Making a cart from scratch...

Are you new to 6502, NES, or even programming in general? Post any of your questions here. Remember - the only dumb question is the question that remains unasked.

Moderator: Moderators

the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

Awesome! Thanks! While I was waiting for a response, I just tried out a different ROM with different chips. I admit to going a bit more advanced than the NROMs I'd done before, choosing to put Dragons of Flame on one of the MMC1 boards. I already knew the ROMs worked fine, so I burned the two ROMs (PRG and CHR), added all the chips, placed caps, diodes and resistors (save for the 27 OHM resistor that Muramasa's board description lists as optional). Solder pads are connected appropriately (CHR ROM is selected instead of CHR RAM, PRG32 is used because it's a 32 pin chip, etc). However, the game does NOT boot. I have not installed a CIC yet, as my lockout is disabled on my main system. The only thing I can think of is possibly a bad WRAM chip, but if my understanding of it is correct that shouldn't affect the game loading, only the ability to save. Another possibility is that I can't see a solder pad for mirroring, but I'm assuming this is because the mapper itself deals with it (Bootgod's database shows that to be the case). Bootgod and Famirom both point to it being an SKROM board, which is confirmed compatible according to Muramasa's product page. I tested the patched ROM, and it works fine, and as far as I can see all my solder points are fine, and nothing is accidentally crossed. Unless the board needs a CIC to boot, which I highly doubt, I can't see a problem that would cause it to simply not load. Any suggestions from the pros?
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

the_wizard_666 wrote: Fri Mar 26, 2021 6:08 pm The only thing I can think of is possibly a bad WRAM chip, but if my understanding of it is correct that shouldn't affect the game loading, only the ability to save.
Due to the reduced amount of memory on the NES console, often games use the extra cart RAM for more than just saving. You may be able to specify a NES2.0 header to disable cart RAM and see if you can get the same crash in an emulator.
Unless the board needs a CIC to boot, which I highly doubt, I can't see a problem that would cause it to simply not load.
No games need the CIC. The top-loader doesn't even have a CIC.

Exactly two games (mapper 37 and mapper 105) are known to use the CIC as a source of a reset pulse, and both are very rare.
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

Well, I confirmed the RAM chips are indeed the correct chips, as it turns out my burner can also be used to verify RAM chips (even though they can't do anything WITH those chips). So I confirmed they are the correct chips, so that shouldn't be the problem. I also confirmed while cleaning up a bit that I actually DID buy the 27K ohm resistors, and accidentally used those instead of the 10K ohm ones, so I removed those, put one in the 27K slot, and installed the correct resistors. Sadly, that still didn't fix the problem, as the game simply fails to load. I'm thinking maybe it's an issue with the ROM. If I accidentally burned the wrong file to one of the chips, that could possibly cause the problem. Aside from that, I can't see any obvious problems with anything else. Attached are photos of the board, alongside an unused one for comparison. Like I say, I don't see any obvious issues, none of the contacts are accidentally combined, all the pads are correctly merged to the best of my knowledge...I really don't know what the problem could be aside from possibly burning the wrong files, which I highly doubt I did (but am not gonna rule out until I can desolder the chips to confirm).
IMG_20210326_210102538.jpg
IMG_20210326_210116424.jpg
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

Unfortunately, it's really hard to narrow down all the places that something could be wrong, especially when if you don't know that you can trust your PROMs. I'd suggest buying a couple modern 32-pin Flash ROMs and using them with sockets before making more-nearly-permanent reproductions using UVEPROMs.
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

You underestimate my determination and sheer stubbornness :lol: The thing is, I already have the stuff and a week off with nothing to do aside from my cousin's funeral, so I'm not too worried about making mistakes - especially ones I can fix in due course. All the boards I've got that didn't work will be stripped, cleaned, and reused, so the only thing it's costing me is time and solder. Practice is what will get me there, and you guys have been great for me when problems do arise. At least, it gets me thinking of things I didn't see before.

I tested my process with a simpler ROM, as my buddy wants a copy of Super Bat Puncher. That one has issues, but actually boots, so at least in that regard I think the problem lies with the files themselves and not the hardware. So while I still need work, I am close to getting it working. Probably shoulda started there instead of a more complex game with a battery, but whatever. I'm confident I can get this going. My buddy has more experience with repros than I, and I'll be heading there tomorrow, so I may have more success with an aide :P

Anyway, I do appreciate everything. I'm definitely making progress in a short span of time, and I'm learning so much here. It may be annoying and old hat to you guys, but it's new and exciting for me. I fully expect a few hiccups along the way, and I appreciate y'all tolerating my newbie mistakes.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

the_wizard_666 wrote: Fri Mar 26, 2021 9:19 pm I tested my process with a simpler ROM, as my buddy wants a copy of Super Bat Puncher. That one has issues, but actually boots, so at least in that regard I think the problem lies with the files themselves and not the hardware.
Super Bat Puncher has been around for a looooooong time now so if you're having problems it's probably the PROMs again.
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

Well, with the NROM board, it turned out to be an incompatible chip causing the issues. I'm using 27C010 chips, which should be compatible with the board according to the specs on the website. Basically it's booting up, but with graphical issues, and it freezes after hitting start. The bat animates on the title screen. I've used the PRG32 pads because it's a DIP-32 chip, and the CHR RAM pads.

What I was referring to was that I think that I may have burned the wrong file on the Dragons of Flame cart. Super Bat Puncher only had one file, so there was no way to confuse it, so burning error is something I can safely rule out on that one. Sorry I wasn't so clear on that.
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

UPDATE: All issues solved.

Super Bat Puncher required two RAM chips, not just one. Adding that in solved the problem.

Dragons of Flame was indeed a badly burned ROM. I re-burned the chips and it also worked flawlessly.

Moved on up to an MMC3 game and built Destiny of an Emperor 2. Everything works except that the saves keep getting corrupted. 99% sure it's because I don't have a 1000pf capacitor, so I'm gonna order those, add one in, and try again. Aside from that, no issues.

So yeah, all that's left for me to get to testing is the VRC boards, but given that I've solved everything else, I'm not worried anymore. Thanks again for everything, the hardest part seems to be troubleshooting, but now that I know the things I've learned, I don't forsee any more problems, at least until I start doing crazy stuff :p

One quick unrelated question though - I'm considering using an After Burner cart for a Maharaja repro. I'm wondering if I have to rewire the chips like other donors, if it's wired differently, or if no rewiring is necessary. It's not something I'm planning immediately or anything,,, but something I have in my project queue. Thanks!
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

the_wizard_666 wrote: Sat Mar 27, 2021 11:22 pm Everything works except that the saves keep getting corrupted. 99% sure it's because I don't have a 1000pf capacitor, so I'm gonna order those, add one in, and try again.
Battery backup is more fragile than other things. The capacitor's absence is a likely culprit, but other things can go wrong too.
One quick unrelated question though - I'm considering using an After Burner cart for a Maharaja repro. I'm wondering if I have to rewire the chips like other donors, if it's wired differently, or if no rewiring is necessary.
If a game was not manufactured by Nintendo, and if a ROM was not exactly 128KB, it usually uses a standard pinout that's compatible with UVEPROMs.

Correspondingly ... if a game used a ROM that was exactly 128KB, it was most often in a 28-pin footprint that will require rewiring. Or if a PCB was manufactured by Nintendo and the PCB could accept a ROM that was 128KB or bigger, if often used Nintendo's own backwards-compatible footprint.

Note that the US release of After Burner only has solder mask on one side, has no place for the save RAM or the extra support hardware for a battery, and uses two 28-pin 128KB mask ROMs for CHR, so reproductions reusing that board definitely counts as "advanced".
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

Quick update on Destiny of an Emperor 2 - capacitor is now installed, but the saves still corrupt. Additionally, in the process of attempting to figure out what's wrong, I tested the battery (despite it being brand new as of installation) and it registered 2.92. When installed it registered ~3.2. So basically the battery drained in a month of being simply installed on the board and not being used. I would expect this level of drain after 20 years, not 20 days, so there's gotta be something wrong somewhere on the board...I just haven't the slightest idea where to begin to look. And since it was mentioned that there could be all sorts of issues with battery backup, I figured I'd check in with the new info to see if anyone knew where I should begin.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

Check the voltage across the 1k resistor. It "should" be a couple millivolts, but if it's higher then something's gone wrong.

Then check the voltages on the SRAM pins 28, 26, and 20 (relative to pin 14).
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

lidnariq wrote: Tue Apr 27, 2021 9:45 pm Check the voltage across the 1k resistor. It "should" be a couple millivolts, but if it's higher then something's gone wrong.

Then check the voltages on the SRAM pins 28, 26, and 20 (relative to pin 14).
Testing on the 200k ohm setting gave me 27 when testing the resistor. On the 2V setting, pin 14 to 28 gave 1.966, to 26 gave 1.45, and to 20 gave 0.93. I'm not sure what numbers should be within the target range, or whether I used the ideal settings, as I typically only use the multimeter to test batteries and traces, which is why I'm giving the settings I used as well as the result.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

the_wizard_666 wrote: Wed Apr 28, 2021 3:37 pm Testing on the 200k ohm setting gave me 27 when testing the resistor.
Not the resistance of the resistor! the voltage across the resistor. The battery is supplying current to the RAM through that resistor, so there will be a voltage drop. 30 millivolt ÷ 1 kiloohm = 30 microamps.

The current should be 10 microamps at most (and preferably 0.2 microamps); anything else indicates that the RAM is either a too thirsty model, or the RAM isn't sufficiently deep in standby.

Point of confusion to me: on the OEM TKROM boards, the resistor is 1kiloohm, but the silk on Muramasa's TxSxROM board instead marks a place for two 10kiloohm resistors. I don't really know whether that could possibly be a problem.
On the 2V setting, pin 14 to 28 gave 1.966
The voltage on the RAM should be close to the voltage on the battery ... as long as current is low enough. Sure sounds like it isn't.
14 to 26 gave 1.45
That's right out. Pin 26 should be driven low strongly by the MMC3 to put the RAM deeply into standby. Voltage should be 0.2V or less.

Do one of the resistors connect pin 26 to ground?

On the OEM board, the resistors should be:
* 1k R2 from diode to RAM pin 28
* 10k R1 from RAM pin 26 to ground??? I think?? I don't actually have a battery-backed MMC3 cart to check
* 27k R3 from M2 to ground??
and [14] to 20 gave 0.93.
That's ... a little weird but not implausible. It could be floating, I suppose.
the_wizard_666
Posts: 73
Joined: Thu Feb 25, 2021 6:52 am

Re: Making a cart from scratch...

Post by the_wizard_666 »

Well, I also appear to have tested the wrong resistor anyway. I looked at a clean board to try and follow the traces. There isn't a 1k resistor on the board. There's a pair of 10k resistors and a 27k. I tested the 27k resistor. Following the traces, pin 28 appears to connect to the 22uf capacitor at C4, as well as pin 1 and the VCC/Battery solder pad. Pin 26 seems to go through the 10K resistor at R2, and pin 20 goes to the MMC1/MMC3 solder pad. 14 goes nowhere, so I am assuming that's ground. If I'm wrong, I'm sure you'll let me know :P

So adjusting for this info and the info you provided, I tested using the same 2V setting on the 10K resistor in R2. The reading was .514. Not sure if that's good or bad. I also tested the battery again and it's at 2.89. I replaced it last week and it tested 3.2, so this is definitely looking like thirsty RAM. I'm going to swap the chip with the CHR-RAM (they're different models) and see if that makes a difference.
lidnariq
Posts: 11432
Joined: Sun Apr 13, 2008 11:12 am

Re: Making a cart from scratch...

Post by lidnariq »

the_wizard_666 wrote: Wed Apr 28, 2021 4:44 pm Following the traces, pin 28 appears to connect to the 22uf capacitor at C4, as well as pin 1 and the VCC/Battery solder pad.
Ok, that's fine. What happens if you then follow it past the solder pad?
14 goes nowhere, so I am assuming that's ground. If I'm wrong, I'm sure you'll let me know :P
Correct. That's why I'm saying to measure things relative to it.
So adjusting for this info and the info you provided, I tested using the same 2V setting on the 10K resistor in R2. The reading was .514. Not sure if that's good or bad.
Definitely should not be that high. There shouldn't be anything other than the MMC3 to pull the pin high.
definitely looking like thirsty RAM.
What's the specific part numbers you're looking at? That should specify whether they're thirsty or know when to stop partying.
Post Reply