TF330 Akiko Rev 0 Issue (SOLVED)

[alenppc]

While I am patiently waiting for the rev3 pcbs to be delivered, the first rev2s are beginning to hit the streets.

Someone told me that the card only works on one of their CD32s but not the other one, while the 328 works on both.
Someone else is also having trouble getting their card to boot while the 328 works fine in the same machine.

Needless to say, no issues on my machines. Not sure what to make of this. Did you experience anything similar supaduper?

Apparently the symptom is flashing led and/or yellow screens at boot.

[terriblefire]

Maybe it’s the PSU? Did they use the same PSU for both tests?

It’s vaguely possible the CD32 needs a recap. Either way a fuller summary of the issue would be useful. Does it fail when disabled for example?

[terriblefire]

Actually. It could be bad contacts at the edge connectors. The TF330 needs way more pins than the 328.. e.g the BG/Br pins .. if they have bad contacts the two CPUs will boot at the same time.

[alenppc]

Underpowered PSU would be my prime suspect, anyway I will keep investigating, but I am curious if supaduper has had similar complaints.

EDIT: I agree with dirty contacts problem, it's a _very_ common issue on the CD32 however I did instruct the relevant parties to try cleaning them.

[terriblefire]

If there is a particular CD32 out there that is misbehaving I’m happy to buy it at cost of a replacement from Analogic... I could then use that machine for future testing. Also happy to just take the main board not the case.

[supaduper]

Hi alen

I have had no one yet say they had problems with theirs...but its only been a week and I have shipped 25 rev2/rev3

I soak test for some hours mine, also I will bet its something like bad contact as its only that one machine, I had a few say with the TF328 that it wont boot, I tell them to clean the contacts then 9/10 its sorted

[alenppc]

I have had no one yet say they had problems with theirs...but its only been a week and I have shipped 25 rev2/rev3

Wow, you must have a ton of free time. Well done!

[nils]

I'm one of Alenppc users

My CD32 is NTSC version, racap was done Dec 2018 was in perfect condition (no leaks)

Original Power brick (I'm trying to source a new power supply/parts now just in case)
With the DisableSpeed jumper on the lower 2 pins I get 6 quick 1 long blink of the green power light then a yellow screen and then what appears to be a roboot/restart.

With the DisableSpeed jumper removed I get the CD32 boot logo then a yellow screen or just a yellow screen (first time I got some glitches on the screen and it crashed).

I've removed the 44 pin IDE/cf adapter still same results as above

I've tried a bit of alcohol to clean the contacts on both the card and the CD32 the TF328 works with either riser (both kipper2k v6)

Once done swapping risers etc the TF328 works fine when I install it

[anton.gale]

alenppc,

I'm one of the guys having issues with my TF330. I've cleaned the edge-connector to the point I'm pretty certain that it isn't the issue. The PSU is pretty beefy and the unit has been re-capped. Please see a snapshot from DiagROM below. This repeats until board 32.

This tells me (and you guys can correct me) that the 68030 is working & some parts of the autoconfig appears to work. The RAM is reported @ 0K Fastram from within DiagROM 1.0, but I don't know how reliable that is.

I am not complaining, I'd like some additional things to try out. I can solder & have a scope, so I'm willing to try whatever you guys throw at me.

autoconfig.png (144.66 KiB) Viewed 5029 times

[terriblefire]

That repetitive AutoConfig is a bug in DiagROM 1.0. It’s fixed in newer versions. If DiagROM says the RAM is good then it’s probably good enough to boot. I can make a non ram firmware later to help diagnose.

EDIT: try running the IRQ and CIA tests. And graphics tests.