Hacking the Rigol DHO800/900 Scope

[shapirus]

Isn't larger diff between the two mean cpu heat is not being sinked out to amb fast enough? cpu temp can be no lower than ambient temp, so isnt smaller diff better?

Given that it was with a larger fan (*and* a higher than a comfortable from the noise perspective air flow), I suppose it was because there was a higher air flow wherever the ambient sensor is located. May well be not even under the heat sink. I may be wrong though.

[Aleksandr]

Do all 900 series oscilloscopes have hardware number 8?

[norbert.kiszka]

Do all 900 series oscilloscopes have hardware number 8?

I can confirm 924S has (originally) HW number 8.

[Randy222]

Do all 900 series oscilloscopes have hardware number 8?

Funny you ask this question, because I came back to ask about the HW #

Do the HW #'s found in 800-900 series uniquely identify a 800 or 900 model? As example, why not use #'s 1,2,3,4,5,6,7,8 to distinguish between 8 different models found across 800/900 series?

Next question I have is a bit more "sinister". Does the HW # actually mean anything in these DHO's, given their current code set? If you make it HW-12 on an actual 914, does that remove scope feature? If you make it HW-8 on an 802 does that provide more features?

An easy test is, for any oem model (non-mod'd), turn off the hdcode KLM in the start script (see recent previous posts) and reboot. The device will reboot with HW-0. Does that impact the scope features in any way? My unproven theory is that rigol perhaps wanted to use hardware config to place certain boundaries on the device, but never really implememted it (for whatever reasons). If HW # means nothing in terms of actual usable features, then the chatter around HW # is a bit moot.

I changed my HW # a few times, 0, 8, 12, 200, and they all seem to change nothing on my 804, at least nothing that I can see. As example, on any 800 if you set it to HW-8, does that give you any banwidth past 100MHz ? We also know that HW-8 does not fully distinguish between 900's. It perhaps seems HW # just ID's the number of channels, so I will test that theory on my 804 using the HW # for an 802.

[ebastler]

Does the HW # actually mean anything in these DHO's, given their current code set? If you make it HW-12 on an actual 914, does that remove scope feature? If you make it HW-8 on an 802 does that provide more features?

I thought Mechatrommer had already confirmed that you need to change the HW ID from 12 to 8 in order to get digital triggers to work?

[Randy222]

Speaking of additional calibration options..  I hadn't noticed this before tonight.  There is a "Detail" drop-down on the SelfCal window with some interesting calibration info. -if it passes- 
(Attachment Link)
BTW: these are all the "additional calibration" options that I found to calibrate with success.  Anything else added will fail. 
(Attachment Link)
FYI: it took 7 hours to test all the permutations, because some would go 98 or 99% before stopping or crashing.

The cal items seen after x3 About press, is a bit confusing. Are the ones selected by default the same items used when you just have the std Cal menu of "start"? Or is the std set something different? Is that in the manual?
Rigol should at least identify which ones selected are in the std set of cal tests. And why does selecting the other tests fail? If they are tests not applicable to the device then they should be grayed out. If they are applicable tests and they fail, why do they fail?

[Randy222]

Does the HW # actually mean anything in these DHO's, given their current code set? If you make it HW-12 on an actual 914, does that remove scope feature? If you make it HW-8 on an 802 does that provide more features?

I thought Mechatrommer had already confirmed that you need to change the HW ID from 12 to 8 in order to get digital triggers to work?

That's it, just the one item? How does that work on an 800 model? Isn't that feature only for the digital analysis port found on the 900's ?
Are we talking about digi triiger selection D0-D15? If so then that's not applicable to 800's. However, I run my 804 as a 914 HW-12 and the digi panel then appears at bottom of screen.

900's have trigger type CAN and LIN, 800's do not. So 800 HW-8 turns on CAN and LIN ? I shall take a look.
800 datasheet shows they have some digi triggers, SPI and I2C are digital signals.

Edit: I just checked my 804 HW-12 running with 914 vendor bin and some lics, CAN and LIN are in my trigger menu.
Are we saying all the digi triggers don't actually work unless the device sees HW-8 ? Or just that CAN and LIN won't work unless there's a digi port installed?

[RAPo]

I'll second that (firmware 00.01.02).

Do all 900 series oscilloscopes have hardware number 8?

I can confirm 924S has (originally) HW number 8.

[norbert.kiszka]

Speaking of additional calibration options..  I hadn't noticed this before tonight.  There is a "Detail" drop-down on the SelfCal window with some interesting calibration info. -if it passes- 
(Attachment Link)
BTW: these are all the "additional calibration" options that I found to calibrate with success.  Anything else added will fail. 
(Attachment Link)
FYI: it took 7 hours to test all the permutations, because some would go 98 or 99% before stopping or crashing.

The cal items seen after x3 About press, is a bit confusing. Are the ones selected by default the same items used when you just have the std Cal menu of "start"? Or is the std set something different? Is that in the manual?
Rigol should at least identify which ones selected are in the std set of cal tests. And why does selecting the other tests fail? If they are tests not applicable to the device then they should be grayed out. If they are applicable tests and they fail, why do they fail?

Enabling "test model" by a taping/clicking 'about' three times is a "top secret" feature, not meant to be found by a customer.

[ebastler]

I thought Mechatrommer had already confirmed that you need to change the HW ID from 12 to 8 in order to get digital triggers to work?

That's it, just the one item? How does that work on an 800 model? Isn't that feature only for the digital analysis port found on the 900's ?

That's the one function we know about. And yes, it requires that you solder the connector onto the PCB (footprint is ready), cut a slot into the plastic housing, and connect an external logic probe (original Rigol or clone).

I would assume that other hardware ID bits or values have a meaning as well. They could encode existing hardware options, options reserved for future expansion, or different hardware versions in case Rigol make design changes to the board, e.g. due to parts obsolescence. (Again, these could be already existing versions or -- more likely since the product is new -- versions reserved for the future.)   

[ebastler]

900's have trigger type CAN and LIN, 800's do not. So 800 HW-8 turns on CAN and LIN ? I shall take a look.

Seems unlikely. I assume that the hardware ID only encodes differences in the actual hardware on the PCB. More likely the vendor ID controls which software options are enabled.

[Randy222]

900's have trigger type CAN and LIN, 800's do not. So 800 HW-8 turns on CAN and LIN ? I shall take a look.

Seems unlikely. I assume that the hardware ID only encodes differences in the actual hardware on the PCB. More likely the vendor ID controls which software options are enabled.

But 800's have I2C and SPI, it does not require the digi port for that.
Is CAN and LIN tied directly to the digi port and the scope can't use CAN or LIN trigger on BNC port?

Also, see my edit comment in my last post behind this one.

[ebastler]

But 800's have I2C and SPI, it does not require the digi port for that.
Is CAN and LIN tied directly to the digi port and the scope can't use CAN or LIN trigger on BNC port?

Also, see my edit comment in my last post behind this one.

Bit of a mix-up there. The HW ID 8 is required to enable triggering from the digital input channels. This is unrelated to decoding serial protocols (or triggering based on decoded protocol signals) captured through the analog channels.

The latter decoder and trigger options do not require specific hardware options on the PCB, and hence are most likely not controlled by the hardware ID. The choice which ones to make available in which models was a pure marketing decision on Rigol's part: Need to reserve some goodies for the more expensive 900 models...

[Randy222]

But 800's have I2C and SPI, it does not require the digi port for that.
Is CAN and LIN tied directly to the digi port and the scope can't use CAN or LIN trigger on BNC port?

Also, see my edit comment in my last post behind this one.

Bit of a mix-up there. The HW ID 8 is required to enable triggering from the digital input channels. This is unrelated to decoding serial protocols (or triggering based on decoded protocol signals) captured through the analog channels.

The latter decoder and trigger options do not require specific hardware options on the PCB, and hence are most likely not controlled by the hardware ID. The choice which ones to make available in which models was a pure marketing decision on Rigol's part: Need to reserve some goodies for the more expensive 900 models...

So, in summary, There are two HW #'s for 800's, 2ch and 4ch model. For 900's there's just HW-8, they are all 4ch w/ digi port. So, HW-8 only seems to indicate that the device has extra hardware (804 more than the 802, 900's are a opened 804 with digi port.)

So I perhaps conclude HW # is not really relevant unless one decides to add a digi port to an 800 model. I do not get any benefits running my 804 as HW-8. The extra CAN and LIN appears to come from running the 914 vendor.bin

HW-8 --> 4ch with digi port (with AFG?)
HW-12 --> 4ch with no digi port
HW-xx --> 2ch with front side EXT trigger and no digi port (I forget the HW # for the 2ch devices, and I am not sure front EXT is hardware or not).

All the 900's are HW-8 ? Does that mean they all have AFG but only S models have it activated? Isn't AFG more hardware, or is that done via FPGA or something?

[ebastler]

So I perhaps conclude HW # is not really relevant unless one decides to add a digi port to an 800 model. I do not get any benefits running my 804 as HW-8. The extra CAN and LIN appears to come from running the 914 vendor.bin

That's sounds right, at least as far as we know now. Maybe at some point we will figure out what magic is worked by the additional RAM in the DHO900...

But in any case, it seems very likely that changing the hardware ID only makes sense (today and in the future) if you also change the actual hardware. As long as your original DHO8xx PCB remains unchanged, there will probably be no benefit -- and possibly some detrimental side effects -- from changing the hardware ID.

[norbert.kiszka]

Three photos of underside of CPU in original & unmodified DHO924S. If somebody wants more photos of something particular - better tell me now.

 

[norbert.kiszka]

I measured resistors around CPU and noted it. Later I will do image of what is what. Values was 10k, 5k, 4.7k, 22, 1.

[S2084]

Three photos of underside of CPU in original & unmodified DHO924S. If somebody wants more photos of something particular - better tell me now.

 

Could you take a photo of the side of the input tract?(Its analog part)If possible, on both sides of the board.  Thank you!

[Randy222]

Is the Sa/s rate bound to the ADC? If so can the ADC actually do better than 1.25 Sa/s ?
I notice via logcat that RIGOL.SCOPE : [drv_scope.cpp] [ConfigAquire] [2619] gets sample rate as ":WavSa :SampleRate".
Maybe hackable via software?

[AceyTech]

So, in summary, There are two HW #'s for 800's, 2ch and 4ch model. For 900's there's just HW-8, they are all 4ch w/ digi port. So, HW-8 only seems to indicate that the device has extra hardware (804 more than the 802, 900's are a opened 804 with digi port.)

So I perhaps conclude HW # is not really relevant unless one decides to add a digi port to an 800 model. I do not get any benefits running my 804 as HW-8. The extra CAN and LIN appears to come from running the 914 vendor.bin

HW-8 --> 4ch with digi port (with AFG?)
HW-12 --> 4ch with no digi port
HW-xx --> 2ch with front side EXT trigger and no digi port (I forget the HW # for the 2ch devices, and I am not sure front EXT is hardware or not).

All the 900's are HW-8 ? Does that mean they all have AFG but only S models have it activated? Isn't AFG more hardware, or is that done via FPGA or something?

Good summation.  To Add:

802 has the same AFE chip as a 804, but the AFE is missing on CH3, and there are a couple "settings" resistors on the top side, near the Ch4 AFE to further differentiate between 802 and any other model.

AFG is more hardware, in the form of some opAmps and power supplies for trigger levels and to power the AFG and LA pods on the main PCBA.  The AFG (digital) data is generated in real-time by the FPGA, then passed to the DAC daughter PCB -typically-only found on S models.

@Mech had to set the HW number because he was experiencing some triggering issues.  It might still be worth looking into a software based bypass, but it seems to me there are other things that could be worked on

[ebastler]

Is the Sa/s rate bound to the ADC? If so can the ADC actually do better than 1.25 Sa/s ?
I notice via logcat that RIGOL.SCOPE : [drv_scope.cpp] [ConfigAquire] [2619] gets sample rate as ":WavSa :SampleRate".
Maybe hackable via software?

It's been in this thread (and a couple of others) more than once. Last time was yesterday, I believe. Do you actually follow this thread, or do you only read the answers to your specific questions? 

The ADC could quite possibly do 2 GSa/s, since it is apparently the same as in the DHO1000 and 4000. However, the FPGA is probably the bottleneck. TurboTom has posted datasheet evidence which makes it plausible that 1.25 GSa/s is its limit.

[norbert.kiszka]

Three photos of underside of CPU in original & unmodified DHO924S. If somebody wants more photos of something particular - better tell me now.

 

Could you take a photo of the side of the input tract?(Its analog part)

Input stages? I already did. First things first. Now Im removing LC filter from channel 4 and adding missing capacitors around dc-dc converters - some of them wasnt populated (one instead of two parallel).

[S2084]

Three photos of underside of CPU in original & unmodified DHO924S. If somebody wants more photos of something particular - better tell me now.

 

Could you take a photo of the side of the input tract?(Its analog part)

Input stages? I already did. First things first. Now Im removing LC filter from channel 4 and adding missing capacitors around dc-dc converters - some of them wasnt populated (one instead of two parallel).

Ścieżka wejścia analogowego

[norbert.kiszka]

Shaky hands and naked eye.

Edit: In this photo I didnt removed this capacitor before LC filters which was a quite big mistake. It should be also removed, to match transmission line impedance.

[Mechatrommer]

So, in summary, There are two HW #'s for 800's, 2ch and 4ch model.

the HW version for all series have been laid out by souldevelop, how he came to the conclusion we dont know, probably from his software rev engineering skill, now he is missing so he cant answer, but from evidences posted in forums, his conclusion is so far correct... except S2084 discovered new HW version extension from resistor mod such as HW4,5,13.. what are they? who cares?

I do not get any benefits running my 804 as HW-8. The extra CAN and LIN appears to come from running the 914 vendor.bin

nobody forces you to change to HW8 if vendor.bin hack provides you with everything you need. some even dont want to upgrade to 900 because they hate the digital and afg tab under the GUI. and for companies they dont even want to touch any of rigol_hack with 10" barge pole due to warranty reason. should i sorry for them? heck no because i understand what i need is probably what they hate. free choice for their own benefit and interest and how much risk they want to take.

All the 900's are HW-8 ? Does that mean they all have AFG but only S models have it activated? Isn't AFG more hardware, or is that done via FPGA or something?

AFG is commanded by FPGA and vendor.bin hack to 900 alone will activate it even on HW12, so far no differences that i can spot. bode plot was working fine in HW12. btw checking start_rigol_app.sh i spotted there is another afg_gpio.ko module is loaded if anyone interested, so far i am not interested in wasting time on it since every AFG functionalities that i expect to work, work as expected either on HW12 or 8. the point we looked for HW8 is because we want the 16CH digital channels can be used as trigger source correctly. this has some applications, but if you dont need it, thats fine for you.

point is, we only report and share how to get the hack done in case readers came to find for the solution, same thing other people reported possible but nonsensical HW number or possibility to burn FPGA pins by switching fpga binary alone, a risk that we wont do nor try. no enforcement whatsoever this is not blind religion. if you dont need it, then dont do it. just as some people are not comfortable enough to edit start_rigol_app.sh and cause the dso to hanged, lucky if they made image backup, if not, they can be danggling in the Windows11 air, ymmv.

attached is CAN and LIN trigger support, but i never find a use for them since the beginning until today and probably many years to come, so i dont care if CAN or LIN work or not in HW12 or HW8... if someone comes up with trigger XXX hack, why would i need to do it if i dont use it?