First, I will confess that I am at the limit of my technical knowledge of the m8x24, so you will likely find one of the masters in this group to be your best source.
I have done what you propose, but for different reasons (trying a software upgrade without losing user settings).
You may not accomplish your intended goal with that course of action. While you would be keeping the same programs (assuming both software cassettes are the same build code), you would not be keeping the same user-defined variables.
Software (non-volatile firmware, no user data stored) cassette mates into the data cartridge (volatile memory, user defined variables) and then mates in the reserved 824 KSU slot(processor).
The KSU creates dialtone and operates data ports and serial busses (like if you add a 12x0 or something) and passes data sets using Nortel's protocols (like the KSU talking to a voicemail)and is your basic switch.
The software cartridge holds a Nortel-defined bundle of "features" that rely on instructions in the form of a processor program, for example, the KSU may detect some Nortel device is broadcasting a signal on port 225, but the software cartridge is needed to know it is a RAD, to recognize it as such, and to create the communications instructions needed to let the RAD talk to the right part of memory to see DN configurations, or passwords, for example.
The data cassette has the user-defined variables in its volatile memory. If you want set 221 to have line appearance for trunk 004, then that info is stored in your data cartridge.
So, if you swap sleeves or data cassettes, what you wind up with is user-defined data from KSU "A" being used to define variables on KSU "B". Unless they are identical units, you really have not gained much, even if you have the password for the other cassette. There will be a lot of re-programming. In addition, you still have a sleeve that you cannot use (the one for which you do not have a password).
Shorting the capacitors with a resistor is certainly not a habit to get into, but the practical effect is the same as pulling the cartridge and leaving it on a shelf for a month (or more or less, depending on the age of the capacitors) and letting time cause the discharge and drop the user-variable settings from the data cassette. At least then you will be rebuilding from factor default baseline. I have done the shorting solution on 6x16 DR5 cartridges and on one 824 DR5. Each time got me back to factory defaults (including passwords).
Again, I must defer to the experts. For what it is worth, my understanding at DR5 is that password data sits in the data cassette and not on the KSU mainboard.
