Hacking the Rigol DHO800/900 Scope

[AndyBig]

I've attached a version of the rigol unlock tools batch file translated to a .sh file for those of us using Mac computers. It still requires having adb and the Go language installed.

However, my DHO914 is not accepting the generated BW15T25 option SCPI command (no generated .lic file), either with the original 00.01.00 or now the 00.01.01 firmware downloaded from the rigolna site.

EDIT: and also not accepting the option with the 00.01.02 firmware posted here.

Has anyone succeeded in increasing the bandwidth of a real DHO914 this way on either of those firmwares?

I think there should be no difference between a real 914 or an ex 814. My ex 814, which now considers itself a 914, accepted the BW15T25 generated option. But the generation tool was different - https://github.com/zelea2/rigol_vendor_bin

[S2084]

I succeeded deceive 814. BW15T25 passed without problems.

[AndyBig]

I succeeded deceive 814. BW15T25 passed without problems.

And I decided that an increase to 914 was enough. I didn’t find any reason to upgrade to 924, especially 924S

[S2084]

And I thought, why not? Now at least there are no empty spaces in the bottom panel.

[Mechatrommer]

I succeeded deceive 814. BW15T25 passed without problems.

can you explain the step how to add LA and AFG without removing your serial number?

[AndyBig]

Now at least there are no empty spaces in the bottom panel.

Good reason, I won't argue ))
...
I discovered a couple of days ago that the keyboard shortcut Win + ] splits the screen

[AndyBig]

I succeeded deceive 814. BW15T25 passed without problems.

can you explain the step how to add LA and AFG without removing your serial number?

Simply change the scope model to DHO914S or DHO924S in the vendor.bin file. But LA and AFG won’t work anyway.

[S2084]

exactly...

[Mechatrommer]

I succeeded deceive 814. BW15T25 passed without problems.

can you explain the step how to add LA and AFG without removing your serial number?

Simply change the scope model to DHO914S or DHO924S in the vendor.bin file. But LA and AFG won’t work anyway.

any specific steps how to do that? i didnt follow this thread closely sorry. i'm working on to make both AFG and LA to work now, still in progress...

[AndyBig]

i'm working on to make both AFG and LA to work now, still in progress...

Oh thats cool! I will look forward to hearing about your results
In addition to this hardware, it looks like you also need to solder 2 missing memory chips onto the main board of the oscilloscope. Presumably this memory is needed for LA, but as far as I know, no one has yet tested this in practice.

any specific steps how to do that? i didnt follow this thread closely sorry.

1. Download the files /rigol/data/vendor.bin and /rigol/data/Key.data from the oscilloscope using ADB:
adb pull /rigol/data/vendor.bin
adb pull /rigol/data/Key.data
2. Using this program - https://github.com/zelea2/rigol_vendor_bin - change the oscilloscope model to DHO914S or DHO924S in the vendor.bin file:
rigol_vendor_bin.exe -M DHO924S
3. Using the same program, generate all possible options using the Key.data file:
rigol_vendor_bin.exe -o
For convenience, you can direct the program output to a file so that all generated options are saved in it:
rigol_vendor_bin.exe -o >>options.txt
4. Load the vendor.bin file back into the oscilloscope:
adb push vendor.bin /rigol/data/
5. Reboot the oscilloscope.
6. Send the DHO900-BODE and DHO900-BW15T25 options from those generated in step 3 via the SCPI interface.

[Mechatrommer]

1. Download the files /rigol/data/vendor.bin and /rigol/data/Key.data from the oscilloscope using ADB:

why i asked is one day when i'm done and ready to do FW hack, this thread will be 100 pages, it will be very difficult to find specific hack like this esp not everyone interested.. hence i saved your post's link with the content offline (local HDD) for my quick reference later, thanks.

In addition to this hardware, it looks like you also need to solder 2 missing memory chips onto the main board of the oscilloscope. Presumably this memory is needed for LA, but as far as I know, no one has yet tested this in practice.

reading rigol diy logic probe threads... i tried quick hack shorting "probe present" detection pin to gnd and LA gui got activated. i saw some up and down in LA signals, maybe due to random floated pins, so i guess this thing doesnt require the 2x DDR ram. i only soldered 2x unpopulated dual opamps near LA port (i guess its for trigger level D0-D7 and D8-D15) and one MPM3630 smps converter (with some difficulty) for 4V rail. i'm planning for external 5V power bank to power that rail from LA probe in case people are not willing to do MPM3630 on DHO800 board, so bare minimum is to solder 2x msop-8 opamps on DHO800 board. one thing i noted is 50Mpts memory for CH1 got halved when LA gui activated, so hopefully LA is using the another half, not from unpopulated DDR3L ram.


https://www.eevblog.com/forum/testgear/low-cost-logic-analyzer-probe-for-rigol-mso5000-easyeda-project/msg5200419/#msg5200419

since its look like brighter hope, i already had a go at doing this and already cut the front panel of my scope and made youtube video to guide others if interested, its irrevesible, and since its not final yet (the LA probe probably wont work as expected), i dont think to publish the link yet so far.

same thing with AFG module.. 2x dual opamps (original TP1282, i replaced with OPA2350, same as the LA) required to install for AFG offset and magnitude (gain) i already have a working prototype module but i have noise/crosstalk issue to fix at low level signal, thats why i asked csuhi17 in fw bug report thread about it, i already redid pcb ver2... now waiting for LA probe module to complete before sending both prototype to jlcpcb... fwiw.

https://www.eevblog.com/forum/testgear/rigol-dho804-bode-plot

[bikeNomad]

I've attached a version of the rigol unlock tools batch file translated to a .sh file for those of us using Mac computers. It still requires having adb and the Go language installed.

However, my DHO914 is not accepting the generated BW15T25 option SCPI command (no generated .lic file), either with the original 00.01.00 or now the 00.01.01 firmware downloaded from the rigolna site.

EDIT: and also not accepting the option with the 00.01.02 firmware posted here.

Has anyone succeeded in increasing the bandwidth of a real DHO914 this way on either of those firmwares?

I think there should be no difference between a real 914 or an ex 814. My ex 814, which now considers itself a 914, accepted the BW15T25 generated option. But the generation tool was different - https://github.com/zelea2/rigol_vendor_bin

I was finally able to get the BW increase using the rigol_vendor_bin, while the rigol unlock tools didn't work for me. I don't know why there was a difference, but I now have 250MHz bandwidth displayed.

[AndyBig]

why i asked is one day when i'm done and ready to do FW hack, this thread will be 100 pages, it will be very difficult to find specific hack like this esp not everyone interested.. hence i saved your post's link with the content offline (local HDD) for my quick reference later, thanks.

I agree, unfortunately, it’s already quite difficult to find useful points in such a large topic. I have seen forums where a topic is assigned a curator who regularly updates the first post of the topic with links to interesting and useful posts from the topic. It would be very convenient
You go into the topic and immediately see a mini-QWO with links to relevant posts:
- how to update the firmware
- how to upgrade to an older model
- how to open options
- useful programs
Etc.

reading rigol diy logic probe threads... i tried quick hack shorting "probe present" detection pin to gnd and LA gui got activated. i saw some up and down in LA signals, maybe due to random floated pins, so i guess this thing doesnt require the 2x DDR ram. i only soldered 2x unpopulated dual opamps near LA port (i guess its for trigger level D0-D7 and D8-D15) and one MPM3630 smps converter (with some difficulty) for 4V rail. i'm planning for external 5V power bank to power that rail from LA probe in case people are not willing to do MPM3630 on DHO800 board, so bare minimum is to solder 2x msop-8 opamps on DHO800 board. one thing i noted is 50Mpts memory for CH1 got halved when LA gui activated, so hopefully LA is using the another half, not from unpopulated DDR3L ram.


https://www.eevblog.com/forum/testgear/low-cost-logic-analyzer-probe-for-rigol-mso5000-easyeda-project/msg5200419/#msg5200419

since its look like brighter hope, i already had a go at doing this and already cut the front panel of my scope and made youtube video to guide others if interested, its irrevesible, and since its not final yet (the LA probe probably wont work as expected), i dont think to publish the link yet so far.

same thing with AFG module.. 2x dual opamps (original TP1282, i replaced with OPA2350, same as the LA) required to install for AFG offset and magnitude (gain) i already have a working prototype module but i have noise/crosstalk issue to fix at low level signal, thats why i asked csuhi17 in fw bug report thread about it, i already redid pcb ver2... now waiting for LA probe module to complete before sending both prototype to jlcpcb... fwiw.

https://www.eevblog.com/forum/testgear/rigol-dho804-bode-plot

This is very interesting and cool! If there are ready-made instructions for making AFG and LA, that will be great

[Mechatrommer]

This is very interesting and cool! If there are ready-made instructions for making AFG and LA, that will be great

it will involve buying lots of parts.. rev engineering, making circuit and pcb and assembling them my target is cheapest 250MHz 1GSps MSO+sig_gen (bode plot and LA probe included) at cheaper cost than DHO924S (or DHO914S) alone...

[AndyBig]

it will involve buying lots of parts.. rev engineering, making circuit and pcb and assembling them my target is cheapest 250MHz 1GSps MSO+sig_gen (bode plot and LA probe included) at cheaper cost than DHO924S (or DHO914S) alone...

Whoever wants to do it will buy it Whoever doesn’t want to or can’t will buy it from you on eBay

[lucian]

Does anyone have the 1.02 firmware for the DHO800 and also when upgrading the DHO804 to DHO924 are all of the DC reading within spec for the DHO924 and is anything out of the spec in the datasheet

[Randy222]

Does anyone have the 1.02 firmware for the DHO800 and also when upgrading the DHO804 to DHO924 are all of the DC reading within spec for the DHO924 and is anything out of the spec in the datasheet

IIRC someone attached it as a zip, and/or gave a link to the zip that has the 01.02 FW, but it's a few posts/pages back. You'll need to do some digging.

[lucian]

But when you upgrade it to 924 does it now act within the specs or is it still off

[Randy222]

But when you upgrade it to 924 does it now act within the specs or is it still off

800's to 900's seem to behave ok. Certainly 802 to 914, or 804 to 924S will have some limitations due to missing hardware/connections.
You gain mainly better scale resolution, mem depth, and BW.
My 804 runs as a 914, seems to work ok (on FW 01.02). It's kinda at the limitations of the hardware at this point.

[lucian]

would you recommend upgrading to 924 or 914 and whats the difference not the S

[ebastler]

would you recommend upgrading to 924 or 914 and whats the difference not the S

Not being willing to read a couple of pages in the forum is one thing. Not being willing to read even the high-level product bullet points at any online seller's website, takes it to another level. I am not sure you are ready to perfom any "hack", since you would need to follow a paragraph of instructions or two. 

[lucian]

I have read through most of the thread and the sellers documentation I understand that 914 is 125 mhz and and the 924 is 250mhz but to my understanding there is a rigol addon which allows you to change the 914 to have 250 mhz aswell. I just wanted to confirm that I didn't miss anything given this thread is 33 pages.

[ebastler]

Yes, you can upgrade within each family (804 to 814, 914 to 924) by buying upgrade keys from Rigol, or generating such keys yourself via a key generator script which is also available in this thread. Going from the 800 to the 900 series requires the other hacking approach, by modifying or replacing the vendor.bin file.

It should be equivalent whether you go straight to the 924 via its vendor.bin, or go to the 914 and then apply the bandwidth upgrade key later.

In either case, it is questionable whether the 924 model makes much sense. The actual, measured bandwidth for the 914 is already higher than specified by Rigol, namely around 200 MHz. And given the limited sampling rate of the small DHOs, they can't really keep up with that bandwidth once you enable more than two channels (and hence run at only ~ 300 MHz sampling rate per channel). So you will get aliasing if higher frequency components than half the sampling rate are present in the signal.

Frankly, I would stick with an upgrade to the 814 model. That one also has the higher measured bandwidth, and it won't annoy you with a display for the non-existent digital channels on the screen.

[lucian]

is it better to change the vendor.bin from 804 to 814 or just do the license key upgrades or is there no difference

[ebastler]

is it better to change the vendor.bin from 804 to 814 or just do the license key upgrades or is there no difference

The resulting functionality should be exactly the same. Personally I would prefer the approach via the license key, since the result looks more "official": The scope will still identify as an 804, consistent with its front panel, just with the 100 MHz bandwidth upgrade applied. In contrast, an 804 which identifies as an 814 can only be obtained via hacking.

Either modification is reversible, of course, e.g. if you need to send the scope in for service. It's a bit easier with the license key approach, where you just need to send an official SCPI command (:SYST:OPT:UNINST).