Rigol DHO800/900 Oscilloscope Bug Reports + Firmware

[maxspb69]

I do have the impression, that manually adjusting the Trigger in Video, that "knob" just does nothing, you never see the "T'D" - triggered - message, while in Sync on Pulse, you can adjust that.

Just a thought: Do you have your probe attenuation set to 1x or 10x?

I've already tried this.    Unfortunately  - NO. 

[axantas]

If you stop acquisition, you get the "screenshot" of that moment, and a clear signal

So it is basically working...

It is not a very important function, but nevertheless strange, they obviously do not care at RIGOL.

Today, few people work with analog video signals. So there were probably few complaints to Rigol about this issue.

And most of those people probably already have an oscilloscope that can do it.

It is "basically" working, just like every watch, even with depleted battery, shows the correct time once a day...
The probe and setup is 10X, yes. It is just a bit stupid, because it is quite a basic thing to fix that.
As I am a proud potential member of the Therapy Gruoup here, I do have an alternative.
I just - you know... try Rigol

[ebastler]

So it is basically working...
[...]
And most of those people probably already have an oscilloscope that can do it.

Yes, sure. Or one could naively assume that an advertised feature, and one which Rigol has managed to get right for the past 10 years, actually works.  

As it happens, I just had to check out the video signal from a vintage computer last week. Nice coincidence that the DS1054Z is back on my bench for now after the DHO1000 went back to Rigol. I would not have enjoyed dragging it out again just to take some measurements which the "new, improved, one class above" successor can't handle anymore.

In my book, a non-working trigger, combined with the ability to stop the acquisition manually at some random point in time, does not qualify as "basically working", by the way...

[maxspb69]

It is "basically" working, just like every watch, even with depleted battery, s

How can you call something that is completely broken: "basically working"? The Video trigger in the HDO series is completely non-working. What you wrote above doesn't require any synchronization. You can stop any oscillogram and study it. It doesn't require a trigger at all. Do not mislead people.

[ebastler]

It is "basically" working, just like every watch, even with depleted battery, s

How can you call something that is completely broken: "basically working"? The Video trigger in the HDO series is completely non-working. What you wrote above doesn't require any synchronization. You can stop any oscillogram and study it. It doesn't require a trigger at all. Do not mislead people.

You might have missed the irony in axantas' post; please re-read it. It was Fungus in his earlier "it is basically working" post who was a bit too generous with Rigol.

[maxspb69]

Maybe. I have been working too much for the last two days )))

[axantas]

Maybe. I have been working too much for the last two days )))

Yes sometimes I tend to be a bit ironic.

[Fungus]

It is "basically" working, just like every watch, even with depleted battery, shows the correct time once a day...
The probe and setup is 10X, yes. It is just a bit stupid, because it is quite a basic thing to fix that.

I mentioned it because the threshold voltage for serial decoders only works when probe attenuation is set to 1x.

(probe should be 10x for max bandwidth...)

[Fungus]

It is "basically" working, just like every watch, even with depleted battery, s

How can you call something that is completely broken: "basically working"?

Somebody said it shows a frame correctly when you stop the 'scope.

This reminds me of a previous post where the scope always shows multiple traces overlaid when it's running and displays a single wave when stopped. Maybe this is the same, I don't know what you're actually seeing on screen.

What's "waveform freeze" set to in your display settings? Play around with that setting.

[axantas]

It is "basically" working, just like every watch, even with depleted battery, s

How can you call something that is completely broken: "basically working"?

Somebody said it shows a frame correctly when you stop the 'scope.

This reminds me of a previous post where the scope always shows multiple traces overlaid when it's running and displays a single wave when stopped. Maybe this is the same, I don't know what you're actually seeing on screen.

What's "waveform freeze" set to in your display settings? Play around with that setting.

This is a screenshot of a video I made this afternoon. If you "Stop" it, there is a clear signal, but the continuous Signal shows like that. Source is a PAL signal from a digital source (test pattern). Markers on both scopes are 64uS - single line. The spikes on the Siglent are just glitches in the Signal. The Rigol shows the correct signal as faint background and something, I did not identify yet in the proximity of this line.

[BRN]

Hi everyone. I just read this entire thread before posting my observation. I have DHO804 with latest FW 00.01.02.00.02 from
https://download.rigol.com/cn/Firmware/Digital%20Oscilloscope/DHO800%26DHO900/DHO800_DHO900(Software)Updatev00.01.02.00.02.zip

I found that scope has several issues I found. My list is below.

When I set scope to remember last settings after reboot, this settings is not saved properly:
1a.) Trigger mode is always "Auto" after reboot (I using normal mode most of the time and I have to set it all the time after reboot)
1b.) Scope can't remember selected memory depth if it's more than 1Mpts. If I set 10, or 25Mpts for example. It always returns to 1Mpts
1c.) Utility>setup>fine/zoom button returns to 'fine' if I select 'zoom'
1d.) Utility>IO>Host Name cannot be changed. When I delete original Host Name to just RIGOL and press 'Apply', it always return to default long name with sn.

2.) When I save the setup to stp file and load it, it loads fine (even selected memory depth), but again, trigger mode 'Normal' is not. It's always Auto.

3.) When I hit 'default' and then time base is set to 2s (roll mode), vertical is set to 1mV/div, set memory depth to 25Mpts, input some data (I touched CH1 by hand) and fill whole screen. Then hit 'run/stop' to stop the screen, zoom horizontal time base to maximum and try to move waveform left or right (does not matter if by touch, or by rotary encoder). Waveform weirdly jumps back and forth. Overall it moves to desired side, but with small jumps with backward direction.

This was my findings as BUGs. And next is suggestions to improve user experience:

|>User should change trigger mode by dedicated trigger button. For example. When user press dedicated trigger button, trigger settings is displayed. Then by trigger button user should change mode and then by long pressing (for example for more than 500ms) the trigger window exits.
|>From Rigol 1054z I have, the dotted mode of display I personally miss. There is only Vector waveform display mode.
|>In Storage>Save, word 'Overlay' should be renamed to 'Overwrite' for better description of what will happen.
|>In Storage>Load setup, it would be nice if it was be enough to just select saved setup by tapping anywhere in the line where file name is instead of just end of line where select circle is.
|>In small on screen keyboard is displayed, there is delete button. That button should delete whole text before when user use long pres that delete button.
|>Dedicated button 'Default' should load default.stp from system, where user can save its own setup. True factory default can be loaded somewhere from system menu.
|>Full/half/none is options to display grid. But I miss the square looking grid instead of what it looks like in 'full' where one div is rectangle looking one.

I hope that this findings will be well considered. My mind is still blown from earlier model 1054z which is stunning product. Well done Rigol. Thank you very much.

[Fungus]

This is a screenshot of a video I made this afternoon. If you "Stop" it, there is a clear signal, but the continuous Signal shows like that. Source is a PAL signal from a digital source (test pattern). Markers on both scopes are 64uS - single line. The spikes on the Siglent are just glitches in the Signal. The Rigol shows the correct signal as faint background and something, I did not identify yet in the proximity of this line.

How are you connecting it to the 'scope? Are you using a 50 Ohm terminator?

[Lathe26]

Good new: created a CSV file that the AWG can load!

I generated a number of CSV files intended for other Rigol AWG models, I found 1 that the DHO924S can load.  Short version is that it can handle CSV files intended for a DG1022Z that use 8192 points in Frequency mode.  Attempts to use CSV files using Sample/s mode or a different number of points failed, as did attempts to use various other Rigol AWG CSV files.  The testing was not exhaustive so the various test CSV file are attached for folks to checkout and/or study.  Hope this helps folks.

Hardware tested on: DHO924S with v00.01.02.00.02 firmware.

[AceyTech]

This is a screenshot of a video I made this afternoon. If you "Stop" it, there is a clear signal, but the continuous Signal shows like that. Source is a PAL signal from a digital source (test pattern). Markers on both scopes are 64uS - single line. The spikes on the Siglent are just glitches in the Signal. The Rigol shows the correct signal as faint background and something, I did not identify yet in the proximity of this line.

How are you connecting it to the 'scope? Are you using a 50 Ohm terminator?

FYI: Most NTSC and PAL video drivers are designed to drive 75\$\Omega\$, and it still doesn't work correctly if it's 75 Ohm terminated.

[Fungus]

FYI: Most NTSC and PAL video drivers are designed to drive 75\$\Omega\$, and it still doesn't work correctly if it's 75 Ohm terminated.

Just asking because it looked like a ghost signal in the line.

Set your trigger holdoff to 12ms and see what happens.

[ebastler]

Set your trigger holdoff to 12ms and see what happens.

This is not how video triggering is supposed to work. You want the ability to trigger on (a) any frame (vertical sync), (b) odd or even frames only, (c) any line (horizontal sync), (d) line #xx in each frame.

A holdoff time alone won't get you any of this. You need a combination of sync pulse duration, levels, and pulse counting, made more complex by the fact that inverted line syncs are sent during the vertical sync time. You can "roll your own" video trigger conditions via filtering, pulse duration and pulse counting triggers, but that's not fun -- there is a reason why scopes still provide dedicated video trigger modes.

[Fungus]

Set your trigger holdoff to 12ms and see what happens.

This is not how video triggering is supposed to work.

Of course.... but the fact that it appears to be very very close to working makes me wonder what happens if you give it a little bit of help.

[ebastler]

Of course.... but the fact that it appears to be very very close to working makes me wonder what happens if you give it a little bit of help.

To which of the four video trigger modes I had mentioned is it "close to working" in your opinion?

You can't even trigger on "any line" in a simple way, since the inverted pulses during the vertical sync period will cause spurious triggers. Pulling out the vertical sync via a low-passs filter is probably the best bet if proper video triggering is not available.

[axantas]

This is a screenshot of a video I made this afternoon. If you "Stop" it, there is a clear signal, but the continuous Signal shows like that. Source is a PAL signal from a digital source (test pattern). Markers on both scopes are 64uS - single line. The spikes on the Siglent are just glitches in the Signal. The Rigol shows the correct signal as faint background and something, I did not identify yet in the proximity of this line.

How are you connecting it to the 'scope? Are you using a 50 Ohm terminator?

This is just a single frame of a video i made, to show the difference between the Rigol and the Siglent. Identical setup without any terminator. On a real screenshot this is not visible.

[Fungus]

To which of the four video trigger modes I had mentioned is it "close to working" in your opinion?

I was told it worked when you press "STOP", that it was just unstable when free running.

If so, that seems "close to working".

[ebastler]

To which of the four video trigger modes I had mentioned is it "close to working" in your opinion?

I was told it worked when you press "STOP", that it was just unstable when free running.

If so, that seems "close to working".

My understanding, in line with what I saw in a quick test with the DHO1074 when I had it: When you stop the acquisition, you see a "stable" signal -- stable in the sense that it is no longer flickering and changing (of course). But you will not catch it in the same, stable phase every time you press STOP; it will be essentially a random point in time.

[axantas]

To which of the four video trigger modes I had mentioned is it "close to working" in your opinion?

I was told it worked when you press "STOP", that it was just unstable when free running.

If so, that seems "close to working".

My understanding, in line with what I saw in a quick test with the DHO1074 when I had it: When you stop the acquisition, you see a "stable" signal -- stable in the sense that it is no longer flickering and changing (of course). But you will not catch it in the same, stable phase every time you press STOP; it will be essentially a random point in time.

Indeed. just some random frame. But clear.

[pakakezu]

Good new: created a CSV file that the AWG can load!

I generated a number of CSV files intended for other Rigol AWG models, I found 1 that the DHO924S can load.  Short version is that it can handle CSV files intended for a DG1022Z that use 8192 points in Frequency mode.  Attempts to use CSV files using Sample/s mode or a different number of points failed, as did attempts to use various other Rigol AWG CSV files.  The testing was not exhaustive so the various test CSV file are attached for folks to checkout and/or study.  Hope this helps folks.

Hardware tested on: DHO924S with v00.01.02.00.02 firmware.

Sir you solved ARB loading on the DHO924S. Tested the format and works perfectly. I wasted days trying and failing. Thank you!

[AceyTech]

Set your trigger holdoff to 12ms and see what happens.

This is not how video triggering is supposed to work. You want the ability to trigger on (a) any frame (vertical sync), (b) odd or even frames only, (c) any line (horizontal sync), (d) line #xx in each frame.

A holdoff time alone won't get you any of this. You need a combination of sync pulse duration, levels, and pulse counting, made more complex by the fact that inverted line syncs are sent during the vertical sync time. You can "roll your own" video trigger conditions via filtering, pulse duration and pulse counting triggers, but that's not fun -- there is a reason why scopes still provide dedicated video trigger modes.

That's so awesome.  I wrote up a big explanation last night, but got too tired to finish and post it.  Good looking out., thanks!

p.s., Video trigger is one of the 2-3 considering factors that influenced my DHO purchase.  I hope they fix it "soon".

[lunix]

Update- I've had some issues with my brand-new DHO924S and the PLA2216 accessory.  I got a call from one of Rigol's FAEs today.  They are aware of some problems and seem to be trying to fix them.  They really want the bug reports, he said.  My scope hasn't crashed since the update (never had any piece of test equipment crash in decades of doing R&D).  And the PLA2216 is detected, seemingly reliably, on startup now, since taking their suggestion of hitting the "default" button.

We'll see what gets fixed and when.  He said that Rigol will post firmware updates on their site as soon as they're available.  I will say that I wish they'd keep the old ones, and give the release date for each so that users can see how often they're released before they buy.