Poll

Can you reproduce the Freeze-up Bug as described in this thread?

Yes, can reproduce the freezing.
42 (39.6%)
No, can't reproduce the freezing.
64 (60.4%)

Total Members Voted: 101

Author Topic: Rigol DS1054Z Freeze-up Bug Confirmed ??  (Read 184504 times)

0 Members and 2 Guests are viewing this topic.

Offline John Coloccia

  • Super Contributor
  • ***
  • Posts: 1208
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #50 on: June 04, 2015, 02:31:54 pm »
I just shot Dave a PM explaining it.  We'll see if he has a chance to try it.  If he can get it to lock up too, I thought it'd make a fun little video. It really is kind of a strange setup, though.  I could see getting in there by accident, but I'm not sure why I'd ever do it on purpose.  Maybe looking for a glitch in some comm protocol, or something like that?
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #51 on: June 04, 2015, 02:33:29 pm »
What are the exact steps you need to follow from Factory Defaults to reproduce the bug.
You cannot have a bug that only happens with some scopes, that smells like a hardware bug.

Please watch my videos linked above, especially the second one.

Or....
1. Power off scope, disconnect probes, or connect CH1 probe to Calibrator or some other signal. My scope reproduces this bug reliably even with no signal input (no probes connected), or any signal to CH1.
2. To enter "factory default" mode, repeatedly press 5th dark grey left menu button while scope is starting up. Screen should come up in Chinese.
3. Select your language (I have only tested with English language.)
4. Note that 1 us/div is already set in Horizontal Scale. Note that "Auto" is already set in Acquire>Mem Depth. If not, set these now.
5. Enter Horizontal Zoom mode by pressing the Horizontal Scale knob.
6. Display>Persis Time> select 100 ms, or anything other than "min".
Scope freezes at this point.

Steps 5 and 6 can be reversed.

Or you can start with Memory Depth set to maximum, then when in Horizontal Zoom with 100 ms Persistence, select Acquire>Mem Depth> Auto and the scope freezes at that point.
The easiest person to fool is yourself. -- Richard Feynman
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #52 on: June 04, 2015, 02:38:22 pm »
I just shot Dave a PM explaining it.  We'll see if he has a chance to try it.  If he can get it to lock up too, I thought it'd make a fun little video. It really is kind of a strange setup, though.  I could see getting in there by accident, but I'm not sure why I'd ever do it on purpose.  Maybe looking for a glitch in some comm protocol, or something like that?

Or looking at a signal with some jitter, etc.  I was usually using "Auto" memory depth, and setting some persistence allows a more "analog" picture so lets one see the magnitude of the jitter a little better. That's how I found it. I don't think it's such an unusual setting combo, even though it took me a month of use before I actually encountered it.

Regardless of that, no legitimate combination of user settings should make a scope freeze up and require a reboot! This isn't Windows we are talking about here....
;)
The easiest person to fool is yourself. -- Richard Feynman
 

Offline OldDogSleeping

  • Contributor
  • Posts: 12
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #53 on: June 04, 2015, 05:52:22 pm »
I put that I can't reproduce the freeze in the poll, as no matter what I did using the probe calibration signal it worked fine without freezing.

However I thought I'd just try a faster signal, so connected up a 100 Khz square wave, and guess what ? It freezes with 100ms Persistence.

Bellow 100Khz, the scope becomes unresponsive and freezes as you zoom in, but it can be recovered by removing the signal, and hitting STOP.

anything bellow 10 Khz, and mine does not freeze.

Newish DS1054z 00.04.03
 

Offline Jeroen3

  • Super Contributor
  • ***
  • Posts: 4067
  • Country: nl
  • Embedded Engineer
    • jeroen3.nl
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #54 on: June 04, 2015, 06:15:07 pm »
Yet again, apart from becoming very slow, it does not freeze.
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #55 on: June 04, 2015, 06:41:17 pm »
I followed John C's sequence on my newly arrived DS1054Z  with no input signal and it 'Froze'. No reaction to any Button pushes!  I still have the demo options and have upgraded the F/W to the latest 00.04.03.  So definitely a somewhat severe bug if one ended up with that combination of timebase and persistance.  Power it down and everything comes up normal again fortunately. I haven't tried other combinations though.

 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #56 on: June 04, 2015, 06:52:19 pm »
I forgot to mention after powering up again the 100ms persistance was remembered and selecting Delayed Timebase again froze everything again. However the Delayed menu did not change to ON  on either occasion despite being selected. ie remained OFF.  Rather weird.
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #57 on: June 04, 2015, 07:05:25 pm »
Last time powering up for now I finally ended up locked up as mentioned in the first post of this thread. A session of pressing the fifth left button to recover in Chinese brought the scope back to life and returned everything to normal.  I hope Rigol take some notice of this thread's comments and update their F/W to cure this as it isn't very good to be able lock the scope up like this.
 

Offline John Coloccia

  • Super Contributor
  • ***
  • Posts: 1208
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #58 on: June 04, 2015, 07:09:20 pm »
When you managed to reboot and come back OK, did you turn the scope off quickly after noticing the lockup, or did you play around with it for a while?  Could be that things are just crawling along so slowly when it happens that it didn't even have a chance to record the current setup, so when you rebooted it didn't remember you'd just gone into the delayed sweep.  I'm almost wondering if after it's frozen, perhaps you can just leave it for 20 or 30 minutes, and it will eventually come back.  It would sort of be consistent with the other slow downs I saw.  Maybe it's not locked...it's just crawling along extremely slowly.
 

Offline Tainer

  • Contributor
  • Posts: 37
  • Country: 00
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #59 on: June 04, 2015, 07:34:52 pm »
Although I voted for "don't have the bug", I do agree that there is something seriously wrong with this mode. I noticed some minor graphical glitches while changing trigger settings, so I tried to take a screenshot of these glitches. It took about 6 minutes to write the image on a usb-stick and the images were corrupted. Tried different usb-sticks - same result.
« Last Edit: June 04, 2015, 07:40:14 pm by Tainer »
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #60 on: June 04, 2015, 08:18:04 pm »
After the first reboot I did not play around for long but just tried to select Delayed Timebase the persistence had remembered it was 100ms.  However on selecting Delayed the screen changed to part display view ie 2 traces but no 'blue - space -blue  window' but did not show Delayed ON.  This first time everything then froze and no buttons did anything.

After the Fifth button boot up and Chinese start up all worked OK again.   Following your suggestion I Set up the 'Frozen mode' again 100ms persistence and 1us Timebase then selected Delayed again. Once again it froze showing the 2 traces as above and Delayed apparently OFF.  But this time I left it for some time.  After some 10 minutes, I didn't keep watching, the full blue- space-blue window appeared and Delayed suddenly showed up as ON.  The scope then continued to operate normally again.  It would appear that the s/w has a bug such that it goes through a rather long s/w loop or perhaps it waits for the Memory to Fill up (Currently 24M at 12.0kpts 1GSa/s).

So your thoughts of a long slow response are correct.
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #61 on: June 04, 2015, 08:30:01 pm »
Although I voted for "don't have the bug", I do agree that there is something seriously wrong with this mode. I noticed some minor graphical glitches while changing trigger settings, so I tried to take a screenshot of these glitches. It took about 6 minutes to write the image on a usb-stick and the images were corrupted. Tried different usb-sticks - same result.

Wow... that is really weird. I have seen some unreasonably long delays in writing a picture to the USB stick if I try to save while the scope is running. Sometimes it is pretty fast but usually not. So now I generally Stop the scope before I save a picture, this seem to work to make it save much faster.  If you Stop the scope, do you still get outrageously long delays saving, and corrupted files?



By the way, I think I've found another difference between firmware 04.02 and 04.03. I think I recall that when the scope is Stopped, the Hardware Frequency Counter stopped showing a valid value in 04.02 and showed "<15Hz" instead. But now in 04.03 the value in the HW Freq Counter stays at its value when the scope is Stopped. This is good.
The easiest person to fool is yourself. -- Richard Feynman
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #62 on: June 04, 2015, 08:35:53 pm »
After the first reboot I did not play around for long but just tried to select Delayed Timebase the persistence had remembered it was 100ms.  However on selecting Delayed the screen changed to part display view ie 2 traces but no 'blue - space -blue  window' but did not show Delayed ON.  This first time everything then froze and no buttons did anything.

After the Fifth button boot up and Chinese start up all worked OK again.   Following your suggestion I Set up the 'Frozen mode' again 100ms persistence and 1us Timebase then selected Delayed again. Once again it froze showing the 2 traces as above and Delayed apparently OFF.  But this time I left it for some time.  After some 10 minutes, I didn't keep watching, the full blue- space-blue window appeared and Delayed suddenly showed up as ON.  The scope then continued to operate normally again.  It would appear that the s/w has a bug such that it goes through a rather long s/w loop or perhaps it waits for the Memory to Fill up (Currently 24M at 12.0kpts 1GSa/s).

So your thoughts of a long slow response are correct.

Also very interesting. I think I've had mine sit for longer than that in "frozen" mode but I don't really have any data. So I'm doing a trial right now. I've got a circuit on the bench that I'm running some tests on and I'm taking a coffee break so it's an ideal time. Here's the current screen I'm looking at. From this, I just selected Auto memory depth and as soon as I selected 100 ms persistence, it froze. Started the stopwatch....   ;)
The easiest person to fool is yourself. -- Richard Feynman
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #63 on: June 04, 2015, 08:37:04 pm »
Just found another interesting thing once you have used Delayed with 100ms and frozen it, waited till it caught up every persistence value and ON/OFF of Delayed works almost instantly.  It would seem the first time one operates the Delayed with 100ms is the only occasion  'Freezing' for a long time occurs.  I haven't tried with other initial values of persistence to see if the Frozen period changes with different values.   Just maybe those other forum members who haven't seen this problem didn't try it from first boot up and had already filled the Memory with other activities.  Perhaps some of those who couldn't create the Frozen mode could try it from cold and see if they can then repeat the problem as well.
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #64 on: June 04, 2015, 08:53:13 pm »
Additional point I hadn't noticed that I was in stop mode when I switched  Delayed ON and OFF.  On pressing Run ie Auto it all froze again and I am timing properly this time!   So far 14 minutes frozen.
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #65 on: June 04, 2015, 08:55:49 pm »
23 minutes and still frozen.....

The easiest person to fool is yourself. -- Richard Feynman
 

Offline John Coloccia

  • Super Contributor
  • ***
  • Posts: 1208
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #66 on: June 04, 2015, 08:57:27 pm »
Mines been on for quite a while too, now.  I'm just going to leave it run while I'm soldering up some products.  We'll see if it ever comes back.  The fact that the slow ups are occurring, and that it sometimes comes back after a while, might help Rigol track down the problem.  Given that people are seeing different behavior depending on the signal going in might even suggest it could have something to do with triggering.

 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #67 on: June 04, 2015, 09:12:49 pm »
Mines been on for quite a while too, now.  I'm just going to leave it run while I'm soldering up some products.  We'll see if it ever comes back.  The fact that the slow ups are occurring, and that it sometimes comes back after a while, might help Rigol track down the problem.  Given that people are seeing different behavior depending on the signal going in might even suggest it could have something to do with triggering.

Maybe... but my scope still shows a perfectly valid, live triggered waveform while it's frozen.

40 minutes now and still frozen. I'll give it a few more minutes then I'm rebooting and getting back to work.

The easiest person to fool is yourself. -- Richard Feynman
 

Offline JohnPen

  • Regular Contributor
  • *
  • Posts: 240
  • Country: gb
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #68 on: June 04, 2015, 09:20:14 pm »
I have just got to 40 mins and it hasn't recovered this time so it looks like the Fifth button might be needed again.  That's it for tonight in the UK!
 

Offline John Coloccia

  • Super Contributor
  • ***
  • Posts: 1208
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #69 on: June 04, 2015, 09:33:21 pm »
Yeah, I gave up at the 45 minute mark too.
 

Offline Loboscope

  • Regular Contributor
  • *
  • Posts: 62
  • Country: de
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #70 on: June 04, 2015, 09:51:01 pm »
I have the DS1104Z (true 100 MHz), Board Ver. 0.1.1, SW 00.04.02.SP4.
No Freeze-up Problem, it seems only to response a little bit slower, but not worth to think about.

Greetings, Jürgen
 

Offline Tainer

  • Contributor
  • Posts: 37
  • Country: 00
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #71 on: June 04, 2015, 10:42:05 pm »
Wow... that is really weird. I have seen some unreasonably long delays in writing a picture to the USB stick if I try to save while the scope is running. Sometimes it is pretty fast but usually not. So now I generally Stop the scope before I save a picture, this seem to work to make it save much faster.  If you Stop the scope, do you still get outrageously long delays saving, and corrupted files?
After pressing 'Stop' screenshots are saved without problems, graphical glitches disappear as well. Glitches appear on 1uS/div when I change trigger level (notice the blue line at the bottom of the screen).
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #72 on: June 04, 2015, 10:55:45 pm »

After pressing 'Stop' screenshots are saved without problems, graphical glitches disappear as well. Glitches appear on 1uS/div when I change trigger level (notice the blue line at the bottom of the screen).

I have seen that blue (or maybe green)  line also, for the first time just today. I can't reproduce it "on demand" but it was definitely there once or twice today.

I also noticed in the Trigger menu the "Sweep" item would be completely blank and there would be no LED lit indicating sweep mode above the trigger "Mode" button. Normally always one or another of the three lights is lit and there is always either "Auto", "Normal" or "Single" in the "Sweep" menu item box.
The easiest person to fool is yourself. -- Richard Feynman
 

Offline alsetalokin4017Topic starter

  • Super Contributor
  • ***
  • Posts: 2055
  • Country: us
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #73 on: June 05, 2015, 09:48:16 am »
So what is happening here? We have a real mystery.

-A show-stopping bug has been found that seems to affect roughly half of the DS1054Z units tested.
-The bug affects scopes running firmware 04.02 or 04.03.
-The bug doesn't care whether the scope's Options have been unlocked or not -- several reports of totally "virgin" never-been-hacked scopes with the bug have emerged. The bug affects stock 1054 units and those that have been "unlocked" to 100MHz status.
-The show-stopper might also have other minor bugs associated with it, like slow-downs, corrupted file-saves, a screen artifact associated with setting Trigger Level, and other strange behaviours associated with the Horizontal Zoom (aka Delayed) mode.
-Some users who initially reported not being able to reproduce the bug, eventually were indeed able to do it by the proper combination of settings.

So how can this be? There must be some difference between scopes that have the bug and scopes that do not, and it doesn't appear to be related to the firmware or the "hacking" history. A hardware issue? Some difference at the component level somehow? That affects only half the units? This is very strange.

I'd really like to hear Dave's take on this.
The easiest person to fool is yourself. -- Richard Feynman
 

Online Bud

  • Super Contributor
  • ***
  • Posts: 6877
  • Country: ca
Re: Rigol DS1054Z Freeze-up Bug Confirmed ??
« Reply #74 on: June 05, 2015, 02:36:17 pm »
So what is happening here? We have a real mystery.
..
A hardware issue? Some difference at the component level somehow?

My friend You may not be too far from the truth.

What is happening is very simple and I have been saying this for months:

Rigol have no idea what they are doing

This was proved in another thread for their bad oscilloscope ADC clock case. It was clear they had no idea how the circuit works, how to select design parameters, how to select components, how to layout the circuit. No quality control was performed, otherwise this defect would have been caught right away.

I myself have a DS2072 and it caused me direct dollar loss in terms of my lost time before I realized what rubbish it was. It only took me less than 24 hours to be hit with at least two defects - lock ups and that the scope did not memorize its 50 Ohm input setting between power-ups. You check the setting in the menu and it says it was on 50 Ohm, where is in fact the input was on high impedance. That mislead me and caused me wasted time which cost money. There is no excuse for this type of shit to be happening.

I bought 2072a scope to make my life easier, but it made my life harder. I have been saying since then anyone buying rigol junk will sooner or later regret they did. Reading this and other threads I can see it is now happening that this junk began affecting their owner's performance. Why do you, I and others have to spend time fighting with this POS instead of using this time for their circuit testing. What kind of instrument is it if it requires fixing it first. Oh year, I can already hear someone singing the familiar "cheap, best bung for the buck !!" mantra. But is it? Cant you guys see that you now waste your time troubleshooting this shit? Is that what "best bang for the buck" supposed to look like ?

In regards to your hardware issue guess, it well may be the case. I have opened my 2072a to fix the ADC clock problem and what I found beside that problem was bunch of other hardware related problems all over the board, along and across. In one instance reviewing a circuit that occupied 1 square inch of space on the  PCB I counted 10 design errors! Wrong component selection, wrong design, wrong circuit layout. 10 errors per sq inch, that tells me something. And that is:

- No more rigol shit of any kind on my bench. Never again.


Facebook-free life and Rigol-free shack.
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf