Hacking the Rigol DHO800/900 Scope

[Aleksandr]

By the way, this is the memory I have on 804.

[ebastler]

Has anyone tried these Puccy 3 Pack Screen Protector TPU Film Guard (Not Tempered Glass) for the touchscreen?
If so, what are the pros and cons of them? Does it cut down on the screen reflection?

The description says "gloss finish", so it probably is (at least) as reflective as the original. It's also plastic film, not glass at all. Good thing they give you three of these so you can replace them when they are scratched again...

[norbert.kiszka]

By the way, this is the memory I have on 804.

On 924S I have different ones. All populated (4 GB visible for CPU). Other ones for CPU and other ones for FPGA.

Edit: I dont know if I uploaded photo around CPU, so here it is:

[norbert.kiszka]

If somebody will need values and placement of resistors (924S), I measured most of it around CPU (photos atached).

[Mechatrommer]

Maybe Im stupid and proper termination is not a problem?

LC filter has its own input and output Zc, maybe that 85ohm you mentioned earlier is its termination? removing the LC filter changes Zc, then you mentioned again 50ohm, how you make to that conclusion? i havent plugin values trace width and distance and gnd plane height into saturn pcb toolkit yet to know what those diff strips' Zc, so btw i cant find any "dho800/900 400 MHz hack" in this forum, that russian forum doesnt show any conclusive method anyway. so your posts seem out of magic for me.

slightest presence of anything more than 625MHz will create false signal on screen

Are you trying to suggest LC filters are kind of magic and not making false signals? Phase shift in pulses are nothing to worry about? Lowering amplitude of high frequency is also never not a problem? Anything between samples will be still on screen - more or less. Im aware of it.

what is the cutoff frequency by shorting those inductors pads?

Only one problem for me here is the noise (as I mentioned earlier) - I will check eventual contamination (making higher capacitance somewhere in the analog path) - if its not the case, then I will bring back second stage filter, which I suspect is around 800 MHz.

either you misunderstand me, or me misunderstand you, i think. please proceed

[norbert.kiszka]

Maybe Im stupid and proper termination is not a problem?

LC filter has its own input and output Zc, maybe that 85ohm you mentioned earlier is its termination? removing the LC filter changes Zc, then you mentioned again 50ohm, how you make to that conclusion? i havent plugin values trace width and distance and gnd plane height into saturn pcb toolkit yet to know what those diff strips' Zc, so btw i cant find any "dho800/900 400 MHz hack" in this forum, that russian forum doesnt show any conclusive method anyway. so your posts seem out of magic for me.

Same impedance must be for DC. That explains why they added capacitor before a filters instead of giving lower value for resistors - with removing coils, this cap (before filter) should be also removed to match same impedance - is that clear or not? Also, resistors just after input stage are 2x 35Ω (70Ω in total). Thats why I removed this cap before filters to match same impedance. 0.1Ω difference is not a problem (measured those resistor before ADC are 85.2 84.9 85.3 85.3).

What You have on mind with 50Ω ? Maybe You misread what I told earlier of what I expected when I was measuring those resistors.

slightest presence of anything more than 625MHz will create false signal on screen

Are you trying to suggest LC filters are kind of magic and not making false signals? Phase shift in pulses are nothing to worry about? Lowering amplitude of high frequency is also never not a problem? Anything between samples will be still on screen - more or less. Im aware of it.

what is the cutoff frequency by shorting those inductors pads?

Much higher than before. More than 1 GHz. Input stage is also a filter - if You like it or not.

Do You have any sort of problem with my modification? I dont see any distortions after this change. Only more noise, as I mentioned three times. Im not Your child, so please dont force me waht I should do with my scope and what I shouldnt. I feel like Im talking with Arduino-child which forces everybody else to put Arduino everywhere (including simple inverter based on single transistor).

either you misunderstand me, or me misunderstand you, i think. please proceed

That didnt sound nice.

[AceyTech]

By the way, this is the memory I have on 804.

Thank you.  That's the same as in my 804, purchased Black Friday from AliExpress -CN.   I think they ran out of the GigaDevice part in November., as a lot of the photos of newer builds show the Samsung -BYMA parts.

[AceyTech]

either you misunderstand me, or me misunderstand you, i think. please proceed

That didnt sound nice.

Wow man., are you seriously so tired of insulting everyone else, that now you're insulting yourself?

[Mechatrommer]

Much higher than before. More than 1 GHz. Input stage is also a filter - if You like it or not.

good! try feeding the scope with ringy square wave, say 100MHz square (since you already have 0.8-1GHz scope) nasty signals, not a good one, say with highly mismatched reflected signal, highly overshooted etc, just a simulation of quite badly designed circuit/dut. show me the gibbs! cheers.

[norbert.kiszka]

Much higher than before. More than 1 GHz. Input stage is also a filter - if You like it or not.

good! try feeding the scope with ringy square wave, say 100MHz square (since you already have 0.8-1GHz scope) nasty signals, not a good one, say with highly mismatched reflected signal, highly overshooted etc, just a simulation of quite badly designed circuit/dut. show me the gibbs! cheers.

Currently I will not put housing back to do Your request. I already compared signal between channels and I didnt see any "nasty" signals, reflections or overshots. Please keep Your imagination to yourself.

[ebastler]

I feel strangely reminded of people modding their old Philips CD players -- disabling DAC oversampling and removing all output lowpass filters. They used to swear by the perceived sound improvement,  while others were swearing at them...

[norbert.kiszka]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

[norbert.kiszka]

while others were swearing at them...

Some people thinks that if a some engineer made something in some way, then completely nobody can change that, because engineer decided to use only possible way and everybody changing design are stupid. So if a some stock radio in a car is X, then I shouldnt change it to other model, because some engineer made best possible decision?

[Mechatrommer]

Much higher than before. More than 1 GHz. Input stage is also a filter - if You like it or not.

good! try feeding the scope with ringy square wave, say 100MHz square (since you already have 0.8-1GHz scope) nasty signals, not a good one, say with highly mismatched reflected signal, highly overshooted etc, just a simulation of quite badly designed circuit/dut. show me the gibbs! cheers.

Currently I will not put housing back to do Your request. I already compared signal between channels and I didnt see any "nasty" signals, reflections or overshots. Please keep Your imagination to yourself.

sorry if i sound "complaining", i'm not... i'm just "asking" or "proposing"... as i said earlier, no compulsion... but we maybe share the same urge.. cheers.

Earlier I told here to measure those resistors. Everybody else assumed those are 0 Ω resistors... I wasnt complaining, but watching everybody destroying their scopes. Next time dont assume anything, especially when somebody tells You opposite. Using multimeter can take couple seconds and can save expensive&precise equipment and save many hours of unnecessary work.

[norbert.kiszka]

Earlier I told here to measure those resistors. Everybody else assumed those are 0 Ω resistors... I wasnt complaining, but watching everybody destroying their scopes. Next time dont assume anything, especially when somebody tells You opposite. Using multimeter can take couple seconds and can save expensive&precise equipment and save many hours of unnecessary work.

When modding CH4 I didnt assume anything. As I posted before, I will repeat myself: I dont see anything wrong with this modded channel beside of higher noise.

[shapirus]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

Where is this signal coming from and how is it fed to the scope? It looks seriously screwed up, I mean all that strong ringing.

[norbert.kiszka]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

Where is this signal coming from and how is it fed to the scope? It looks seriously screwed up, I mean all that strong ringing.

Cheap and ugly resistors to match impedance and same thing with coax.

[ebastler]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

I would say that the signal we see is nicely bandwidth-limited to about 100 MHz, in both cases: We see pronounced ringing at the 5th harmonic of the 20 MHz fundamental, so there is not much frequency content above that. And I would eyeball the risetime at 3.0 to 3.5 ns, which is also consistent with not much more than 100 MHz.

Presumably your source already limits the spectral content. In which case one would expect to see neither a benefit, nor any problem, from the lack of an input lowpass filter.

How do things look with a signal which actually contains higher frequency components? Could you show the benefit? And what about aliasing from spectral components above the Nyquist limit?

[norbert.kiszka]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

I would say that the signal we see is nicely bandwidth-limited to about 100 MHz, in both cases: We see pronounced ringing at the 5th harmonic of the 20 MHz fundamental, so there is not much frequency content above that. And I would eyeball the risetime at 3.0 to 3.5 ns, which is also consistent with not much more than 100 MHz.

Presumably your source already limits the spectral content. In which case one would expect to see neither a benefit, nor any problem, from the lack of an input lowpass filter.

How do things look with a signal which actually contains higher frequency components? Could you show the benefit? And what about aliasing from spectral components above the Nyquist limit?

Signal coming from this stupid hand made generator is not bandwidth limited. Beside of overshot, voltages are as expected. Measured rise time is very close to what You told, exactly 2.3 ns.

Later I will make some faster generator, beside of fact I have one (faster than this).

[Mechatrommer]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

Where is this signal coming from and how is it fed to the scope? It looks seriously screwed up, I mean all that strong ringing.

16MHz (17MHz?) is some sort of arduino clock? 10Vpp source? if its 5Vpp, its unterminated.

[norbert.kiszka]

Ref from CH1 and modded CH4 with exact same settings (including sample rate) and same coax:

Where is this signal coming from and how is it fed to the scope? It looks seriously screwed up, I mean all that strong ringing.

16MHz (17MHz?) is some sort of arduino clock? 10Vpp source? if its 5Vpp, its unterminated.

17.7 MHz, 5Vpp terminated on both ends. In latest firmware there is a 2x probe selection.

[AceyTech]

If anyone is interested, I have attached two files. These are boot logs read from the UART connectors of the processor and FPGA.

Thanks for posting these.  Have you diff'd these to the original logs Dave posted that he took during the teardown video?  Has much changed?  I'll check in "the morning" when I get up.

No, I didn't compare. I didn't even look closely at what I recorded from my oscilloscope

Huh.  Interesting. 
Aside from the Garbage characters at the beginning and the occasional line break diff, not much changed in the FGPA console outputs.  Not totally surprising, but a tad bit disappointing that they haven't found something to fix.  Maybe their VHDL team is that good....?

On the RK side;   Even though dates/versions are same, there are many more white space diffs, and there seems to be several lines not matching up  -Altho', admittedly I'm just doing a quick compare using a subpar diff tool--Chrome extension POS.  Hard to tell what's missing, and not.

Did you edit/add anything in your bash scripts?
If you don't mind sharing; What is your model # and month built?

[AndyBig]

Did you edit/add anything in your bash scripts?

As far as I remember, I only changed the time zone setting in the oscilloscope application launch script

If you don't mind sharing; What is your model # and month built?

Model - DHO814. I don’t even know how to look at the month of release, but I ordered it at the end of November 2023
The firmware is the latest, 00.01.02.00.02

[AceyTech]

Did you edit/add anything in your bash scripts?

As far as I remember, I only changed the time zone setting in the oscilloscope application launch script

If you don't mind sharing; What is your model # and month built?

Model - DHO814. I don’t even know how to look at the month of release, but I ordered it at the end of November 2023
The firmware is the latest, 00.01.02.00.02

Ahh, Thanks.  This is from when mine was "new":  See build date.

[AndyBig]

Ahh, Thanks.  This is from when mine was "new":  See build date.
(Attachment Link)

Well, the build date is the build date of the application, it depends only on the version of the application Now I have a modified application in which this date is set to February 28, 2024
I don’t even remember what I had there from the factory It seems the same as what you have in the screenshot