512K Red Boards and the MBM27C4000

[+Ksarul]

A lot of homebrew cartridge images are tailored to run on the 512K Red Boards. It has been a while since there have been any changes to it (the last one added the capability to use W29C040 and BQ4015Y chips at Rasmus' request a few years ago). Fast forward a bit, as I recently added two new jumpers to the Red Boards to add support for MBM27C4000 chip. Some of the boards used in the Realms of Antiquity cartridge are ready to use it, and are the first examples of this board revision in the wild.

 

Why did I do this? A while back, I acquired a large quantity of non-JEDEC 512Kx8 Fujitsu chips (MBM27C4000), that weren't in the programmer libraries for pretty much any programmer with a reasonable price (anything under 1K). There is a JEDEC chip that is very close to it (MBM27C4001) though, and that one does show up a lot more often. I sussed out the differences between the two chips, and it came down to two pins being different, so I decided to build an adapter to try and program them as 4001s.

 

My initial tests failed--the algorithms for the MBM27C4001 in the TL866 seem to be incorrect to actually program one of them--a problem I've sometimes encountered with other low-price solutions when trying to program uncommon chips. The algorithms for the uncommon chips aren't nearly as well-tested as the common ones are. 

 

I then moved up the scale to my BK Precision 844USB programmer. With the adapter properly inserted, it quickly accepted the chips once I deselected the Chip ID box. It blank checked and programmed one of the chips on the first try--and the programmed chip worked flawlessly in the newest Red Boards, so I can now successfully program the MBM27C4000 chips and use them. I suspect I won't be needing any PDIP 512Kx8 chips for a long time now.

 

I have yet to test my rough adapter with the other four programmers I have to see how they do, but I plan to do so as time allows and will post the results here. It may be that no one else will ever need to program these, as the chips aren't common--but because of the difficulty finding compatible programmers, the prices are somewhat lower than those for other 512K chips when they do show up, and our community now has a way to use them. This may also be of use to other communities looking for ways to program these, as in my researches I found several threads on various fora looking (mostly unsuccessfully) for a usable solution.

 

I also have a completed layout for a robust, standard adapter for these which I plan on using to make a small number of EPROM adapter boards. I'll probably build some of them for others if there is interest, as I have quantities of all of the necessary parts (including high-quality ZIF sockets) on hand. It is just a matter of finding a programmer with a good MBM27C4001 algorithm. Also of interest, Fujitsu also seems to have made MBM27C2000 (256Kx8) and MBM27C1000 (128Kx8) chips which also follow the pinout differences of the MBM27C4000, making these adapters potentially suitable for programming these two additional chips as their MBM27C2001 and MBM27C1001 variants when otherwise unsupported by a programmer.

 

One other note: only Fujitsu made these MBM27C4000 chips. A couple of other manufacturers that use the 27C4000 designation for one of their chips, but they are not the same--their pinouts are different.