Rigol DHO800 [814] failing to boot up

[norbert.kiszka]

CPU (RK3399) and RAM closest to CPU.

[Hix]

Thank you.
Can I ask, are these known to have hardware issues?
I've been trying to scour the interwebs for any examples of DHO800 failing and not found anything.
It just seems strange that it was working before it wasn't. I would have thought that if there was a solder issue it would have prevented it booting form the start.

[norbert.kiszka]

Can I ask, are these known to have hardware issues?

Couple years ago, I was living from repairing mostly laptops. In older and newer ones, problems with balls are quite often.

Anyway, malfunctions are pretty often caused by something simple, like shortened cap. But diagnosing it is a another story.

[Fungus]

Can I ask, are these known to have hardware issues?

No.

I've been trying to scour the interwebs for any examples of DHO800 failing and not found anything.

There's your clue, right there.

[Hix]

Well, I've used a hot air station to try to reflow the ICs and memory chips, to no avail.
adb is showing nothing.
I've tried the image that has been recommended from the first post in the hacking the DHO800 thread, that hasn't worked. Though when I later tried to flash the SD with a different imaging tool [balenaetcher] this tool flags that the archive may be corrupted. This could be an issue, but I cannot find any other images and trying to find on other threads is seemingly irksome to other users.
I'm at a loss as to what to do.

[Fungus]

Do these 'scopes flash any lights on the front panel or click any relays when you turn them on?

[norbert.kiszka]

Are You using Windows and Windows tools? For a company known as Microsoft (or maybe Microsh*t), everything not designed by them (including disk format) is corrupted.

My advise is, change Your OS for something better, like BSD or Linux. On that two, You can use disks from almost all other operating systems.

[norbert.kiszka]

Do these 'scopes flash any lights on the front panel or click any relays when you turn them on?

I can't understand why You are trolling this forum for so long. Those questions are dumb as hell.

Buttons are driven by a separate IC and after powering it up, it will always flash them for couple seconds. Relays are driven by CPU GPIO and AFE (I2S communication based on CPU GPIO). Both will be silent, when CPU can't boot.

[Hix]

I've just used  dd on mac terminal, no errors flagged, but the image doesn't boot.
Are there any specific points on the board that I can probe with a DMM to ascertain if voltage is present where it should be? I've measured random points and found 3 and 3v3 present, but am wondering if there is somewhere that would flag dead or alive more quickly?

Are You using Windows and Windows tools? For a company known as Microsoft (or maybe Microsh*t), everything not designed by them (including disk format) is corrupted.

My advise is, change Your OS for something better, like BSD or Linux. On that two, You can use disks from almost all other operating systems.

[Fungus]

Both will be silent, when CPU can't boot.

Right, so this is a test for...?

[norbert.kiszka]

Both will be silent, when CPU can't boot.

Right, so this is a test for...?

This is a test for something that we know already.

[thm_w]

There should be many more rails for the CPU/FPGA. 1V, 1.2V, 1.6V something like that. Any inductor you see would likely be a rail. They will lead to a large ceramic cap. Who knows if all rails are on on boot though.

If it does end up unrepairable, sell it to someone here so they can use the ADCs to attempt to upgrade the DHO1000.
Hopefully OP didn't reflow those, and only tried on the CPU.

[Hix]

Nope didn't reflow anything else other than the CPU and memory chips immediately adjacent.
I've measured voltages at many points on the board. There are lots of points around 1v. a few 3v and 3v3 there was also one point at 8v?!?
When I have time outside of work, I'll photograph the board and document what I have measured and where. I'm living in hope I can resurrect it.
Failing that I would sell it to someone wishing to develop something from it. Think of it as a kidney donor

[ptluis]

UPDATE! UPDATE! i received a message stating that this model doesn't have a HELP button! but the model i recovered does in fact has this button located on the place where the QUICK button exists! My customer bought it directly from china it was one of the first units to be released, could this be a counterfeit unit? demo?

In this case I don't know if it will apply to other scopes without this key. Anyway I leave this post and my experience on this
------------------------------


I was able to revive a non working Rigol DHO814 from a
customer. This customer tried to hack this device and he told me when he was trying hack through adb there's a power failure when he was sending data. After that the scope stop working. 
I tried other SD cards with different firmwares and no luck.

Symptoms: lights turn on, relays don't click, Rigol logo not displayed, fan working, nothing more work, no adb access.


I leave the original sd card on scope before I try the following:

Somehow I was able to enter booloader mode of the scope, but logo isn't displayed, I read on other post that logo is controlled by bootloader, but honestly i can't explain why it worked.

I tried entering bootloader mode 5 times, keep reading.

This is how I did it, if you have a broken scope you don't have nothing to loose, maybe it also work for you. The only negative point is that the Rigol logo stop being displayed!

1. I used a 16GB usb flash drive formatted in FAT16

2. Download the latest firmware file for the DHO814 from the official RIGOL website. Extract the GEL file from the zip and place it in the root directory of the USB drive.

3. Insert the USB drive into the USB HOST port on the front panel of the DHO814.

4. DHO814,  From power off state,  press and hold the HELP button and power it back on, and release the power button BUT don't release the HELP button until you see the led on the flash drive blinking, it must be continuous blinking, if it just blinks briefly and stops, it won't work!
 I HAD TO REPEAT THIS 5 TIMES, i was almost quitting and suddenly it worked.

5. there's no information displayed on screen,  when the flash drive led stops blinking, I waited another minute and then power off the scope.

6. I remove the usb-c power cable and the flash drive and wait a minute

7. i turn the scope on, nothing happened, but i decided to leave it on because i remember i recovered some smartphones waiting longer at the beginning and after around 1 minute the scope become alive and boot almost perfectly it took a total of around 3 minutes.

8. I rebooted the scope several times BUT The RIGOL LOGO is NOT displayed!
Calibration took around 45 minutes

My customer accepted the scope the way it is and took it back. At the moment of writing this text he didn't complain.

I hope this helps those who have similar problem and could recover you units, you have nothing to loose, just a little of your time.

[Hix]

That sounds very positive Luis. It isn't working on my scope, but that could possible be that the GEL file I have is already installed. I wonder if I try and older firmware, will that flash? I'll have to see if I can download a previous GEL file and try that. A d it is the Quick button on the upper right, below Clear that you held down until the flash drive stopped flashing?
As luck would have it the only drive I have that lights up is the very one I was using for the DHO814. There's serendipity there for sure.

[Hix]

Damnit. Rigol only lists the latest GEL file to download. I don't suppose anyone has anything previous they could share do they?
I have 00.01.04.00.02 so anything previous to that would be great. Assuming that is, that you can 'downgrade'

[norbert.kiszka]

I think I have all firmware versions, starting from 00.01.00.00.00. But I have mess on a laptop disk, Im tired and that will not work, since this recovery method works with bash scripts. Bash will not works If You don't have at least Linux kernel (Android kernel is basically Linux) booted. If there is complete silence on UART, and You tried other SD card with other image (assuming that was correct - on Windows You can't be sure for anything - that's why Im not using Windows), then this will not help, because You have hardware issue, not with software.

Failing that I would sell it to someone wishing to develop something from it. Think of it as a kidney donor

How much?

There are lots of points around 1v. a few 3v and 3v3 there was also one point at 8v?!?

I don't know if there is a 8 V somewhere (Im aboard and far from my scope, so I can't check), but two things. There is AFE IC, which has opamps - Im preety sure, it use negative rail also. Second thing, AFG uses +12 and -12, but I don't remember where was source of that voltages (maybe that AFG board has own converters, but Im not sure).

[ptluis]

That sounds very positive Luis. It isn't working on my scope, but that could possible be that the GEL file I have is already installed. I wonder if I try and older firmware, will that flash? I'll have to see if I can download a previous GEL file and try that. A d it is the Quick button on the upper right, below Clear that you held down until the flash drive stopped flashing?
As luck would have it the only drive I have that lights up is the very one I was using for the DHO814. There's serendipity there for sure.

Hi i tried with the latest firmware 00.01.04.00.02

 I add an update note at the beginning of my post. look like the unit I recovered is not the same sold right now. the one I recovered had an HELP button in the place of the QUICK button. Ive been searching for pictures all around but can't find one with the help key. I wonder if it was a prototype? demo? fake? I discovered this by trial and error maybe on the new units some other key does the same? can't help you more because i don't have the scope anymore.

[Hix]

I've tried pretty much all of the other buttons but no success. I already had that version of the firmware, so it likely wouldn't do anything anyway.
But it is looking increasingly like hardware failure.
Annoying as reading your post, it sounded like a carbon copy of what I am faced with.
Running out of hopes and dreams 😂

[Hix]

Are there schematics available for this model?
Given it looks like there is a hardware failure I might as well learn some fault finding skills. If I have half an idea where to look I stand some chance of possibly finding something that isn't as it should be.
The fact that one CPU is seemingly working and spitting out data via its UART, then something down chain from there is seemingly the issue?
What would those wiser than me say about it being a passive that has failed rather than the main CPU? Are there lines from the CPU which would indicate that it is alive, other than its UART?
Ita an expensive paperweight so I may as well use it as a learning tool at least.

[thm_w]

Are there schematics available for this model?
Given it looks like there is a hardware failure I might as well learn some fault finding skills. If I have half an idea where to look I stand some chance of possibly finding something that isn't as it should be.
The fact that one CPU is seemingly working and spitting out data via its UART, then something down chain from there is seemingly the issue?
What would those wiser than me say about it being a passive that has failed rather than the main CPU? Are there lines from the CPU which would indicate that it is alive, other than its UART?
Ita an expensive paperweight so I may as well use it as a learning tool at least.

No, schematics aren't available.
I thought it was the FPGA UART that was working, and the CPU UART was not outputting anything?

There are common failures you can always look for:
- shorted ceramic capacitors
- shorted FETs (though these are more likely only after device is running for a while)
- hiccuping/dead rails (may require an oscilloscope)
- Oscillators not oscillating (look for an EN pin, see where that leads you)
- maybe stuck buttons or issues on the front panel board?

Then beyond that its tracing things around and figuring out the design for yourself.
Would start from voltage input, follow it down through the converters.

[Hix]

Are there any differences between any of the PCBs?
I’m just wondering about a hypothetical scenario whereby I were to purchase a unit from an EU supplier that strangely developed the same fault.

[Fungus]

Are there any differences between any of the PCBs?

Not in theory.

Maybe there's revisions where they changed components for price/supply reasons but they should be functionally the same.

[Hix]

I managed to log the output of the FPGA UART. I have attached this to the post as a txt file.
The CPU UART remains silent.

[Hix]

Having finally found Dave's logs of what his 814 was doing when he did the teardown, it seems that my FPGA is stopping a few lines short of his. Apart from his boot log being seemingly duplicated, mine is missing the following lines towards the end of the boot. I am guessing that this is the handover to the CPU?
My bootlog finishes here:

Code: [Select]

DMA Done !

Handoff Address: 0xFFFF0000
In FsblHookBeforeHandoff function
SUCCESSFUL_HANDOFF
FSBL Status = 0x1
This is MultiBoot Image 3!!!
FlashID=0xEF 0x40 0x18

Whereas Dave's bootlog finishes with these additional lines

Code: [Select]

Recv Command Boot!!!
Target Addr: 0x00400000
Running Command Prog!!!
Try MultiBoot
Send Software Reset!!!