exxos's Atari, Amiga & retro forum

Well, the LCIII is home.

First checks:

• No shorts.
• Power rails good.
• The CPU is getting a 25MHz clock.
• Reset stays low and then goes high.
• /HALT not asserted.
• A single HIGH->LOW on address lines.

I've yet to check /DTACK etc. though surely if /DTACK wasn't asserted the address would stay on the bus and if the CPU was running at all.

More investigation later.

P.S. Note: This is all being measured from the top of the pins on the CPU, so it's what the CPU sees and not what the PCB may be connected to if there are bad connections.

D12 and D13 seem to be shorted on the CPU and on the PCB at the CPU but not on the processor direct slot. Very strange.

Anyway, the next task will be removing the CPU and cleaning up all the pads. For a supposedly 33MHz part it's getting a little too warm in one corner for my liking at 25MHz. Maybe it's a 16MHz part which has been relabelled.

@exxos this is one of the CPUs you sent to me with the adapter board. Did you ever get any of them to work?

stephen_usher wrote: 07 Mar 2024 20:03 @exxos this is one of the CPUs you sent to me with the adapter board. Did you ever get any of them to work?

The CPUs we tried with the TF536 did work okay, so you should be fine at 25mhz

Thanks. I'm always distrustful of "new old stock" chips these days.

Looking closer.

Somehow D8 - D10 are all shorted to ground and there are some missing pads too, e.g. /ECS pin.

Looking as pictures, there are four missing pads and one which is a bit skew-wiff. I'm going to have to be very careful getting this chip off over the weekend.

stephen_usher wrote: 08 Mar 2024 14:20 Looking as pictures, there are four missing pads and one which is a bit skew-wiff. I'm going to have to be very careful getting this chip off over the weekend.

Assuming that the solder is probably leaded since it was removed before, that may make it easier. I see much flux in your future...

rubber_jonnie wrote: 08 Mar 2024 14:37

stephen_usher wrote: 08 Mar 2024 14:20 Looking as pictures, there are four missing pads and one which is a bit skew-wiff. I'm going to have to be very careful getting this chip off over the weekend.

Assuming that the solder is probably leaded since it was removed before, that may make it easier. I see much flux in your future...

So do I, and kapton tape.

stephen_usher wrote: 08 Mar 2024 15:09

rubber_jonnie wrote: 08 Mar 2024 14:37

Assuming that the solder is probably leaded since it was removed before, that may make it easier. I see much flux in your future...

So do I, and kapton tape.

Indeedy

Managed to get the chip off this morning.

Thankfully the missing pads don't seem to be connected to anything and it's not as bad as it looked. There are two floating pads but they're still well connected.

I've managed to get the new CPU mostly on.

It looks like D20 and D21 are shorted somehow but I can't see anything in the area I've soldered. I should have checked before I replaced the chip.

The biggest problem I have is that the pins on the pin 1 side of the chip got more and more out of alignment as you go along the chip. It's only a very small amount of distortion but by the time you get the full way across the chip it's out by one pad. Also, the legs aren't all level with the board, by a tiny amount. I do hate surface mount at the best of times but a fine pitch where the chip is effectively second hand is a nightmare. I had to re-tin all the legs due to tarnish too.

I may have to remove the chip again, which risks breaking the pads. *sigh*