Rigol's New DHO800 Oscilloscope unbox & teardown

[Warhawk]

The integrated bode plot is actually a very useful feature. I use it sometimes with R&S scope and it works well. It is not for RF stuff but definitely great for audio or loop stability investigation.
My next scope has to support this. Either externally or with the in-built AWG.

[Mortymore]

A scope that I could integrate with my AWG is in the shop list, for the Bode plot option that has to be in the audio region.
A nanoVNA is also in consideration. But acquiring certain things takes it's time

This new scopes (at least DHO1000 and 4000) that do not integrate an AWG, should be capable of using an external AWG to enable Bode Plot option, as Siglent does in many scopes

[svetlov]

Owon is releasing new oscilloscopes on Android - there is also a 14-bit model - with a generator (FRA) - with a built-in power supply
https://www.owon.com.hk/products_owon_fds1000_series_multifunction_dso

 

[Nikki Smith]

Have you done equal length differential signal processing? The signals I understand may have frequencies of tens of megabytes

My first attempt doesn't length-match the differential signals. I looked at previous clone boards and no one bothered! The popular design by Sergey Petrukhin for example has up to 3mm difference within a differential pair.

But Dave's teardown video clearly shows Rigol made an effort with the logic analyser signals on the scope motherboard, so I'm updating my PCB design to length-match to (much) less than 0.1mm difference. It should be good enough for the 200MHz max bandwidth that the (official) Rigol logic analyser is rated for?

[Fungus]

Hi,
I have managed to install the key, got the 4th option in the menu (BODE PLOT) but I can't find the BODE option in the menu.

Maybe it checks for the signal generator hardware.

[Fungus]

But even for the difference beterr 914 and 914S it's just such a meh AWG (single channel, 25mhz)...

Especially if you got the 804. Then your remaining budgetary coins might have left over even yet for something like nanoVNA also (to do the bode plots with).

The 914S will do audio frequencies. The NanoVNA won't.

[dreamcat4]

But even for the difference beterr 914 and 914S it's just such a meh AWG (single channel, 25mhz)...

Especially if you got the 804. Then your remaining budgetary coins might have left over even yet for something like nanoVNA also (to do the bode plots with).

The 914S will do audio frequencies. The NanoVNA won't.

You misquoted my comment by entirely removing the line in-between. My point was that the price jump is so much that you could afford both a standalone AWG and then also have even room left for something like a nanoVNA. That was to double highlight the relatively poor value. But if you remove that line you are misrepresenting the intentions there... yes standalone AWG vs integrated is really the thing here that is what you are paying for. And maybe I don't value that as much as other people do seem to here (what actually is the relative hike to the 914S? $400 vs $700?).

[dreamcat4]

OK so let's take a step back and re-examine the fundamental nature of the problem here...

Nobody really wants to add themselves back in the 914S's awg into an 800 series. Because it's such an undertaking, a lot of components. The complexity and for $300 of a person's time (at least as some individual / standalone 1-off project). It's just generally considered by a lot of us as an unattractive prospect. Compared to paying over some dough and getting a product with a build in AWG. OK agreed then.

Problem is bode plot isn't enabled. Ok then. But the scope is digitally controlling the AWG, checking if it's responding (present). And then telling it to do it's sweep etc. for whatever built-in functions. To get the body plot.

So how about instead of re-creating fully ourselves the entire original 'meh' (as i would put it) 25mhz 1ch AWG. Why don't we just put in a simple hardware interface. To fool the scope into thinking the AWG is present. And make a low level digital interface that then sends those commands to an external standalone AWG? For example the one I mentioned earlier was something like a UNI-T UTG962E (which is 2ch, 60mhz).

Then we would still not have such an integrated all-in-1 solution. But we could then upgrade our 804s by buying a BETTER awg, and then basically the scope has become a bit like a remote control for it? Could this be a realistic prospect? Because if yes then... I'm definately all up from that.

There still might be a few people who need to use these features on battery powered for mobile / field work applications. That having it all neatly inside the scope is more attractive... those people still would not be served by this type of a solution so well. And might prefer to pay those extra monies for the Rigol 914S instead of the cheaper 804.

However for everybody else it's a cheaper / more incremental step-by-step upgrade path to follow. That is then assuming it's not an infeasible concept that's not going to work. Probably needs more investigation and looking into. But also needs some example 914S. To work from.

[RAPo]

If we can run python on the oscilloscope we can make a bode plot like described
here
with a generator that accepts scpi commands 😊

[UK]

Just found out some details about Rigol RT1642IQ used in frontend of DHO series... if someone interested

[dreamcat4]

well yes, there is the choice of hacking the firmware alone, to trigger some actions that then communicates over one of the general purpose digital interfaces on the scope (whether that is via the ethernet, or usb, or something else). the advantage of that is then should not require any hardware mods / hacking whatsoever. which gives itself a potential to keep a viable product warranty.

then there's the idea of minimally hacking the pcb awg section there, to install some low level mcu digital interface within the scope. to then relay the commands (like you say, perhaps scpi, but whatever the specific choice is...). this approach then requires some degree of hardware hacking. but then you don't need to fight the firmware. because you are fooling the mostly-standard 900 series firmware (and/or the android app). or it is being altered more minimally.

then there's the prospect of doing either of those above things. but more heavily hacking (for example) the menu choices for the built in capabilities to fool the scope even further. to unlock awg above the built in 25mhz. all the ways up to whatever higher specs your external awg actually supports.

for taking something like a bode, my understanding is that it's however many data points gets plotted. that the awg is given an instruction for each data point. then there is some short delays for the signal to settle before the scope actually takes a snapshot of the measurement. before moving on... so then any extra processing burdens (within the scope's running software, to be sending these commands)... it should not negatively affect the measurements?

rather any hardware side mods on the awg section would need to be carefully considered for digital noise within the scope. which might or maybe not be needed after all. depending whether or not this is feasibly done all over the existing built in communications interfaces. or rather which route has the fewer obstacles overall.

[Fungus]

standalone AWG vs integrated is really the thing here

That assumes the 'scope can use an external AWG, which I don't think it can.

(at least, there's nothing in the manual).

Will it be hackable? Possibly. We're getting into big hacks now though.

[Martin72]

That assumes the 'scope can use an external AWG, which I don't think it can.

I don't believe that either, at least I'm not aware of that with other Rigol models.
If they had integrated Bodeplot, then always when there is also a built-in AWG.
That you can also use external AWGs would only be a finger exercise in the software - But rigol and software...

[Fungus]

That you can also use external AWGs would only be a finger exercise in the software - But rigol and software...

Maybe in that APK there's a function "setAWGFreq();" that can be replaced. 

[Dacian]

I picked up my Rigol DHO804 from UPS a few hours ago. This is from the Aliexpress seller.

I had to play for about 1h so this is the first impression with me mostly testing the 12bit ADC resolution.
Did not even connected anything to the input it still has the plastic caps.

Bad part is the nasty plastic smell hopefully that will go away but is very strong.

Here is what I got in the box (seems it has the Lenovo power supply). And the plug looks like Australian version with a US/Canada adapter.


The hardware and software version


Settings I done to save the waveform in .csv format and from memory not screen.


First I set the channel 1 to 50mV/div and 20ms/div stop the acquisition and then zoomed on screen 100x to 0.5mV/div
As it can be seen 12bit data was captured thus still some points at 100x zoom as many as expected


Then I reduced the vertical resolution to 500uV and reacquired the signal to see if is a true 500uV at 12bit and yes it is as you can see from the csv if you want to download the step is about 1.1uV as expected with 4mV per screen (8div * 500uV) and 4mV / 4096 = 1uV step.


download the 500uV csv file from here https://electrodacus.com/RIGOL/RigolDS3.csv
download the 50mV csv file from here https://electrodacus.com/RIGOL/RigolDS2.csv


I have a DS1054z if anyone wants any sort of comparison. But there is no discussion that DHO804 is superior on any aspect. The user interface is super responsive no comparison with DS1054z. I think the fan is not as noisy as some made it look and not sure but I think DS1054z is noisier by quite some margin.
Form factor is excellent also. Touchscreen is useful and responsive but buttons are also good.

[gslick]

I picked up my Rigol DHO804 from UPS a few hours ago.

Now that you have a unit to try out, one minor thing I was curious about is if there is any support for a USB keyboard? The user guide mentions USB mouse support, but I don't see any mention of USB keyboard support. If you plug in a USB keyboard, does it do anything when the onscreen virtual keypad is displayed, for example to rename the waveform labels from the default CH1, CH2, etc? Do the arrow keys do anything to scroll or zoom the waveform display, or move cursors, or do anything else?

[Dacian]

Now that you have a unit to try out, one minor thing I was curious about is if there is any support for a USB keyboard? The user guide mentions USB mouse support, but I don't see any mention of USB keyboard support. If you plug in a USB keyboard, does it do anything when the onscreen virtual keypad is displayed, for example to rename the waveform labels from the default CH1, CH2, etc? Do the arrow keys do anything to scroll or zoom the waveform display, or move cursors, or do anything else?

I just tested using my Logitech MX mini keyboard and mouse combo so a single receiver and while the mouse works the keyboard will not do anything in any menu.

[Fungus]

the keyboard will not do anything in any menu.

Not really surprising.

[EEVblog]

Was shooting the review video and this happened, WTF!
EDIT:  It's NOT the DHO800 at fault! Follow-up video shot on the 2nd channel.

[EEVblog]

Bad part is the nasty plastic smell hopefully that will go away but is very strong.

There is an operational "hot" smell. It's a bit disconcerting.

Then I reduced the vertical resolution to 500uV and reacquired the signal to see if is a true 500uV at 12bit and yes it is as you can see from the csv if you want to download the step is about 1.1uV as expected with 4mV per screen (8div * 500uV) and 4mV / 4096 = 1uV step.

Rigol specifically said the 500uV range is software magnified 1mV, so 

[Martin72]

Even the much more expensive DHO4000 does not have a "real" 500microvolt, rather few scopes on the market do.

[Dacian]

Rigol specifically said the 500uV range is software magnified 1mV, so 

If you look at the .csv file the steps are about 1.1uV for the waveform captured while at 500uV/div
In any case 1.1uV is not that useful due to noise floor but I think it is there and they have a real 500uV range.
A strange thing I notice is that I can have 25Mbps of memory only on channel 1 (not one channel but only specifically channel 1). The max is just 10Mbps if I have only channel 2 enabled, same for 3 and 4.

[Dacian]

Even the much more expensive DHO4000 does not have a "real" 500microvolt, rather few scopes on the market do.

If it was just 1mV and then digital zoom I will have expected to see about 2uV steps in the csv file and not 1.1uV.
It is a bit strange that is 1.1uV and not 0.98uV but is much closer to 1uV than 2uV.
It is not really useful and likely the reason they did not implemented that in the DHO4000 even the 1mV is a bit of a stretch (it is OK on 8bit oscilloscopes but you can not take advantage of 12bit at 500uV and 1mV unless you want to measure the scope noise).

Here I used math function to be able to have a 20x vertical zoom on the CH1
I also experimented with the inverse color screenshot.

[EEVblog]

Pricing has changed!

Before:


Now:

[Dacian]

Did you change the trigger to the active channel as well?

Peter

I just tested now and yes I have 25Mpts available if I enable the trigger to the single enabled channel.
Not sure why that will be needed as I can stop the acquisition without using a trigger especially if it is a repetitive signal.
In any case is good to know. Is that similar on other oscilloscopes ?