My first repro wont boot.

After a lot of issues with a home made tsop to willem adapter, the chip finally would ID and program. I then tried to make an Earthbound repro, for personal use only. The donor cart is Madden '96 which has 16mb mask rom with a 64k sram chip. I removed the header with snes purify, but don't think there was one on it to begin with. Programming went smooth and it verified. Soldered it in copying the orientation of the original mask rom, tried to boot her but got a black screen. Reflowed all my solder joints but that didn't help. I'm not sure what it could be really. The only thing I can think of is the rom is 24mb and the tsop is 32mb, but someone said you don't need to pad roms if they are smaller.

What tsop adapter are you using? I think type II from buyIC had a few addresses reversed.

I routed my own a year ago based on this and just finished comparing it to the type III adapter, all traces go to the same places. The flash chip is an m29f032d 70ns 4mb. Maybe it needs the rest of the hi addresses filled to work as the original rom is only 24mb. Would lunar expand work for something like that or is there a special way to do it?

I made my own too a while back....


It's been my experience that padding or filling the unused space has no effect one way or the other.

Did you apply the rom to your adapter before programming or do you program it after its been mounted on your adapter?

I like to program mine through the adapter that way, upon verification, I'll know if my mounting was good or not.

That's pretty slick how you mounted the tsop underneath like that. I'm programming mine with the tsop mounted on the adapter. Just tried to burn a madden 96 with no header and unpadded to a tsop and play it on the madden 96 donor cart. It still gave me a black screen. Next I wondered if damaged the donor cart. Soldered the original mask rom back in and it played fine. At this point I don't know what the problem could be. Has anyone else used these m29f032d 70ns with a reproduction successfully?

Do you have a pic of your TSOP board? Maybe if we could see it it would be easier to see what you did wrong.

Found the original brd file and took a screen capture of it. Also took a picture of my adapter with a m29f032d mounted.

Drop whatever solder gun you use, get something: 1. Leaded, if it isn't. 2. With flux. 3. Less heat on the gun that you use. 4. practice more on something else. It doesn't work because that solder is terrible. Probably cracked everywhere, and traces melted off the board.

Used a weller soldering station that has controlled heat. There's a lot of left over flux on there, but that shouldn't affect the functionality should it? It programmed and verified fine using a willem, which is why I'm stumped as to why it's not working on my madden '96 donor.

Try programming a smaller game, like super Mario world or all stars (not the + world one)

If your programmer verified the file on the adapter, then it sounds like it programmed correctly. Is the reset pin on the tsop pulled high?
I haven't traced your design out but, yeah, work on your soldering techniques. Your solders should be smooth, shiney, and round. I've never worried about heat though, my iron goes up to 840 degrees and I always have it on max heat. Many many years of soldering, never had a heat failure. I can't even remember the lat time I used flux (yes, I know it's your friend) but I don't use it. Tsop, dip, whatever... I don't know what to think about your situation... Maybe you have address lines in the wrong place which would explain why it would verify but not work in the cart....

Aren't both those games on low rom boards? All I have is a madden 96 donor at the moment, and actually tried to bun the same madden game to a tsop and it wouldn't boot. I'll try to redo my soldering so all the points are shinny and clean up the left over flux. I'm not sure what you mean about connecting the reset pin to high but the rest pin connects to pin 10 VCC and pin 31 VCC.

Oh, right. Hi rom.

Well, I just traced your sch out and it all looks right.... But I've never understood how these boards program with the /WE tied high. /WE has to be able to go low for the chip to enter programming mode and /WE either floats or is tired high through a resistor on these BuyIC adapter boards (and clones). I have to tie a line to my /WE to get mine to program.

Maybe email me the game file you are trying to run and I can test it on my set up....

Edit: yeah, I see the reset pin tied high so that's ok....

Whats left on there is flux that has burned, not really good flux. Also, mirror the extra ROM data correctly. EB has more security checks than any game out there.

I tested the game file you sent and it works fine.

So it's a hardware or programming issue.

Thanks for checking that Mark. This is my donor cart socket I'm using. It's not the prettiest, but something similar to this works for my tsop to willem adapter. Could those ribbon leads being too long cause issues with the snes reading the rom maybe?

I doubt that those wires would be too long for it to not even boot the game. Maybe put a heavier gauge wire for Vcc and gnd.... But I doubt that's the issue.

Since you are attaching the wires on the back side of the cart, are you sure they are going to the proper pins? It would be easy to get it upside down bringing wire from the back side and rom top side...

It's in the proper orientation after triple checking. Tried to double up on the 5v and gnd lines, but that wasn't it. Tried to use a 12v 3amp dc power supply to program the willem, didn't make a difference. I even soldered shorter wires directly to the tsop adapter instead of using the ide socket, same thing. The only time I saw something besides a black screen was with a final fantasy 3 burn. It would turn the screen to some red garbled graphics when the black and white logo screen should come on. Im loosing my mind here

I remember Ramsis had a timing issue with his Willem. His Willem timed out before the rom was programmed. He extended the timing somehow.... I don't own a Willem so I can't offer much insight.
What's throwing me is the verification after programming.
Can you look at the hex data? Is there a bunch of zeros or FF's?

It's interesting you say that because my chips erase routine times out normally for me. I have to drag the program window around a bunch causing the countdown timer to freeze and then it will erase properly. I'll go compare the original hex with a chip read hex.

Edit: Well the original 4mb Killer Instinct rom is identical to the dump that was saved from the chip. Has anyone here had success using these m29f032d chips on the tsop adapters? Maybe these chips need the 10 ohm resistors?

I think you might be well served if you tie a wire from the Willem /WE to the adapters /WE pad (opposite of the 5v line to where the resistor would be - or is). That might cure your erase issue.

I don't know if Ki has protection or not, but you might want to try a rom file that matches your host cart with no protection schemes.

It sounds like your chip is programming correctly. So the only issue I can see is hardware. Either the adapter isn't making good connection somewhere or maybe you have a solder pad that has broken loose from its trace.

Can you do a continuity test from the adapter solder legs to the carts edge connector? That'll tell you if you have any trace-pad issues. Remember that OE goes through the MAD1 chip first. Not sure if CE goes through mad1 first.... But all the address lines and data lines should go directly to the cart edge.

Regarding the m29f032d. They made that flash rom in a reverse pinout. You might want to verify that your chip ISN'T the reverse pinout version. Your flash rom shouldn't need any resistors except the pull up resistor on the WE line. But since there is no connection from the programmer to the WE line, putting a pull up resistor there will keep the chip from going into programming mode. Make sense?

Continuity test checks out. Wonder how I would test if it's a reverse package. Should an ohm test with an unmounted tsop on the two VSS pins show continuity?

Edit: Regarding the WE line, yes that makes sense, though pausing the erase window timer seems easier than tying a wire from the willem to the adapter. Forgot to say this was on a KI donor board I didn't want to mess with, but was too anxious not to. Just tried to test the two VSS pins on the right side of the tsop and they have continuity, so presumably this is not a reverse package.

Just look up your exact part number on google. Get the datasheet. Check it out.

Mail me on of your adapters (if you want). It would be a good idea to have someone else check it out too.

Thanks for all the help you offered Mark. Figured out what it was, had my a21 and a19 lines swapped on my tsop programming adapter.