exxos's Atari, Amiga & retro forum

Ok, so after some serious chip removal then testing , here is my best guess what happened to me

I swapped out the Ram first and managed to get it to run...briefly died again,so I swapped out the CPLD and ..got nothing

So then I swappped out the IDE to SD card and put a bog standard cheapo IDE to CF that we all normally use with the latest Firmware and it booted !!

So in conclusion did the IDE to SD card die and take the Board with it ? ( I know it was dead at the time because I did try the IDE to CF as well).

I do want to put the SD card adapter back in in case I kill it again. as you can see in the picture below that IDE to SD adapter has lot more going on than just the standard pass through IDE CF adapter we normally use.

Steve what I will do now is put it on soak test again with the latest firmware and with just a standard IDE to Cf ok, and lets see how that goes

If you can.. measure the voltage of the data pins on the ram side of the bus buffers.

I would recommend the pre alpha firmware for checking for dead ram.

My concern right now is that the RAM is getting 5V.

terriblefire wrote: 18 Dec 2018 10:45 If you can.. measure the voltage of the data pins on the ram side of the bus buffers.

Just measured random data lines on the Bus Fets and on the A1 to A8 data pins i get under 1v, and on the B1 to B8 data pins I get highest reading of 2,8v and lower
The Vcc is a constant 3,3v :)

So yesterday I didn't get much testing done since the kids were temperamental however... now the gold CPU I was using originally doesn't boot at all anymore. It looks like it might have died, I have to try it in a TF530.

Another gold CPU decided to start working with or without MMUDIS. The same CPU wouldn't get ram recognized the day before.

It's all very puzzling. Looks like completely random behaviour to me.

alenppc wrote: 18 Dec 2018 12:52 So yesterday I didn't get much testing done since the kids were temperamental however... now the gold CPU I was using originally doesn't boot at all anymore. It looks like it might have died, I have to try it in a TF530.

Another gold CPU decided to start working with or without MMUDIS. The same CPU wouldn't get ram recognized the day before.

It's all very puzzling. Looks like completely random behaviour to me.

Could be a bad PCB Alen ,wont be the first :)

supaduper wrote: 18 Dec 2018 12:26 Just measured random data lines on the Bus Fets and on the A1 to A8 data pins i get under 1v, and on the B1 to B8 data pins I get highest reading of 2,8v and lower
The Vcc is a constant 3,3v :)

Which side is connected to what ? (not familiar with TF designs sorry all!) I assume one side is the CPU side with pull ups, so assume that should be switching to 5V ?

The other side of the buffer, I assume is 3.3V side connected to the RAM.

So are you saying the supposed 3.3V RAM side is only switching to 1V ?

exxos wrote: 18 Dec 2018 13:06

supaduper wrote: 18 Dec 2018 12:26 Just measured random data lines on the Bus Fets and on the A1 to A8 data pins i get under 1v, and on the B1 to B8 data pins I get highest reading of 2,8v and lower
The Vcc is a constant 3,3v :)

Which side is connected to what ? (not familiar with TF designs sorry all!) I assume one side is the CPU side with pull ups, so assume that should be switching to 5V ?

The other side of the buffer, I assume is 3.3V side connected to the RAM.

So are you saying the supposed 3.3V RAM side is only switching to 1V ?

Ram side are pins B1 to B8 and give a highest reading of 2.7v

supaduper wrote: 18 Dec 2018 13:23 Ram side are pins B1 to B8 and give a highest reading of 2.7v

That should be fine then for 3.3V stuff.. a little low, but that might be the drop across the buffer (i've not checked the pdf).

Here are the data PDF`s for the Bus fets and the 258mb ram
all looks good regarding the tolerances from what little i know lol