Developing 29F032 to DIP42 adaptor

[mikey]

Hey everyone. Just wondering if anybody can lend me a hand with verifying this, letting me know if I'm doing it all completely wrong, giving pointers etc. Basically, I'm trying to develop an adaptor for the 29F032 to conform to the standard DIP42 package like a 27C322 would use. The point is to be able to easily repair 32MBit V roms on Neo carts. That said, I've never used Eagle before, so this is my first crack at things. I had to create the DIP42 and TSOP40 packages, so hopefully I did everything correctly. I'm assuming there are likely errors that I need to fix, but need another pair of eyes at this point. Based upon my reading, I wired it up as follows:

29F032 A0 = Q15 DIP42
29F032 A1 = A0 DIP42
29F032 A2 = A1 DIP42
29F032 A3 = A2 DIP42
etc...

D0-D7 connected between the two, and the rest of the data lines on the DIP42 end left unconnected

29F032 RESET and WE lines tied to VCC, and RY/BY left unconnected (only used when writing?)

Here is a picture of the board layout so far, I'll likely clean it up a little more as I go:



If this is possible and I can get it working, I'll just upload it all to OSH Park for everyone to use. It's all based on the assumption that the 29F032 can be used in this application, but I don't see why it can't seeing as its a 32MBit 8-bit rom.

That's all for now. I'm working on this when I have spare time, so my apologies if it takes me a bit to respond to suggestions.

Thanks guys.

 

[aha2940]

I do not see the issue with using 27C322 to fix carts having issues with its 32Mbit ROMs. It would seem more interesting (to me) to develop something like this to fix carts having 64Mbit ROMs (Ganryu, Metal Slug 3/4/5, Strikers 1945+, KoF 99 & up, Garou, etc.).

Thanks!!

 

[NeoHomeBrew]

Found these adapters inside a Neo-Bomberman bootleg. Unfortunately, the first one is TSOP-48 and the second one is TSOP-56 but I hope it could help a bit.

Do you maybe also plan to make an adapter for a common programmer (like Willem GQ-4X) for the chip adapter (for re-programming the chip without desoldering)?

TSOP-48



TSOP-56

 

[mikejmoffitt]

Just a quick EE note - you might get more reliable performance at higher speeds if you change your 90 degree turns to two 45 degree ones with rounded edges. Probably not an issue at 5V with these slow speeds, but it is good practice.

 

[kraquepype]

I do not see the issue with using 27C322 to fix carts having issues with its 32Mbit ROMs. It would seem more interesting (to me) to develop something like this to fix carts having 64Mbit ROMs (Ganryu, Metal Slug 3/4/5, Strikers 1945+, KoF 99 & up, Garou, etc.).

Thanks!!

I agree here. The C322 is common enough. I'd like to see a 64mbit chip solution.

 

[mikey]

Thanks for the info and pointers so far guys, appreciate the help. As for 27C322's, they can't be used as drop in replacements for V roms because they are 16bit only, and V roms need to be 8bit. In order to use one as an 8 bit rom, you would have to add a few components and do a fair bit of wiring. Below is an example of what would be needed to use a single 27C322 for an SNES repro, which requires an 8bit rom. I started working on this because I have a Sengoku 3 cart where a couple of the V roms are failing tests, and need to be replaced. Wiring up 27C322's as replacements would be pretty sloppy and complicated, hence why I'm looking at a cleaner solution. I agree that a 64mbit drop in would be great for C rom replacements, and I'd like to see that developed. I personally don't have the knowledge to do that at the moment, but maybe down the road I'll give it a shot.

 

[bustedstr8]

Looks good so far mikey. Another note- Signal vias can be very small since they carry little current. It looks your using through hole pads right now, those can be replaced with something like 0.25mm ID x 0.45mm OD vias. Might make layout easier and help avoid the shorts in your first pic.

64Mb replacement is easily doable using two 29F032. You just need to split the file to odd / even bytes.

 

[mikey]

Alright, I cleaned up the design a bit. Here is the most recent revision. I tried to implement all the suggestions, making 45 degree turns rather than 90, smaller vias, etc. I also plan on making a programming adaptor for it as well, which means I'll need to make further changes to this pcb design to allow that. I'll work on it a bit more when I get a chance.

 

[mikey]

Ok, so I made some more revisions and created a programmer adaptor for it at the same time. I'm a little fuzzy on what needs to happen for programming requirements, so I referenced the designs of the TSOP adaptor III from buyicnow with the programming adaptor from retrohacker for my design. Basically, I used the D14 pin on the DIP42 package to route the CE pin from the programmer to the WE pin of the 29F032 for the purpose of programming (pin D14 is not used in this application because it's 8bit only). There are now jumpers on the actual adaptor that connect WE to CE for programming, or WE to +5v for reading. The 29F032's CE is tied to ground to keep it active on the programming adaptor. Does this sound right, or did I screw something up? Other than that change for programming, Everything else is hooked up normally. I couldn't think of a way to do this without the need for jumper changes, if anyone has a suggestion to fix this I'd be happy to hear it. The "mode" pin connects to the "read" or "program" pins to select which is required. The programmer adaptor has 36 pins to accommodate the GQ-4x (which I have), so you don't need the extended address header, but there are still solder pads to connect them externally.

The adaptor:


The programming adaptor: