REVIEW - Rigol DS2072 - First Impressions of the DS2000 series from Rigol

[Wim13]

I never use the trigger outs, as Marmad said, just use one of the signals to trigger the other DSO also.

So i have 4 signals on the DS1000, and 1 of these signals is also connected to the ext trigger input of the DS2000


The jitter of the DS1000 is 8 nSec, independed of the input frequency, see picture below, displayed on the DS2000
the other signal is the orginal, that i can vary to any freq. the jitter will not change

[pa3bca]

But I don't think the annoyance of the jitter would be related to the frequency of the input signal (or DUT).

Yes of course, the jitter is always 8 ns, independent of the timebase and/or the observed waveform. What I meant to say is that (in my opinion) this method of locking both scopes is not very useful for displaying waveforms above (an arbitrary chosen frequency of say) 1 MHz as then 8 ns becomes a significant percentage of the analyzed and displayed frequency's period, resulting in a visible and annoying smear of the waveform.... For lower frequencies the jitter is just less or not at all visible.
But Wim's  method of syncing the scopes works much better.

[marmad]

[pa3bca]

Hi Marmad,
Thanks for this interesting video. I was just struggling with an I2C problem, and learned a few things how to optimise use of the segmented memory (bummer it will not decode in analyze mode, maybe in the next FW update?). Maybe I should fire up the hex editor.
Looking forward to seeing this functionality in RUU.

[marmad]

Thanks for this interesting video. I was just struggling with an I2C problem, and learned a few things how to optimise use of the segmented memory (bummer it will not decode in analyze mode, maybe in the next FW update?). Maybe I should fire up the hex editor.
Looking forward to seeing this functionality in RUU.

You're welcome, pa3bca.

I think not-decoding in Analyze is a bug of some sort - the DSO seems to want to decode the data, but gets stuck on the 1st frame. But it might be my settings as well - I haven't spent much time playing around with different sample lengths (which would change the granularity of the data), so maybe a different size would yield different results. For now, I've just been making a mental note (or writing down) the match frames - then leaving Analyze and returing to normal Playback (where the decode works fine).

[marmad]

Thanks Mark
That's a good trick to search recorded frames of data
It is great that you will incorporate it in a future version of RUU.

Here is a tip to use with manual Hex editing; I think the hex editor will fill a selected data block with a given byte pattern. That is fill 240 bytes with  x' 21 3C E1 FF' repeated to fill the block, for example. That is what I used to patch Rigol waveform files. I hope that will help until RUU.
Thanks again.

You're welcome, Len.

Thanks for the tip with the hex editor - of course, it's so obvious     It's been a long time since I used one and forgot the basic functionality 

[NikWing]

hey guys

could anyone please help me out (I hope one or all of these problems weren't asked yet, it would mean my search skillz suck)

kinda annoying and I don't know why this happens all of the sudden: I power off my DS2072(2302) normally and pull the plug. Now, whenever I put it back in the next day, the DSO resets itself (like it does after self cal for example)
how can I stop it doing that? (and why does it reset all settings after self cal, it's so annoying, too)
(I don't believe the backup battery only lasts 6 months or something like that ...)

the 2nd problem is about measuring. I try to measure/calibrate an ECG circuit, setting it to 5 mV/DIV. If I don't  use high res, it's really impossible to measure anything.
(I even tried avg@32 to be able to see a line on the DSO)
increasing the time base to something like 10 ... 100 ms/DIV makes the DSO totally slow, no matter what sample memory I select (the lesser the memory, the weaker the visibility of the curve on the screen)
I still remember my old HP DSO which was nothing compared to the Rigol and I was able to measure and display what I wanted to.
What do I have to do to get a nearly clear curve on screen?
I can't see an ECG curve on the Rigol because it refreshes so slowly :/

thanks so much for any help!

[Teneyes]

kinda annoying and I don't know why this happens all of the sudden: I power off my DS2072(2302) normally and pull the plug. Now, whenever I put it back in the next day, the DSO resets itself (like it does after self cal for example)

Nikwing. , did you set power on to 'Last' and NOT. To 'Default'. In Util menu?

[Teneyes]

the 2nd problem is about measuring. I try to measure/calibrate an ECG circuit, setting it to 5 mV/DIV. If I don't  use high res, it's really impossible to measure

You may wish to set intensity higher

Try to set trigger mode to normal and on each heart beat (trigger to large part)

Try recording and segmented memory

[Teneyes]

@Nikwing
Is this what you like to see?  Here is 70 BPM for 70 Seconds.

[Teneyes]

@Nikwing

You may wish to try Roll mode at slow scan rate (10s/div)
then stop the scanning
and change Time Base faster to see more detail (100ms/div),
and scan though the stop display data for more detail
See : pics

[Teneyes]

@Nikwing
You may wish to try triggering on each Pulse and recording the frames
Here is an Animating '.Gif' of 60 pulses at 120BPM
The playback is at 5 frames per sec so that is =  2.5 times the actual pulses rate

[Teneyes]

@Nikwing
At this setting you can record 8128 Pulses

[NikWing]

wow Teneyes

well, at first I try to measure just 5 mV to see if the gain stages work (using a 1:1 probe)
the curve is so "fat" that it's hard to see it move when I apply 5 mV to the circuit
I'm not yet to the point where I see a nicely looking heart beat like you show in your examples

I tried both, "default" and "last" in the settings, it happened with both (more than 1 try for both settings)
I'll give it another try

[marmad]

I tried both, "default" and "last" in the settings, it happened with both (more than 1 try for both settings)
I'll give it another try

Are you using FW v.3? If so, the new firmware has the new option 'Open' in the 'Power On' settings - although I have no idea exactly what it means. But you might try upgrading anyway, and see if it solves your problem.

[Teneyes]

@Nikwing
  You may wish to try Limit BW to 20MHz to limit the noise.
and I think you will need an Amplifier to boost the 5 mv

Note to All: my displays of Cardiac pulses are from a Rigol DG4000 , using 16000Pts Arb Function (built in). All displays captured easily with Marmad's RUU (a Plug)

[Purevector]

I think I may have found a bug.  Since upgrading to 3.00.01.03, whenever I plug in the scope, it automatically powers up.  Before it just used to fade the power button.  The power mode in the settings is Default, not Open.  I did try both.  Going back to V2 did not fix the issue.  When I turn the scope off and unplug it it takes a few seconds for the power led to completely turn off.  If I then plug the unit in, it boots up on its own.

[Galaxyrise]

the 2nd problem is about measuring. I try to measure/calibrate an ECG circuit, setting it to 5 mV/DIV. If I don't  use high res, it's really impossible to measure anything.

  You shouldn't have a "fat" trace at 5mV/div (it should be about 1/10 a division.) If you unplug everything from the scope, is the trace still fat? If not, then you're letting noise in from somewhere and should adjust your measurement setup.  It's pretty easy to pick up 5mV of noise.

  How large is the signal you're looking for? In general,  if you're trying to measure anything smaller than ~2mV you should either be measuring something slow enough that you can use hires (I've captured a 100uVpp ripple this way) or you should use a preamplifier (and everything being amplified has to be super low noise.)

increasing the time base to something like 10 ... 100 ms/DIV makes the DSO totally slow, no matter what sample memory I select (the lesser the memory, the weaker the visibility of the curve on the screen)
I still remember my old HP DSO which was nothing compared to the Rigol and I was able to measure and display what I wanted to.
What do I have to do to get a nearly clear curve on screen?
I can't see an ECG curve on the Rigol because it refreshes so slowly :/

100ms/div means 1.4s per capture.  Has nothing to do with memory depth!  You have a couple of avenues to make your life easier here, but these time bases are always a bit clumsy because making adjustments (like vertical offset) clears the screen.

1) As others suggested, use roll mode.  This is what I usually do for my slow measurements (and I've had it all the way to 1000s/div!)  Hires is really good in roll mode.
2) Set the trigger point to exactly the left side of the screen. 

In normal y-t mode, the scope needs to have captured enough data to fill in from the beginning of the waveform to the trigger before it can actually trigger.  So if your trigger is in the center of the screen, and you're at 100ms/div, then the scope needs to have 700ms of data before it can arm the trigger.  By moving the trigger point to the left of the screen, the trigger can arm "immediately" after the previous waveform is captured.  When the time base is slow enough, the scope will even display the in-progress waveform, making it feel a lot more responsive.

[jlmoon]

Hello,
I know this might be a bit off topic, since we're discussing the 2000 series scopes,

Has anyone noticed any vertical amplifier offset (no signal applied with 50 ohm termination on input AC or DC coupling selected) in the range of 100 - 200 uV when in the 500uV / div range.  I also noticed that channel 1 has a bit more than channel 2.  My particular model is a DS2202, are we splitting hairs with a Chinese scope playing in the grass?

[marmad]

1) As others suggested, use roll mode.  This is what I usually do for my slow measurements (and I've had it all the way to 1000s/div!)  Hires is really good in roll mode.

Peak detect is also very handy in roll mode.

2) Set the trigger point to exactly the left side of the screen. 

In normal y-t mode, the scope needs to have captured enough data to fill in from the beginning of the waveform to the trigger before it can actually trigger.  So if your trigger is in the center of the screen, and you're at 100ms/div, then the scope needs to have 700ms of data before it can arm the trigger.  By moving the trigger point to the left of the screen, the trigger can arm "immediately" after the previous waveform is captured.  When the time base is slow enough, the scope will even display the in-progress waveform, making it feel a lot more responsive.

This isn't precisely correct (although it works this way in many circumstances). In Y-T mode, the DSO captures the pre-trigger buffer data from the beginning of sample memory to the set trigger position - but sample memory doesn't always correspond to screen memory (what you see on the display). That's why there's the bar above the screen center - to show you how much of sample memory you're seeing - and your trigger position within it. In the attached screenshot, I have my trigger position set to screen left as you suggest - but it would still take 7 seconds before the buffer is filled (and I would start to see the trace). So you have to keep an eye on the location of your trigger position within the current sample memory depth - not just the screen memory.

[marmad]

I know this might be a bit off topic, since we're discussing the 2000 series scopes,

Not off topic at all - questions like these are precisely the reason for the thread. 

Has anyone noticed any vertical amplifier offset (no signal applied with 50 ohm termination on input AC or DC coupling selected) in the range of 100 - 200 uV when in the 500uV / div range.  I also noticed that channel 1 has a bit more than channel 2.  My particular model is a DS2202, are we splitting hairs with a Chinese scope playing in the grass?

Yes, I also have some offset at 500uV, but it's within the DSOs specs: ±0.1 div ± 2 mV ± 1% offset value

Keep in mind, 500uV is just a magnification of 1mV. So any offset at 1mV will just be doubled.

[jlmoon]

Okay  Marmad,

Thank you for your input, I had not referred to the the specs as of yet.  At 1mV & >  all is good.

[Galaxyrise]

In the attached screenshot, I have my trigger position set to screen left as you suggest - but it would still take 7 seconds before the buffer is filled (and I would start to see the trace). So you have to keep an eye on the location of your trigger position within the current sample memory depth - not just the screen memory.

Good catch.  To get the effect I was after, that means also choosing a memory depth so that the waveform is the entire screen.  That's what I get for giving usage advice away from my unit!  As another usability caution, you can actually set the trigger position to be far to the left of the waveform.  You want the "T" chevron to still be pointed down, not to the left.

Another tip that just occurred to me for measuring low frequency stuff like heartbeats is to turn on the bandwidth limit for that channel.  That'll probably cut the trace thickness in half.

[Purevector]

I think I may have found a bug.  Since upgrading to 3.00.01.03, whenever I plug in the scope, it automatically powers up.  Before it just used to fade the power button.  The power mode in the settings is Default, not Open.  I did try both.  Going back to V2 did not fix the issue.  When I turn the scope off and unplug it it takes a few seconds for the power led to completely turn off.  If I then plug the unit in, it boots up on its own.

I am pretty sure the newest firmware does not change the scope back to regular power on mode if it is changed to "open" mode. Has anyone tried this? Is there some way to report this to Rigol. The official 3.00 version doesn't have this setting so I am hesitate of simply writing to Rigol NA support.

Also, if there are any smart hackers out there that can figure out how to change the setting back, I would greatly appreciate it.

Cheers.

[Purevector]

I went back to FW 00.01.00.05 and main power connect still starts my DS2072
I use SCPI command to get to LAST

That setting controls the setup information at startup; that's not what I'm talking about.  In 3.01 there is a new setting called PowerStatus (you can see if if you scroll to the next page)  The options are Open or Default.  According the newest DS2000 manual on the Rigol website, Open mode powers up the scope when you plug it in, and Default just "energizes" the unit and you have to press the power button to power on the scope.  I am stuck in Open mode after testing that setting and cannot get back to Default.  I can change the setting and it appears to save the value, but the scope still powers up by itself.

We just had a power failure last night and the scope was on this morning because of this.  I'll have to keep it unplugged to prevent it running if power is lost