Have you taken apart the system unit to look for battery leakage? When those batteries leak they tend to dump their guts on the traces leading to the rear expansion connector

Thank you for the reply!
I haven't taken it apart yet.
As far as I know, this IBM 5140 has only got a removable battery pack and no CMOS battery.
So, before eventually doing something else, I have checked the battery pack and it was already disconnected from the system board. I don't see any battery leakage from the connectors: they seem good.

I don't know if I will soon take all apart this computer, since it still works quite well and I don't want to damage anything else due to my lack of experience with older portable computers. Its case is so fragile that one little piece fell off when I took the front panel away to check what I could see; in addition to this, I suppose that it was disassembled by someone else before me, since there was already another broken part of the case (I'm referring to the two upper joints).

Anyway, from what little I've seen, there is no sign of battery leakage.
Could this all mean, along with the fact that this computer has no internal CMOS battery, that there are no damages caused by battery leakage on the board?

the battery pack is NiCad and it can leak just as well as a barrel battery can

however in this case it would appear to be clean

taking the system the rest of the way apart is very easy - the only fiddly part is separating the floppy drive edge connector sockets from the top case and putting them back without breaking a brittle plastic clip

For now, as a newbie, I am not really confident about my capabilities, even if this procedure could be easy to do. For me, this isn't the same thing as disassembling a common laptop that could be easily replaced: I really care about it! However, I'll give a try in the future when I have enough patience and confidence and I'll answer in this topic.
By the way, I have an another idea: I need to test other accessories that should be connected to the same I/O connector (a display adapter and a serial / parallel adapter), so it could be another opportunity to check the port before disassembling.
I'll update here!

Little update for this IBM.
I've tried its display adapter but it doesn't seem to work either for other reasons (I've tried a port which should be composite video). Firstly, the connection to the composite video port is somewhat loose; secondly, some of its capacitors seem faulty. When I powered on the computer with this adapter connected to the I/O port, the adapter started to smell bad, but it didn't smoke. I think that it is due to old tantalum capacitors which are prone to breaking in these IBM display adapters, according to what I've read surfing the Internet. I know that I should replace them, but I would do that after I properly learned how to solder.

In other words: the last attempt to test the I/O port, before eventually disassembling the system unit, could be the serial/parallel adapter, but I need the right peripheals to connect to it, which I haven't got for now.

When I posted this topic, I naively hoped that the printer required a "simple" fix like my previous few other repairings.
At least, I'm quite happy for what still works. And, well, I'm sure I will have an interesting challenge when I will be more skilled (and more confident) in the future. 🙂