New Buddha Flash causes A4000 to hang during boot-up.

erique · Sep 3rd 2021

Hi,

I have just taken a brand new Buddha Flash (20th version) out of its box to try it out in my A4000. (The Buddha was purchased in March 2021, but has been sitting in a box waiting for the A4000).

When the card is plugged in as-is (no jumpers changed, DOM inserted, J2 closed - basically "factory settings" - and no other IDE device attached to it) the A4000 will not boot up, but instead sit silently with a black screen.

As it was suggested in another thread that removing the PLCC would perhaps at least bypass the early hang, I removed it.

This made the A4000 boot-up, but the card is not visible autoconfig-wise (which makes sense).

But trying to flash the PLCC, by hot-inserting it, and running the BFlashTool ("initial") has so far been unsuccessful.

The tool outputs:

Code

1> buddhaflashtool initial buddha_ROM_52-101.bin
BuddhaFlashTool 1.14 (07-May-2018)
(C) 2018 individual Computers Jens Schoenfeld GmbH
Assuming $ef0000 initially.
Flash upgraded:no - CPLD upgraded:no
ROM loaded. CRC32 checksum: f1eed6c4
Press Y to start, CTRL-C to abort
Flashing now ...
Write error address: $0800
Error flashing ROM - Possibly flash protection is on?
1>

Display More

Any ideas what I could try next?

Cheers,

erique · Sep 3rd 2021

Hah, after attempting this a 5th time, and trying all imaginable voodoo-maneuvers, the flash tool suddenly succeeded with a "ROM flashed" message.

And yes, now the card seems fully functional (I can see the contents of the DOM at least).

I guess I'll just leave this thread here in case anyone else happens to get a board with an "empty" PLCC.

erique · Sep 3rd 2021

Oh well - it seems I spoke too fast: after power-cycling to connect the IDE drives and some other Z2/Z3 boards I'm now back where I started:

Black screen, that can be bypassed by removing the PLCC, but hot-inserting and trying to flash it yields the same output as in the first post.

Faulty PLCC perhaps?

Jens · Sep 4th 2021

Before I jump to any conclusions, I'd like to know if there is an boot device connected to the A4000's internal HD controller. I'm asking because the DOM has a low boot priority, and therefore will not be accessed if there are other boot devices.

Further, most problems we've had with these PLCC flashes were not reproducable here. We have even updated the QC procedure of the Buddha because of these reports: We are leaving at least two days between flashing the chips and testing the controllers, so short-lived flash contents are caught by QC. So far, only a tiny amount of bad chips were caught by this - which shows that the update to the procedure was good and needed, but it's not in line with your observation of "losing contents after a single power-cycle".

Next step would therefore be to double-check the power supply of your A4000: Does it maybe output high ripple? Or is it's startup-ramp not a ramp, but a swing-in with lots of over/undershoot? This is about the worst thing you can do to a flash chip, and old output capacitors or unsuitable PSUs (such as ATX types) can cause this very symptom.