tests, bugs, questions
« Reply #49 on: December 08, 2019, 09:44:42 am »
One more quite significant one, I guess:

While in the decoding menu, the available signals are named correctly (MOSI, MISO,..), in both the search and trigger menus they are names "i2c-ish" (SDA, SCL). I think, this typo has been mentioned before somewhere. But actually, it doesn't appear to be just a typo. SPI search and trigger do not work in any way. Decoding works fine.

FW 01.01.04.08

Hope this will be fixed in the release that appears to be about to come.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf