Posts: 217
Threads: 4
Joined: Mar 2010
The RAM chip used in there is a standard pinout 2Kx8 CMOS static RAM so there are many other type that would be suitable replacements. From the datasheet I found online for the TC5516 it would seem to be a 250nS part, so any 2K part with an access time equal to or faster than that would be suitable. For instance I have in front of me a Sony CXK5816 and a Fujitsu MB9416 and looking at the specs for them either one of them would be suitable.
As for a schematic, there only seems to be the 4 SRAMs in the module so I would suspect that all the address and data lines and R/-W are connected to pins common to all 4 chips and then there would be separate -enable lines for each chip. Since there is no glue logic in the module It seems that the -enable for each chip comes from the main logic of the 75, likely from the chip called 'MAR' this chip seems to provide all of the RAM control logic, and the SRAM chips on board are also 2K x 8 so it would seem likely that this chip divides the RAM space into 2K pages and provides -enable lines for each of them. unfortunately the copy of the service guide I found is missing some of the schematic pages so I cannot confirm that.
Since the -enable is supplied by the 75 this raises the possibility of the problem being inside, or perhaps even a connection issue between the module and the main board, did you try reseating it? It is possible, depending on how the 75 'discovers' the RAM that only one chip is at fault. If it reads through the memory until it detects that it does not read back what it wrote and from that assumes it is the end of memory then if the 3rd 2K chip was bad you would still loose 4K.