I don't think you have been reading my post.
Anyway, part of the address decoding is different in the RAM and I/O part of the -002 and the -003. Only the ROM part is socket location decoded. Moving the -002 to the -003 location does not change the chips RAM or I/O location decoding, that is done in the chip.
You can see a picture of the two boards in post:
https://forum.vcfed.org/index.php?threads/debugging-a-kim-1-computer.55589/page-2 #38
They essentially fit under the 6532 chip. It does raise the 6532 by about 3/8th inch or so.
I recommend reading the entire post from end to end.
This arrangement uses an EEPROM to replace the ROM part of the replacement. The address changes for the I/O and the RAM are done with the PAL chip.
The EEPROM is programmed by the debug board. in two steps. The first step is to load the code into the KIM1's RAM of the code that executes the programming. It is run, using the next block of EPROM data to initialize the EEPROM that is under the 6532.
The EEPROM I used was intended to be mounted with the leads of a flat pack. This was too wide so I fold the leads under the chip, like a PLCC. I planerize them and pre-tinned them.
In the kits, I did the step of soldering them to the PCB, as the spacing is quite small. I left the construction of the rest of the circuit to the person that bought the kit.
Like I said, to load the EEPROM, you need the debug board. It can be done with the proper attachment of the flat pack to a regular programmer but I do not use the addresses in typical order as the routing of traces was way too constrained. I don't recall if I swapped data bits either. Remember, it was intended to be programmed in place. With the debug board. I don't know if I still have the PAL code or if any of the PALs I have in my box of parts have the correct code.
I do have several of the prototype under 6532 chip boards that is known to have one of the top, easy to get at traces missing.
As I mentioned before I have several of the prototype debug boards that would need a three pullup resistors added and the edge connector pads scrapped clean for the bottom side. ( both not too hard to deal with). I shipped the kits with an edge connector but I don't have any spares. These are hard to find.
If you are interested, I can send a blank EEPROM, programmed PAL ( believed to be correct ) and the two boards.If you want to do both 6532s you'll need two sets.
I would like to recover some of my cost. I'm not in the business of doing this. I created these kits that I sold at close to my cost for the creation of the kits and did not add extra cost for my prototypes, since I needed to fix my board as well.
You can PM me to determine what you expect of me and my time.