I have a
Viper 1230 (photo) and the real time clock appears to be going (all I have in workbench is the correct year and month which I guess it could be storing somewhere other than an RTC?). Note it does not have this
extra RTC board (on left edge).
The problem as you can see from the photos and
this closeup is that the button cell appears to be soldered onto a flimsy piece of metal. I have no idea what the piece of wire going across the board is for btw - I have that on mine

The cell appears to be a CR2032 (unconfirmed as I cannot get access to the number inscription).
Can someone tell me if...
a) Whether the Amiga holding the time within a few days of what I set is the RTC working properly or a software bug (which I hope is more likely)?
b) There is some software I can use to test the RTC (other than just switch it off for a few minutes

)?
c) The battery is replaceable (I have a soldering iron although I'm not too good at it and don't want to explode the battery

) and whether a standard CR2032 is suitable or I need to find a rechargeable variant?
The RTC jumper is already set, the battery has not leaked.
Thanks,
Benjamin
[EDIT] Bizarrely I can use my PCMCIA port without problems and have an 8MB SIMM installed so I think the docs on the first link are wrong. Before I covered the 68030 with a heatsink I noticed it mentioned it was a 33mhz chip - chip id ended in 33) which is overclocked to 42Mhz. This might explain why I get regular yellow screens after reboots after it has warmed up (not a problem - I've always had them, resets solve it - I bought the board for £25 as a student in 2000).
[EDIT 2] Ha...
"but the poorer thermal characteristics of the QFP package limited it to 33 MHz and below;". That definitely explains the yellow screens.