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

[marmad]

Can the firmware be backed up just in case some mysterious power outage takes out the scope for no apparent reason?

[tlu]

That was just an inquiry, my dso is still perfectly fine. Just curious as to if it is even possible to back up the firmware as a safety precaution. Just wondering....

[marmad]

That was just an inquiry, my dso is still perfectly fine. Just curious as to if it is even possible to back up the firmware as a safety precaution. Just wondering....

Since firmware does not fail because of power outages (unless, perhaps, one happens at the exact moment you're upgrading the firmware) - and since many manufacturers generally don't want people fiddling with their firmware - and since one of the reasons manufacturers supply warranties is as a safety precaution against device failure - I'd guess that, no, there is no method (or at least no publicly available method) to copy the firmware from the DSO to a PC.

[tlu]

Since firmware does not fail because of power outages (unless, perhaps, one happens at the exact moment you're upgrading the firmware) - and since many manufacturers generally don't want people fiddling with their firmware - and since one of the reasons manufacturers supply warranties is as a safety precaution against device failure - I'd guess that, no, there is no method (or at least no publicly available method) to copy the firmware from the DSO to a PC.

Thanks marmad, I figure it will be some time before someone can dump the firmware from this dso. I recalled in Dave's tear down of this scope, there are jtag pins available so maybe in the future this is an option for dumping the firmware? Anyhow, the previous inquiry was like you said, what if power went down during a firmware update and your warranty is expired? It would be nice at that point to have something to load back onto the dso.

I noticed on channel 2 of my dso I'm getting 10mV of ripple noise as opposed to channel 1 which is about half that. Is this normal or maybe there is grounding issue for channel 2? This is done with the respected channel in question and nothing hooked up to the probe.

[marmad]

I noticed on channel 2 of my dso I'm getting 10mV of ripple noise as opposed to channel 1 which is about half that. Is this normal or maybe there is grounding issue for channel 2? This is done with the respected channel in question and nothing hooked up to the probe.

A probe can act like an antenna. To check actual noise levels of the DSO, unhook the probes. Also, posting a screen cap here (try to use .png format to save Dave's BW) would help in identifying problems.

Attached are pngs of noise levels of my scope at the 500uV vertical setting:
No BW limit: ~640 uVpp average
20M BW limit: ~176 uVpp average

[EV]

A probe can act like an antenna. To check actual noise levels of the DSO, unhook the probes. Also, posting a screen cap here (try to use .png format to save Dave's BW) would help in identifying problems.

Attached are pngs of noise levels of my scope at the 500uV vertical setting:
No BW limit: ~640 uVpp average
20M BW limit: ~176 uVpp average

I also checked the average noise values and got:

680/660 uVpp with no BW limit and horisontal scale at 20 ns
180/170 uVpp with 20M BW limit and horisontal scale at 20 ns

1.27/1.21 mVpp with no BW limit and horisontal scale at 10 ms
597/560 uVpp with 20M BW limit and horisontal scale at 10 ms

It looks that the noise value depends on the horisontal scale value possition.

[tlu]

Thanks again for the info marmad. You are absolutely right about the probe acting like an antenna. After removing it and checked again. Normal ~400uV and with 20Mhz BW limit ~160uV. I did not wanted to post a picture just to save Dave's BW just for this. In the future, I will with other issues so debugging and locating discrepancies would be easier.

[tlu]

I got so excited when receiving and playing around with the scope to check its labeled spec that I've did a self cal and lost the trial options.... . I'm mad at myself as I really wanted to try out the i2c decoding on a small project. I'm kind of bummed for doing that...

[marmad]

It looks that the noise value depends on the horisontal scale value possition.

It's not the noise level - it's the measurements themselves. Most lower cost scopes (including the Agilent X series) take measurements from the displayed data - not the full-resolution captured data. Like analog scopes, measurement is most accurate when the waveform fills the screen without overshoot, and you have at least 3-4 cycles on screen.  If you play with the vertical or horizontal scale while keeping auto measurement on, you can watch the measurements change dynamically, starting with the LSD of the values.

[tlu]

It looks that the noise value depends on the horisontal scale value possition.

It's not the noise level - it's the measurements themselves. Most lower cost scopes (including the Agilent X series) take measurements from the displayed data - not the full-resolution captured data. Like analog scopes, measurement is most accurate when the waveform fills the screen without overshoot, and you have at least 3-4 cycles on screen.  If you play with the vertical or horizontal scale while keeping auto measurement on, you can watch the measurements change dynamically, starting with the LSD of the values.

Taking overshoot into account, the measurement reported from the dso is the average of the all the captured display data? How does it do the calculation for the reported Vpp, does it take a fixed sample size from the displayed data and average that?

[Sparky]

I got so excited when receiving and playing around with the scope to check its labeled spec that I've did a self cal and lost the trial options.... . I'm mad at myself as I really wanted to try out the i2c decoding on a small project. I'm kind of bummed for doing that...

This is one of the known bugs with the current firmware.  It has been mentioned (by marmad and perhaps others) that you can request a new set of trial license codes for the optional modules from your distributor.

[marmad]

Taking overshoot into account, the measurement reported from the dso is the average of the all the captured display data? How does it do the calculation for the reported Vpp, does it take a fixed sample size from the displayed data and average that?

I'm not sure what you're asking. The measurements are made on the displayed data, which is limited to 700 x 400 points - not the entire sample data.

You can see this effect clearly by setting your sample depth to 14Mpts, then feeding a fixed waveform into the scope and adjusting the vertical and horizontal scales until 2 or 3 cycles fill the screen. Then turn on the Vpp measurement. Then change to the smallest timebase scale (i.e. flattening the waveform) and you will see the Vpp measurement flatten correspondingly. Then STOP the scope and increase the horizontal scale - you will see the entire captured waveform reappear (from stored sample memory) and the Vpp measurements will again increase as the waveform fills the screen.

[drieg]

Since firmware does not fail because of power outages (unless, perhaps, one happens at the exact moment you're upgrading the firmware) - and since many manufacturers generally don't want people fiddling with their firmware - and since one of the reasons manufacturers supply warranties is as a safety precaution against device failure - I'd guess that, no, there is no method (or at least no publicly available method) to copy the firmware from the DSO to a PC.

...Anyhow, the previous inquiry was like you said, what if power went down during a firmware update and your warranty is expired? It would be nice at that point to have something to load back onto the dso.

DS2000 has a bootloader (unlike DS1000E), so in case of e.g. power failure during firmware update, the unit may not start again, but you are still able to re-initiate FW update by pressing HELP button during boot process.

In addition to that, importand unit specific data is backed up in another part of memory, so in case they gets corrupted or overwritten, they are restored automatically.  I'd say there is enough protection implemented. Anyway, in case anybody get into trouble and need my help, feel free to contact me.

As I already mentioned in another post here, I suggest anybody with FW version 00.00.01.00.02 to upgrade first to 00.00.01.00.05 or later.

[tlu]

I'm not sure what you're asking. The measurements are made on the displayed data, which is limited to 700 x 400 points - not the entire sample data.

You can see this effect clearly by setting your sample depth to 14Mpts, then feeding a fixed waveform into the scope and adjusting the vertical and horizontal scales until 2 or 3 cycles fill the screen. Then turn on the Vpp measurement. Then change to the smallest timebase scale (i.e. flattening the waveform) and you will see the Vpp measurement flatten correspondingly. Then STOP the scope and increase the horizontal scale - you will see the entire captured waveform reappear (from stored sample memory) and the Vpp measurements will again increase as the waveform fills the screen.

Sorry I've misunderstood initially but now I understand what you meant by the display data. Thanks for the detailed info and clarifying things up for me.

[tlu]

DS2000 has a bootloader (unlike DS1000E), so in case of e.g. power failure during firmware update, the unit may not start again, but you are still able to initiate another FW update by pressing HELP button during boot process.

In addition to that, importand unit specific data is backed up in another part of memory, so in case they gets corrupted or overwritten, they are restored automatically.  I'd say there is enough protection implemented. Anyway, in case anybody get into trouble and need my help, feel free to contact me.

As I already mentioned in another post here, I suggest anybody with FW version 00.00.01.00.02 to upgrade first to 00.00.01.00.05 or later.

Great information there drieg and thank you for pointing this out to me. I believe I'm on firmware 00.00.01. I haven't checked the Rigol site for newer firmware yet but it seems they are out since you mentioned the two firmware update. I have to look into this.

[tlu]

Just double checked and the software version is 00.00.01 and hardware version is 1.0.

[drieg]

Check full/long firmware version  00.00.01.xx.xx as described here.

[paolo]

... take measurements from the displayed data - not the full-resolution captured data.

Aha! That explains a few things that were confusing me initially.

...I received mine at the weekend. My first scope, and initially thought - oh dear, I've overbought, should have bought something cheaper/lower end as a starter. But I'm getting much more comfortable with it now.

I'm not the person to give an opinion on tech performance, but it's ticking the boxes that made me think it was the right scope for me - UI and screen size were the main factors. Especially UI. Chinese products generally can struggle a bit in that area. The videos of the Owon I'd seen - despite being on a nice big screen - well, the UI seemed to be all over the place. The Rigol on the other hand makes good use of lots of soft buttons, which minimises sub menus and all that.

The screen is also nice. TBF it's still a far cry from other more general tech. When we're seeing 'retina' displays etc becoming common on sub $500 consumer devices, this is an area where scope manufacturers (one and all) seem to be a few years behind where it's at. But, relative to other scopes at this price, it's pretty good I think, and doesn't have that horrible Owon 'margin' space that chops off a whole load of usable area.

(Dreaming... One day budget scope designers will offer HDMI out, and let us plug in our cheap 720p or 1080p displays. The built in screen will be considered a basic integral screen for scenarios when an external display isn't practical or available. Perhaps there won't even be a built in LCD on some models. For the hi-res out of course they'll need more video memory and more CPU/GPU horsepower to plot those muuuch nicer resolutions, but that's not insurmountable.)

[T4P]

Of course they are common 500$ devices, they have 10-20$ chips that do everything in the phone ... vs 100$ alone for the ADC
AND a proper industrial LCD panel, not a crap one that starts going wobbly in 2-3 years

[marmad]

...I received mine at the weekend. My first scope, and initially thought - oh dear, I've overbought, should have bought something cheaper/lower end as a starter. But I'm getting much more comfortable with it now.

Check that you have the latest firmware version 01.00.05 (using the instructions provided above and in other threads).

The screen is also nice. TBF it's still a far cry from other more general tech. When we're seeing 'retina' displays etc becoming common on sub $500 consumer devices, this is an area where scope manufacturers (one and all) seem to be a few years behind where it's at. But, relative to other scopes at this price, it's pretty good I think, and doesn't have that horrible Owon 'margin' space that chops off a whole load of usable area.

You have to realize that every extra pixel displayed horizontally (and to a lesser extent, vertically) is a major challenge for DSO manufacturers - it's always a tradeoff between resolution, waveform update rate, and blind time - that's why the screen resolutions lag behind other digital devices that cost similar amounts of money. This article gives you an introduction to the idea: http://www.eetimes.com/ContentEETimes/Documents/Schweber/C0906/C0906edited.pdf

For example, the DS2072 displays 700 pixels horizontally for the waveform. At the 20ns timebase setting, with 14k sample memory, it's processing 700MB of data per second (@ 50k wfrm/s) to those 700 pixels - each one representing ~560ps of time, with an intensity level indicating frequency of occurrence.

Edit: Sorry, my math above was not correct - I wrote it when I first woke up  - although the principal of increasing the resolution also increasing the blind time remains. But the math is as follows:

The DS2072 displays a MAXIMUM of 700 points horizontally for the waveform. At the 20ns timebase setting, with 14k sample memory, the scope is capturing the entire sample depth in 7us (500ps per sample at the 2GS/s rate), but it's only displaying 280ns (or 560 points) of the captured sample. So the DPO processing that takes place is happening on 28MB of data per second (560 * 50k) - likely the highest throughput that the scope can achieve. If you would increase the screen resolution to a higher number (e.g. 1024 pixels), the displayed points would increase - and the waveform update rate would have to drop (and the blind time increase) correspondingly in order to maintain the same throughput.

I think I've got it right this time 

[EV]

I suggest anybody with FW version 00.00.01.00.02 to upgrade first to 00.00.01.00.05 or later.

I have now also upgraded to this 00.00.01.00.05 version. No problems. Thanks for tips!

[Sparky]

Hi folks!

I finally received my DS2072 today.    I ordered from Tequipment.net, and they were out of stock, so it took forever to get, and in my case, it shipped direct from Rigol's warehouse.

I've only had about 30 mins with it(!), but my initial reaction was "NICEEEEE!"  The first thing that really stands out before turning it on is the quality (and weight!) --- way exceeds general expectation of China brand equipment.  Even before using, you feel this is a very well engineered equipment, and I really feel I'm going to enjoy using it.

I'm not sure if there's a way to find the manufacture date, but mine was calibrated on July 20.  I had a quick look at the software/hardware versions I got:

Software version: 00.00.01.00.02
Hardware version: 1.1.0.0
FPGA version:
SPU 03.01.02
WPU 00.06.00
CCU 12.29.00
MCU 00.05

So, the older 01.00.02 firmware is still shipping.  I'm going to contact Tequipment.net for the 01.00.05 firmware update before I use it much.  Most likely I'll get the update on Monday (as it's late Friday here).  I'll post back here with the outcome and more impressions!

Overall, very happy and can't wait to get using it!

[marmad]

Good to hear you finally got it - it is definitely a pleasure to use so you should enjoy yourself   Remember, because of bugs in 1.00.02, it's best if you avoid doing anything that writes to internal memory (self-cal, saving references waveforms or setups, etc.) until after you've updated.

All of the EEVBlog members zibadun, branadic, and EV have copies of the 1.00.05 firmware - you probably can get it from one of them. Just remember, because of the bugs, it's important you do the update during the boot process - and not from the menus - as detailed by drieg here: https://www.eevblog.com/forum/blog-specific/eevblog-369-rigol-ds2000-oscilloscope-playing-around/msg153700/#msg153700

[Sparky]

Thanks marmed!  I contacted a few of our great forum members, and have the firmware to update.  But I have some sad news regarding the update...

I followed the instructions from drieg and tried many times to start the update process during boot, but it would not enter "update mode."  Always, the system would complete boot, and then show a message of "new firmware detected, and select OK/CANCEL to do the update.  I select CANCEL, and power OFF, and ON again.

I tried about 6 times to start the update from boot; I pressed "HELP" at various stages during boot: immediately after power on before the logo is shown, and on other attempts when the logo is shown.  Sometimes I pressed "HELP" briefly, other times I held it longer, and sometimes I pressed it multiple times during when the logo is shown.  I never seemed able to enter the update process.

After many attempts, I think may be the only option is to select OK soft-button option from the GUI...so eventually I did.

The update started, and proceeded to 60%, but now it has stopped there. 

It has stopped at 60% for at least 15 mins now. 

One member suggest I start over, but I'm not sure the system will recover if I remove power during the boot process.  So, it is still plugged in.

I'm not sure what to do...I'm quite worried 

Any suggestions?

[Sparky]

Well, it is booting now...but not without loss of trial options...

After 1 hour, no progress beyond 60%, so I powered it off.  I powered it on, and unsurprisingly, it did not boot.

Despite failed firmware update, if the boot loader were okay, then I assumed it may be possible to perform an update via the boot process, if I could get it to work.  So (in desperation and with no alternative option!) I tried again to press "HELP" button after power ON during boot.  I tried a couple of times, and again no luck, but then it worked!!!  I think it is worth stressing that HELP must be pressed immediately after ON button, and well before RIGOL logo appears.  It seems that the timing of pressing "HELP" button is very specific, so it may take several attempts to do it right.  If you find that the system boots normally despite pressing HELP button, just turn it off and back on, and try again!

This time, update continued with CH1 flashing, and eventually all LED were on.  I powered OFF, remove USB flash drive, and power ON.  System booted, okay, and System Info says:

Software version: 00.00.01.00.05
Hardware version: 1.0.1.0
FPGA version:
SPU 03.01.02
WPU 00.06.00
CCU 12.29.00
MCU 00.05

So, software version is updated, and hardware version has decreased! (hardware version was 1.1.0.0 before the update).

So, I'm lucky it is working now, but unfortunately I've lost all my trial options.  The "Installed" soft-button is grayed out under UTILITY->OPTIONS menu.  I hope I can get the trial options back from a new trial license from Rigol.

Also, I'm worried what other system information (calibration, etc.) might have been lost from the failed system update.  Does anyone know what are consequences of failed update on system info?

Edit: Emphasis added regarding timing of pressing HELP button.
Edit 2: Improved readability and added more emphasis!