Rigol DHO800 [814] failing to boot up

[Hix]

Hi,

Thanks for trying to help, currently the scope it sat in a warehouse in China, so I am unable to do anything. But it looks like I will have to recall the shipment as UPS have stated that it could be destroyed if not resolved in three days. Will let you know if I get it back into my hands.
I had tried with a couple of different firmware versions. Card was correctly seated and the contacts cleaned with IPA. This is what led me to believe it was hardware related and return it due to the amount of time I had to do so before any warranty period expired.

[Hix]

Well, shes back in my hands. Am going to try flashing a few firmwares and see if that does anything. Some of the inductors were reading 1V but UART was 3v3 so that looks positive

[Hix]

Flashed the image from https://mega.nz/file/UjBC3KRY#Kqv1BCHNQdPcUGMfR8IqbuUwHUsUhU4GpO1keTAXqf8 but still doesn't show the Rigol screen on boot. The fan runs and the buttons do their thing. Tried HDMI just in case but notihng, but then again I'm 99.9999% sure it needs the OS to decode and port the HDMI out.
Are there any more image files out there? I don't hold out much hope.
Will see if I can jury rig something to try and probe UART, any things to look for if I do this, or am I just looking to see if it is spitting out bits?

[Hix]

This is the back of the PCB, I see nothing untoward, bar some crap in the solder on the upper mount peg of the USB header, lower right.

[ptluis]

This is the back of the PCB, I see nothing untoward, bar some crap in the solder on the upper mount peg of the USB header, lower right.

Try flashing on a different SDcard

[thm_w]

Is there any difference in behavior with the SD card installed or with it removed?

If the above suggestion doesn't work, maybe inspect the SD card slot for damage, and pins for solder. From the image it looks OK. But its really easy to damage those SD insertion detection tab things.

[TurboTom]

Did you remove the screw in the heat sink or has it been missing right from the start? If the latter is the case, I'ld disassemble the whole scope very carefully and look if the screw has been lodged in there somewhere, maybe causing the trouble you're observing.

[Hix]

Same behaviour with and without the micro SD. I've flashed the firmware listed to no avail and tried another known working card too.
I've looked at the solder joints with a loupe and they look sound.

[Hix]

Yes, I detached the main board from the controls board and removed and reseated the ZIF for the display, in the hope that it could have just worked loose. I just placed the screws in easier holes to prevent loss whilst the case was removed.
All screws are accounted for and not shorting the PCB.
This error predates the removal of any screws.

[FixyMixy]

Does anyone know if there is a hard reboot sequence for the DHO814? I cannot get it to boot up. I've tried multiple power blocks including a MPB 130W jobbie.
Once I hit the red power light, it turns green and that's as far as she goes. Absolutely nothing on the display. It hasn't been used for more than 10 minutes. Any ideas? I'm all ears.
Here's the issue https://youtu.be/foNty4EBuQ0

Oscilloscope boots into adb early at boot sequence. You should try to backup your personal files. Download "minimal adb for windows".
Try (from the adb folder):
1) connect:
adb devices
adb connect <your scope IP>:55555
adb devices
2) pull it:
adb pull rigol/
3) Download rigol dho800/900 sd card image https://mega.nz/file/06VDFaBI#3pmlujrkarmJ63_5Vh9VDz-VOt_nxBVmmBv0GXrSxZo
flash it to a (working ) sd card.
4) Push your personal files (but only from the /rigol/data/ folder not to owerwrite executable files) to the scope.

[Hix]

I have no personal files, so is there any value in trying to pull anything via adb?
I have already flashed that image and it is not working, have tried two known working as cards with that same image, to no avail.

[ptluis]

If there's no Rigol logo then it's badly messed up. The only thing to try would be to open it up and reflash the SD card (or just reseat it).

If it's new though, send it back. It could be a component failure, too.

Does anyone know if there is a hard reboot sequence for the DHO814?

No, there isn't. Not when things are that bad.

Fungus did you manage to know what failed in your scope?

[FixyMixy]

But still not seeing anything on boot. It is doing some kind of POST as the buttons light up on first powering.

Oh, I see you've already tried to flash fresh system image. And it kind of boots up. Try to download logs with the adb to see what's wrong.

[FixyMixy]

I have no personal files, so is there any value in trying to pull anything via adb?
I have already flashed that image and it is not working, have tried two known working as cards with that same image, to no avail.

Without calibration files you lose some accuracy, but the scope should be usable anyway.

[Hix]

That relies on getting it to boot, which is the primary point of failure. Calibration can be handled once it actually fires up, if it ever does.

[FixyMixy]

There is a special script on the scope to copy logs to the USB drive.
adb shell /rigol/shell/copy_logs_to_udisk.sh

[Fungus]

Same behaviour with and without the micro SD. I've flashed the firmware listed to no avail and tried another known working card too.
I've looked at the solder joints with a loupe and they look sound.

At this point it's fairly clear that it's NOT the SD card or flash image.

It's a hardware failure. Maybe a bad chip, maybe bad soldering, maybe bad power.

Power supply is easy to check but there's lots of little voltage converters on the PCB. Look for voltage test points.

[Hix]

------Edited------

Fungus, I've proved probed a few points on the board seems to be legit
So, the UART to the right of the SD car on the upper edge of the board shows 3v
The six port header just below that reads 3v3
Some of the pins on the SD reac 3v (well some 2.9xxx but deemed that good)
Is there anywhere in particular that would be wise to probe?
I've got a pi pico somewhere so will try and use that to see if there is any UART data when it tries to boot.
I've tried the HDMI to another screen, but suspected it was a futile idea and was proven right
Any other ideas on how to unbrick this?

[Hix]

So, I've managed to dig out an old Saleae clone 8CH logic analyser. I've hooked it up to the GND TX RX on the uppermost header on the right hand side of the board. TX line shows as HIGH and RX shows as LOW but there don't appear to be any packets. 
Happy to listen to those wiser than I as to what data settings to use, or anything else to try to read any output
I am a rank amateur at UART so could be me getting things wrong.

[norbert.kiszka]

5 $ UART-USB converter and some terminal (Im using Linux with it and without problems). Even if You made wrong settings, You should be able to see early U-boot logs as some bad chars - at least. If not and voltages are ok, then next step will be to heat up all BGA chips with hot air, to the point of melting solder joints, because single bal can be a cause.

And I repeat myself again: If You want to buy new expensive staff, buy it from certified reseller instead of Amazon, Ebay or something like that. It's more expensive, but much less problematic.

[Hix]

Connecting to another header on the PCB

Yields some results. There are strings that mention reboot, so something is working in there....


It's all well and good reiterating about purchase selection Norbert, but without the ability to move time, it is a little fruitless. I'm stuck with what I'm stuck with so am trying to make the best of what I have. Anything that is going to resolve this is very gratefully received

[norbert.kiszka]

I think, that UART is from FPGA (which have own CPU with two cores). Use UART which is located literally on top right. Even with broken FPGA, screen is connected to the CPU (via MIPI-DSI) and You need to boot CPU to make anything (including installing and playing games...) with this scope.

[Hix]

Yeah, there is no output from the UPPER RIGHT UART.
Would this point to a failure of an IC in some way? Surely if there was a bad BGA it wouldn't have booted up all those times before?
I'm trying to think what the last thing I was doing with it was. I have a feeling I was testing WiFi dongles using a USB hub plugged into the front USB port.

[norbert.kiszka]

Check 5 V at USB socket and heat up those damn BGA chips. All of them, including memory around CPU. Check all voltages from RK808 (not only inductors, because it has also LDO), if You didn't already.

[Hix]

Do you have a diagram of where the BGA chips are? I don't really want to risk reflowing ICs that I don't need to.