Rigol's New DHO800 Oscilloscope unbox & teardown

[asmi]

If you're concerned about signal integrity problems, an LA is an especially blunt tool.

Unless you thought about debugging in advance and added some kind of inline high-speed connector, SI is always going to be a problem. I don't know about you, but in my experience in 99% cases I had to use flying-leads, hookup wires and similar debugging "devices" which obviously only work for slower signals. Which is kind of my point - in most practical cases I've encountered I simply could not utilize high sample rates as mere act of probing destroys signal to the point that what I see on the screen and what actually happens on the line when the probe is off are totally different things. So even if LA had > 9000 GS/s, it would be just as useful as another LA which can do only 500 MS/s.

[tv84]

Modern PC-based LA have UNLIMITED triggering capabilities because they can stream data to software in real time. The only limitation is software, but that is an easy problem to solve.

With such certainties I guess you have no problems... IMO with concurrent systems that doesn't seem so linear. Just my 2 cents...

[Warhawk]

if they cannot deal with IPA to clean LCD, cross stitch is a better hobby. but yeah, large thumbprint is an issue causing inaccurate clicking..

FYI to my knowledge, there are plastic materials (e.g. Acrylic glass - PMMA) that should not be cleaned by any kind of alcohol as they cause damage.

FYI2: Touch-enabled buttons can be a problem for people with dry skin or people that work manually. I work on house renovations and have my palms and fingers like true working class (and I say this proudly).

[Warhawk]

Modern PC-based LA have UNLIMITED triggering capabilities because they can stream data to software in real time. The only limitation is software, but that is an easy problem to solve.

Triggering and streaming data real time are two different things. I am not sure if devices like Saleae do the trigger in software.

I've debugged communication issues in systems when the problem happened few times an hour or even less. Additionally, we had to monitor voltage rails to find the culprit. Nevertheless, I assume that once you work with complex systems you use a different test equipment.

[nctnico]

Modern PC-based LA have UNLIMITED triggering capabilities because they can stream data to software in real time. The only limitation is software, but that is an easy problem to solve.

Triggering and streaming data real time are two different things. I am not sure if devices like Saleae do the trigger in software.

I've debugged communication issues in systems when the problem happened few times an hour or even less. Additionally, we had to monitor voltage rails to find the culprit. Nevertheless, I assume that once you work with complex systems you use a different test equipment.

I agree. In most cases I use a logic analyser in a capture - analysis workflow. An MSO is just less work to setup and can show signals in realtime so that is typically my first choice when debugging digital systems. But if you are hunting a really illusive bug that happens seldomly (like once in an hour) there just isn't enough space on a disk to capture at high rates (like 200Ms/s). Not to mention the time it takes to sift through the data (99.99999% of the data will be useless). In such cases the wonderfully elaborate trigger system of a high end logic analyser together with record-on-change or segmented recording are good tools to have around. I always say: I rarely use my Tektronix logic analyser, but when I do it just pays itself back once more.

[tautech]

I think the "Default" button below the "Single" (triggering) button is going to be the most useless button for me.

Yes, I have wondered what Rigol were thinking there. A bit like placing a "Reset" button right in the middle of a computer keyboard...

You have to confirm that's what you really want to do, so... 

Maybe they are heavily targeting education with these models and wanted a quick way for the lab instructor to sort out a messed-up scope? (Or to restart the student exercise "and now set everything up properly for your use case!"  )

Yep. Or in any environment where 'scopes are shared and you have no idea what state the previous user left it in.

It's probably very useful for telephone support, too.

nb. Other 'scopes have that button too (in green so it stands out and invites people to press it):

User configurable too or can be left as Factory Default.

In addition these have 5 recallable User Setups so with a minimum of fuss and bother you can quickly have the scope set for specific tasks.

[Mortymore]

Many scopes have a "Default" button (reconfigurable), even if under other description, like the Tektronix 5 Series B MSO Mixed Signal 8 Channel Oscilloscope "Default Setup" button, or the R&S®MXO 4 "Preset" that can be configured to set either factory defaults or a user-defined preset configuration.

And this are not educational or hobby oscilloscopes.

[2N3055]

Many scopes have a "Default" button (reconfigurable), even if under other description, like the Tektronix 5 Series B MSO Mixed Signal 8 Channel Oscilloscope "Default Setup" button, or the R&S®MXO 4 "Preset" that can be configured to set either factory defaults or a user-defined preset configuration.

And this are not educational or hobby oscilloscopes.

Actually, as scopes get more complicated with time Default button is sometimes fastest way to start measurements from well defined starting point.
Especially if user can define what default is.

[Howardlong]

My original hypothesis that the front end attenuates in line with Nyquist to avoid aliasing was incorrect. It doesn't attenuate as the sample rate decreases.

I'd originally derived this from measuring rise time - fail on my part!

Today I hooked up an Agilent E4433B RF signal generator, set it to 200MHz 10dBm with a 50 ohm through termination at the scope.

Here are the results, 2ch, 3ch, 4ch and 4ch+LA. Vrms doesn't change by any significant amount.

Note the differences between Freq and Counter measurements when the LA is turned on. I thought the counter measurement might always give the correct answer, but no.

[ebastler]

Actually, as scopes get more complicated with time Default button is sometimes fastest way to start measurements from well defined starting point.
Especially if user can define what default is.

Alright, I'll make my peace with the Default button on the DHO 800/900 then. 

Although I still think a top-level menu entry on the touch screen would have sufficed, and Rigol could have given us a Normal/Auto trigger toggle or a dedicated Zoom button instead of the Default button. There's limited real estate on that small control panel...

[Howardlong]

Nothing beats real-time data streaming, no segmented memory not any triggering is going to replace grep. Or Python. Or C#. Or <insert your favourite data analysis language>.

I respectfully disagree. As I've said before, it's horses for courses, use the tool for the job that gets you there the quickest considering one's individual experience and skill set.

If you find frequently find yourself having to hack some code to do your search, I don't consider that to be either "real-time", or a particularly efficient way of working, especially when there's a trigger setup immediately available to you.

For the avoidance of doubt, occasionally I do hack some code to do exactly what you're suggesting on a large dataset extracted from an LA, but it is rare, because most of the time I can use the various triggers available to me to achieve the same goal.

Furthermore, IME near-unlimited streaming of bulk data into a PC from LAs is typically restrictive in terms of sample rate, unless you have a particularly expensive LA. While a reasonably-priced LA might be OK for oodles of I2C traffic, high speed SPI not so much.

[Howardlong]

...but in my experience in 99% cases I had to use flying-leads, hookup wires

What "flying leads" are you using and why are you probing with "hookup wires" in this case? Typically if there isn't a testpoint, I solder a header pin as close as I can get it, certainly within millimeters if not directly to the device's pin itself.

The pods that are supplied with decent scope LAs enjoy low capacitance, adjacent grounds, and use tricks like distributed resistance and damping/compensation to minimise their loading effects.

Remember, we are talking about relative timing here such as setup and hold.

... So even if LA had > 9000 GS/s, it would be just as useful as another LA which can do only 500 MS/s.

Respectfully, you're heading into the world hyperbolic projection now.

While I fully accept that it's never going to be as good as having multiple multi-GHz active probes and a boat anchor scope to match, properly designed digital passive probes can and do have a place in resolving sub nanosecond timings.

I encourage you to read this article on the subject. https://www.tek.com/en/documents/application-note/magnivu-technology-provides-500-ps-timing-resolution-0

Since that was written, the same technology has been pushed down to 60ps sampling period that I'm aware of, and I use it on the Tek MDO4104C on the bench. While I'm no fan of the UI on Tek scopes, their scopes do offer some features unavailable on competing scopes at the same level. While I'm only showing a single channel here, it will do 60.6ps resolution on all 16 channels, albeit at only 10kpts in this mode.

[Mechatrommer]

My original hypothesis that the front end attenuates in line with Nyquist to avoid aliasing was incorrect. It doesn't attenuate as the sample rate decreases.

of course it doesnt, its been like that since ds1000e and ds1000z, you dont have to activate all channels. enough with one channel, dial to large enough timescale, sample rate will be reduced, if signal frequency is high enough, you'll see aliasing, if you dont be carefull, you can register a wrong frequency on report paper, we learnt the hard way, in case of suspicion, dial back the time scale to the smallest where sample rate is highest, if you see suddenly signal jump back to higher frequency while dialing down, you experienced aliasing. same thing when you are driven nuts wondering why you cant get stable trigger however you try changing mode and level. try to satisfy nyquist by auto attenuation will not avoid you from confusion anyway imho. at large enough timescale, when you put your dut signal with hi freq, you may not see anything and got into confusion thinking you dont have a signal, but infact it is there, its just highly attenuated in scope. ymmv, fwiw.

[thm_w]

Nothing beats real-time data streaming, no segmented memory not any triggering is going to replace grep. Or Python. Or C#. Or <insert your favourite data analysis language>.

I respectfully disagree. As I've said before, it's horses for courses, use the tool for the job that gets you there the quickest considering one's individual experience and skill set.
***
Furthermore, IME near-unlimited streaming of bulk data into a PC from LAs is typically restrictive in terms of sample rate, unless you have a particularly expensive LA. While a reasonably-priced LA might be OK for oodles of I2C traffic, high speed SPI not so much.

Yes but in the context of this thread, we are talking about the DHO900. We have other threads you can argue about PC LA vs MSO that are more relevant.

You are paying $200 (PLUS adapter $50-300) for a LA that:
- takes away from your analog sample rate
- will have limited protocol support compared to sigrok

IMO that ~$300 is better spent toward a known good PC LA or dedicated AWG or something else. Maybe in the future that might change, if rigol puts a lot of effort into the LA (which they have not done in the past).
99% of hobbyist use cases can get away with using a few analog channels for I2C or SPI.

[rpro]

Even though it is a good value for the beginner, the people migrating from say their basic 2ch CRT scope to something more digital, and people requiring a 2nd, portable scope for occasional work battery powered, but - all the users here reviewing the scope are the advanced users - and it could easily be after a month or two many of them would consider this specific series a flop..

On my DHO804, I did the 924 vendor.bin file override @Azusa did early on, at the beginning of this thread, and I am not seeing any significant offsets anywhere, with very good noise figures (<24uV 50ohm term.) and rise-times (1.30 ns), using the 00.01.00 2023/07/21 build the scope came with. (I believe the 800 ps rise-time that @Azusa had reported earlier, may have been bypassing the LC filters, as he had done and mentioned in another one of his posts.)

In my opinion, despite its first-release faults and limitations, the DHO804 is an overall good value as a low-noise scope complement to the much more capable, but noisy, MSO5000. 

Edit: Added color-graded pic of Ch1 (50ohm terminated). 

[2N3055]

My original hypothesis that the front end attenuates in line with Nyquist to avoid aliasing was incorrect. It doesn't attenuate as the sample rate decreases.

of course it doesnt, its been like that since ds1000e and ds1000z, you dont have to activate all channels. enough with one channel, dial to large enough timescale, sample rate will be reduced, if signal frequency is high enough, you'll see aliasing, if you dont be carefull, you can register a wrong frequency on report paper, we learnt the hard way, in case of suspicion, dial back the time scale to the smallest where sample rate is highest, if you see suddenly signal jump back to higher frequency while dialing down, you experienced aliasing. same thing when you are driven nuts wondering why you cant get stable trigger however you try changing mode and level. try to satisfy nyquist by auto attenuation will not avoid you from confusion anyway imho. at large enough timescale, when you put your dut signal with hi freq, you may not see anything and got into confusion thinking you dont have a signal, but infact it is there, its just highly attenuated in scope. ymmv, fwiw.

This is the problem. With 4 channels enabled on 200 MHz DHO900 you cannot satisfy Nyquist at any timebase or memory setting. If you enable digital, you have sampling enough for a 50Mhz scope...

DHO800 with 100MHz BW would be OK at first glance, but unlike DS1000Z that had 130 MHz -3dB point, DHO800 has it at higher (almost 180 MHz) and is also 12 Bit meaning it will alias on signal levels you would not have to worry with 8 bt scope...

But we could argue DHO800 should be OK.
OTOH DHO900 is not... You cannot have 200MHz scope with less than 500MSp/s and even that is on the limit..

[UK]

But we could argue DHO800 should be OK.
OTOH DHO900 is not... You cannot have 200MHz scope with less than 500MSp/s and even that is on the limit..

Do you mean buying the DHO924S model is not a good idea?! DHO914S will be a better choice if I need an LA option?

[2N3055]

But we could argue DHO800 should be OK.
OTOH DHO900 is not... You cannot have 200MHz scope with less than 500MSp/s and even that is on the limit..

Do you mean buying the DHO924S model is not a good idea?! DHO914S will be a better choice if I need an LA option?

Look here:

https://www.eevblog.com/forum/testgear/rigols-new-dho800-oscilloscope-unbox-teardown/msg5119131/#msg5119131

With 4 ch on and LA enabled it will drop to 156.25MSa/s   and  1Mpts of memory...
If you need to do digital work stick with Rigol MSO5000 or Siglent SDS2000X+ if you want both digital and analog.
If you are doing mostly digital, 12 bit will not be so important.

[Mechatrommer]

This is the problem. With 4 channels enabled on 200 MHz DHO900 you cannot satisfy Nyquist at any timebase or memory setting. If you enable digital, you have sampling enough for a 50Mhz scope...

obviously DHO800/900S is not the right tool for you. you are asking too much. you need to step up to > $5K scope range.

OTOH DHO900 is not... You cannot have 200MHz scope with less than 500MSp/s and even that is on the limit..

for some people yes it is... trying to probe at its limit with 2 channels at 2.5X sampling rate is possible. if its not acceptable, probe with only 1 channel (1.25GSa/s) should be more comfortable 5X sampling rate... then again if you need 2 channel at > 1GSa/s... DHO800/900 is not for you.. fwiw.. (ps: why this kind of basic discussion has to happen sooo many times? )

[TurboTom]

My original hypothesis that the front end attenuates in line with Nyquist to avoid aliasing was incorrect. It doesn't attenuate as the sample rate decreases.

I'd originally derived this from measuring rise time - fail on my part!

Today I hooked up an Agilent E4433B RF signal generator, set it to 200MHz 10dBm with a 50 ohm through attenuator at the scope.

Here are the results, 1ch, 2ch, 3ch, 4ch and 4ch+LA. Vrms doesn't change by any significant amount.

Note the differences between Freq and Counter measurements when the LA is turned on. I thought the counter measurement might always give the correct answer, but no.

Howard - thanks a lot for pointing this out. A few days ago, I made similar findings here regarding high frequency aliasing, albeit without plugging in and considering the effect the digital probe may have (actually, I assumed it wouldn't change analog sample rate since it's not been documented). The counter discrepancy that you observed between the so-called "Hardware Frequency Counter" and the "measurement window counter" when enabling the digital channels is quite interesting and may give us a hint where the "bottleneck" is. May I suggest to perform the same test with the digital channels while only one and two analog channels are active to see if this discrepancy between the two counters is present as well (obviously, with high enough input frequency to cause aliasing)?

My take on the shortcoming of the scope hardware is memory bandwidth! The "hardware" counter operates on the data stream from the ADC directly, before it gets sent to the sampling memory while the measurement window counter probably operates on the decimated shadow memory which resembles the sample memory (from a temporal point of view) if the time base is set fast enough.

What really surprises me about this is that there's two additional RAM footprints available that -- at least in the photos of the pre-release version of the DHO900 -- had been found to be populated. Assuming that these additional memory chips provide storage for digital input sample data and waveform data for the AWG, I wouldn't expect this to slow down the analog channels. An FPGA should be able to process all this independently in parallel, as long as it's chosen with a sufficient amount of logic cells... So we actually may be up to a surprise if we look into a recently produced DHO900 scope -- the additional RAM foorprints may still be unpopulated.

If it's due to the lack of FPGA resources, Rigol will have to accept the charge of having skimped on the choice of the Zynq version that they used on the DHO900. Whatever the reason for this observation is (I mean the further reduction of sample rate when enabling the digital channels), I consider this a major flaw that most probably cannot be corrected with a firmware update. This was the last "nail in the coffin" that finally triggered my decision to return the DHO914S after the Bode Plot disaster.

I may get a DHO804 at a later time since there are many quite nice solutions in this scope as long as it's used purely as an "analog" scope without the additional functions that the DHO900 series is supposed to provide. The U/I is quite fast compared to Rigol's legacy instruments, the FFT is lightning-fast and provides quite decent performance (IMO), and the over-all appearance makes it a decent, modern instument to "carry around" in the lab / workshop.

What didn't seem to get better during the ten days that I had to "play" with the DHO914S is the "baked electronics smell" that it emanates and that may actually be quite distracting when evaluating a new electronics design, not knowing where the "burning" smell comes from. In a home lab, maybe even in a living area, I'ld consider theis smell an absolute no-go. Rigol should definitely address their PCB manufacturing/flux cleaning processes to eliminate this problem.

"My" DHO900 is back in its box and will get returned these days. It's been an interesting experience but also one that cost me way too much time for an unfinished product that in it's current configuration (IMO) is not worth spending the time and the money on.  At least as long as none of use is getting paid by Rigol to do their work!

[Howardlong]

Many scopes have a "Default" button (reconfigurable), even if under other description, like the Tektronix 5 Series B MSO Mixed Signal 8 Channel Oscilloscope "Default Setup" button, or the R&S®MXO 4 "Preset" that can be configured to set either factory defaults or a user-defined preset configuration.

And this are not educational or hobby oscilloscopes.

Actually, as scopes get more complicated with time Default button is sometimes fastest way to start measurements from well defined starting point.
Especially if user can define what default is.

Agreed, the default button is one of the most commonly used on the front panel IME. The difference with the more expensive scopes is that they have probe readback and sensible default LA thresholds as part of the default setup.

Having said that, one non-default setting I frequently set after hitting default is the horizontal trigger position to the left 10% and expansion around the trigger point, but that's much less fiddly than having to reset four probe ratios.

I can get used to restoring from a stored config, but having a quick button user preset would be useful. I think the Siglents do offer that, but I haven't had one on the bench for a while.

[2N3055]

obviously DHO800/900S is not the right tool for you. you are asking too much. you need to step up to > $5K scope range.
(ps: why this kind of basic discussion has to happen sooo many times? )

I don't know ? Maybe if people would not repeat wrong statements all the time that need to be corrected.

Again:

For DHO800 i would accept it is OK. It is not that critcal

But DHO900 : 3 or 4 channels enabled  and digital on : 156.25MSa/s     / 1Mpts memory

On a scope that has 200 MHz BW on inputs feeding the ADC...

Again: With 4 channels enabled on 200 MHz DHO900 you cannot satisfy Nyquist at any timebase or memory setting. If you enable digital, you have sampling enough for a 50Mhz scope...

DHO900 with that sampling should be a 60-70 MHz realtime scope. Only 70 MHz version should be sold. Or they should have made progressive filters that will limit BW as sample rate drops. That would make it slightly dishonest (not really selling aa 200 MHz scope if it delivers it only in special occasions) but would insure that signal would be shown properly (with limitations of BW). And for low price that would be worth of compromise.

They didn't do that, and how the decided to do it will create problems all the time. And funny stuff, especially for their target market of hoby users and tinkerers. People that have experience could work around because they know how. But also with experience comes realization that it's not worth it... You just buy what actually works. DHO800 has good price and does not violate basic principles. BUt DHO900 is not a good, honest product.

It is funny to me how much coping is there.. It is wrong and should be fixed. And funny enough it can probably be fixed with filters in channels that are already there. It probably just needs fix in software..

In attachments 1MHZ squarewave. Input channel limited to 200MHz. Fixed sampling at 1GS/s, 400MS/s, 200MS/s, 100MS/s.
Representative of SLOW digital clock

In attachments 16MHz squarewave. Input channel limited to 200MHz. Fixed sampling at 1GS/s, 400MS/s, 200MS/s, 100MS/s.
Representative of digital clock on Arduino UNO (ATMega328)

[2N3055]

Many scopes have a "Default" button (reconfigurable), even if under other description, like the Tektronix 5 Series B MSO Mixed Signal 8 Channel Oscilloscope "Default Setup" button, or the R&S®MXO 4 "Preset" that can be configured to set either factory defaults or a user-defined preset configuration.

And this are not educational or hobby oscilloscopes.

Actually, as scopes get more complicated with time Default button is sometimes fastest way to start measurements from well defined starting point.
Especially if user can define what default is.

Agreed, the default button is one of the most commonly used on the front panel IME. The difference with the more expensive scopes is that they have probe readback and sensible default LA thresholds as part of the default setup.

Having said that, one non-default setting I frequently set after hitting default is the horizontal trigger position to the left 10% and expansion around the trigger point, but that's much less fiddly than having to reset four probe ratios.

I can get used to restoring from a stored config, but having a quick button user preset would be useful. I think the Siglents do offer that, but I haven't had one on the bench for a while.

Yes, that is my gripe with 3000T, that position can't be set in default.
And yes, Siglents have full config of default. In fact you can have dozens of favorite setups and recall them as needed.
And then you choose which one gets called when default is pressed.
If you need full reset, factory reset is in maintenance manu.

Best,

[iMo]

A high time for Rigol to open the 800/900 series such it becomes the first Rigol's "HW platform" for an "open source" software development.

Rigol shall publish the necessary information such the broad community of the talented developers may start to fix and develop new stuff. It is pretty obvious Rigol does not have enough resources for the sw development in their low cost DSO segment.

Rigol may still let some proprietary details closed, sure, but playing with, for example, UI, Bode, wifi drivers, keyboards, decoders, LA, etc., etc. could be left open..
That would be the game changer, imho..

[Howardlong]

But we could argue DHO800 should be OK.
OTOH DHO900 is not... You cannot have 200MHz scope with less than 500MSp/s and even that is on the limit..

Do you mean buying the DHO924S model is not a good idea?! DHO914S will be a better choice if I need an LA option?

With the current firmware, if you need an LA option, I can't recommend the either. Having said that, most of the problems are firmware, but there are so many issues I fear it'll take months rather than weeks to resolve.

The analogue bandwidth thing is a separate issue. When you need it, you need it. The choice is either to pay the $100 or hack it. If you pay $100 you also get the four higher bandwidth probes... not that you can enjoy more than two at the same time at that bandwidth of course!