Another IIci ROM hack

[Trash80toHP_Mini]

IT WORKS!!!!

KAWABUNGAAAAAAAAAAAAAAAA!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!

The Pirate's of Ancient Macintosh Modification version really ought to be BLACK!
. . . w/RED LED(s)!

IMHO, of course!

If you want vinyl graphics for the Proto-Boards, I'll fire up the 'ole 40" vinyl cutting/inkJet plotter!

BTW, put me down as your first customer, I'd also suggest that you order the ROMs for the two run set if you can get a price break. Shipping the entire assembly w/unburned ROMs should be no problem and it will cut way down on end user cost by cutting out the second round of shipping costs from Mouser & Company.

If inventory cost will be problematic, just ask the PoAMM gang!

edit: a transparent black Silk Screen would be asking a LOT!!!!

p.s. TEN freakin' pages/225 replies and 2885 views . . .
. . . for a new recruit's very first post! Congrats, buddy!

[tt]

Anyway, I will reserve a board for tt just in case because I know he was interested way back when...let me know for sure tt.

The board looks beautiful! Great work. Thanks so much for thinking of me, yes I would like to buy a board, I will send you a PM.

[olePigeon]

I'd love to buy one with the pirate on it. I wanna change my startup sound to the "Arrrr!" from Pirates! Gold, and put a Pirate IIci for the boot icon.

[dougg3]

The Pirate's of Ancient Macintosh Modification version really ought to be BLACK!
. . . w/RED LED(s)!

IMHO, of course!

Black PCBs are awesome, but Seeed doesn't do them They do have white soldermask with black silkscreen as an option, though...

If you want vinyl graphics for the Proto-Boards, I'll fire up the 'ole 40" vinyl cutting/inkJet plotter!

That's pretty cool! I have to admit though that I'm not very versed in this stuff -- how do you put vinyl graphics onto a PCB?

BTW, put me down as your first customer, I'd also suggest that you order the ROMs for the two run set if you can get a price break. Shipping the entire assembly w/unburned ROMs should be no problem and it will cut way down on end user cost by cutting out the second round of shipping costs from Mouser & Company.

If inventory cost will be problematic, just ask the PoAMM gang!

Good idea! I see what you mean about taking away the shipping cost from Mouser, so I'll go that route instead. This will add about $8 to the cost of the assembled board, maybe a buck less if I can do a bigger order of chips. So a total of $23 unless I can discount it more. Who is PoAMM? I'm not sure that inventory cost will be necessarily problematic, but I don't want to order 100 ROM chips when I only have 3 people interested

edit: a transparent black Silk Screen would be asking a LOT!!!!

Transparent black?? I have nooo idea how that could even be done!

p.s. TEN freakin' pages/225 replies and 2885 views . . .
. . . for a new recruit's very first post! Congrats, buddy!

Thank you, thank you! It's been a blast doing this!

The board looks beautiful! Great work. Thanks so much for thinking of me, yes I would like to buy a board, I will send you a PM.

Awesome, thanks!!

I'd love to buy one with the pirate on it. I wanna change my startup sound to the "Arrrr!" from Pirates! Gold, and put a Pirate IIci for the boot icon.

that will rock! I'll keep you in mind as soon as I get new boards with your awesome pirate drawing (and hopefully LED(s))!

[Trash80toHP_Mini]

Who is PoAMM?

Pirates of Ancient Macintosh Modification

[dougg3]

I just got done testing out the SIMM with the SST39SF040, one of Microchip's flash chips which they did not spin off to Greenliant. It's a 4 megabit (512 kilobyte) chip, so with four of them I have 2 megabytes of ROM space available. That's the largest size that will fit in my SIMM. Luckily, everything worked A-OK! The checksum only applies to the first 512K of the ROM. I used most of the remaining 1.5 MB to contain a very large chunk of the Super Mario Bros song, and set it as the startup chime. It goes on for over a minute before the system finally boots

Here's a video of it (I show off the SIMM at first):

http://www.youtube.com/watch?v=Yen0omvBo2Y

Each chip takes several minutes to program in my programmer. I guess once I start getting up that high in capacity it isn't too surprising. In case anyone is wondering, I'm using a Sivava Willem programmer board on a Windows 7 computer with a PCI Express parallel port card. It's not officially supported past Windows XP, but I found lots of info online about it and actually ended up adapting other people's DLLs to get it to work on mine. I have a blog post about that endeavour here if anyone else needs to get it running.

[tt]

I used most of the remaining 1.5 MB to contain a very large chunk of the Super Mario Bros song, and set it as the startup chime.

Wow, that is awesome potential (great choice of musical score). It makes me wonder if a slim version of OS 6/7 could be added to the ROM like how the Mac Classic has a hidden boot disk image that can be activated by holding down some keys.

[PowerPup]

How the Classic's ROM disk image is implemented would probably require dissembling the ROM. I did find the following which sheds a little light on the subject:

http://www.eeggs.com/items/534.html

Actually, the ROM Disk in the Classic was a secret project that only a couple engineers knew about until it shipped. The reason it was added was because the ROM chip was 512K (16x256K), the original SE ROM (which this was based on) only required 256K bytes, and the engineer who did it had some extra night cycles to spend doing something fun. It was not part of any official Apple project. So this free 256K, plus around 50K from the SE photos, was used for the disk image (which was compressed) and ROM Disk driver. Apple-Talk was added as well, and the ability to save some info in PRAM. This enabled it to remember a network startup disk (plus any password required), and also saved the app name on the network drive and auto launch it (not sure if this was ever mentioned before , making a turnkey system. It was also nice in case your hard drive crashed and you want to run some repair tools from a floppy. XO was the Classic code name, from a Cognac.

[Trash80toHP_Mini]

COOL!
I just LOVE tangential topic tidbits, the most interesting information turns up!

edit: a transparent black Silk Screen would be asking a LOT!!!!

Transparent black?? I have nooo idea how that could even be done!

Bad choice of terms, substitute translucent for transparent. If Black PCBs look cool, the soldermask MUST be translucent to show the traces.

If anyone has a link to a Black PCB, please post it, my sign making/graphic design addiction just kicked back in and I'm wondering how we can use every color/color tone available on the final PCB design to enhance the Pirates of Ancient Macintosh Modification LOGO!

Post a dead on shot of the Jolly Roger side of your PCB and I'll do a little playing around in GraphicConverter.
Better yet, if you can also export some kind of Graphics or CAD file from your PCB layout of the Jolly Roger PCB design, email that and the pic to me!

. . . the SIMM should work in any of the II series Macs that have a ROM SIMM socket with the exception of the IIsi but that can be fixed by cutting a trace, and also the SE/30.

. . . and which trace might that be?

I just tested my IIfx ROM SIMM in my IIsi without damage to either, but without a boot chime from the IIsi/IIfx ROM SIMM.
Maybe I can do a jumper hack or a reversible bared trace hack to my IIfx ROM SIMM to test out your theory.

You might want to add provision for a three hole (on-off) jumper header to your Jolly Roger SIMM with a triangular/bow tie cut point for a less complicated, but readily reversible IIsi modification. Soldering a jumper wire to fix the cut would be the simple reversing procedure, while soldering in the three header jumper would make the mod process quickly reversible.

[olePigeon]

A suggestion: Name it ProAMM (like Pro/Amateur). Pirates of Ancient Macintosh Modification.

[olePigeon]

I'd like to buy one of the original boards if you're selling them. I can use it to test my soldering skills so I don't muck up the Pirate Edition.

[Trash80toHP_Mini]

Since the best price for end users will be the option of buying all parts from you for a single shipping cost anyway, I'd suggest assembly be included as well.

Amortize the cost of a Solder Paste Mask and a decent toaster oven processing unit across the first two runs. Include parts, paste, your assembly/testing/debugging and handling time, etc. for a final, ready for programming, SIMM & ROMs deal. The only variable would be shipping levels for the outer two states and foreign countries.

That way nobody need bork a socket while removing the backing plate, you avoid that extra labor for removal for basic assembly.

You could even farm out the assembly/shipping work to another hacker (NOT ME!!!!!!!!!!!!!!!!!) and concentrate on having more ROM hacking fun!
Unless you'd like the extra toy(s), of course!

[dougg3]

Well, I have some solder paste on the way. Even if I just use an iron, it may help with assembling the boards. I also got a really weird bent shaped tip for my iron, and I'm hoping it might make it possible to reach the leads. It's pretty tough, though -- once I come in contact with the plastic it starts melting pretty fast. So having someone with better tools assemble them might be the better answer. I'm sure you're also laughing at my soldering job on the capacitors on the back side I know I can do a much better job than that, but I was in a hurry to get it working and try it out

I'd like to be able to preserve the bottom plastic plate so you can't push the chips too far down--once they go too far, it's a pain to remove them with the PLCC extractor tool. I've been just barely pushing the chips in, then turning the board upside down and pushing on it, so the chips end up flush with the top of the socket. That seems to work well.

The oven thing may work, and in fact, I just recently ordered a cheap convection oven online, but I'm a little worried about melting the plastic sockets. Plus, I haven't decided whether I'm going to use it for pizza or PCBs yet

My plan is to allow anyone who just wants the bare board to get it that way if they want it, and otherwise fully assembled. I suppose it makes sense to include parts with the bare board too though, you're right...

[Trash80toHP_Mini]

The oven thing may work, and in fact, I just recently ordered a cheap convection oven online, but I'm a little worried about melting the plastic sockets. Plus, I haven't decided whether I'm going to use it for pizza or PCBs yet

Lets see if trag or techknight chime in with the skinny on any possible Dali-esque Consequences for Plastic Sockets when using the Toaster Oven Method.

Will a convection oven even work for solder paste reflow?

My plan is to allow anyone who just wants the bare board to get it that way if they want it, and otherwise fully assembled. I suppose it makes sense to include parts with the bare board too though, you're right...

As a novice level wielder of the soldering iron . . .
. . . I've been doing it a tad better than badly, since the first application of my wood burning iron . . .
. . . to Plastic Model DC Motor Connections, something on the order of 45-50 years ago!

I think it's probably best to let you do the job!

[kite210]

That video is just too awesome, I've got a quick question.

I've noticed that the 2 Nubus Power Macs I have (6100 and 7100) have removable ROM DIMMS, would it be possible to do a similar hack to change their startup sounds?

I don't know if there is any documentation on the later machines, but it seems to me that it could possibly be done.

I would look into it but I'm very educated in software or hardware hacking like this.

Anyway keep up the great work dougg

[dougg3]

Hey, thanks!

My understanding is a large chunk of the PPC Macs (Beige G3 and before) have ROM DIMMs (or do they all? I don't know for sure). I'm pretty sure it would be totally possible to do the same thing with DIMMs, although the board probably gets a bit more complicated because since it's a DIMM it has connections on both sides, which means many more traces to route. I think there's a pinout somewhere in the wiki, so it is definitely possible, it's just that the board may be more complicated than this one. I have definitely thought about it I think I have a 6100 here somewhere too, so that may be another project! In fact, looking at the hardware tech notes, Apple mentions that the 6100 has a ROM SIMM socket. No info on the pinout, but I'm sure that can be figured out.

As for models after the Beige G3 (I also have one of those, so that may be a project at some point too!), those have a ROM chip soldered onto the motherboard to my knowledge. I happen to ALSO have a Blue and White G3, so maybe if I get REALLY crazy that might be a hackable machine in the future too.

Update:

To satisfy the awesome people who want to get a board from me, I just ordered enough parts on Mouser for the first batch of SIMMs. I'm going to charge $5 shipping in the US, using the flat rate boxes from USPS. So the final price, including shipping to the USA, for an assembled SIMM supplied with four empty 4 megabit EEPROM chips ready for you to burn will be:

$28

PM me if interested...I still have 6 or 7 of the original boards unclaimed!

[Trash80toHP_Mini]

IIRC, the DIMM layout has already been done, one of the comrades has been working on the PEx ROM DIMM for quite some time. There are too many of these system boards out there that don't have the DIMM that's needed to boot them up . . .
. . . including mine!

I'm sure he'll chime in soon!

[kite210]

That would be great, then I could change the Startup sound on my 6100 to something a little better (like the 5200 sound) or even something like HAL from 2001: A Space Odyssey.

[Trash80toHP_Mini]

I'd like to put my PEx's boot and bomb chimes in the SuperIIsi.

[techknight]

You have to use a grade of plastic thats close to bakelite, it wont melt at reflow temps.

You have to think, that most boards are used in reflow ovens with plastic parts in place. So its possible, but you have to pay VERY close attention to temperatures. I have ran reflow profiles on boards up to 217c for lead free, and havent melted any plastic parts yet. They just have to be reflow grade plastics. Check the datasheets.

[bigmessowires]

Have you given any more thought to a reprogrammable version? That would be awesome, since most people probably don't have an EPROM programmer.

I was musing about this some more, and while reprogramming the ROMs from inside the Mac *might* be possible, I think initial programming wouldn't be. You can't even address the ROM SIMM unless the jumper is set, but if the jumper is set, you can't boot from a blank ROM, right?

So I think USB programming is the most likely path. You'd need to find room for a micro-USB connector, a small microcontroller, and probably a few shift registers. Assuming you used the smallest SMD packages you could find, do you think there's room on the SIMM? Or is there enough clearance to make the SIMM taller? Or you could put a 60 pin edge connector on the SIMM, and build a second board that mates with it (outside the Mac) that contains the microcontroller and USB stuff, maybe.

[bigmessowires]

Actually you could just use the existing edge connector, with one trick: connect the write-enable line to one of the VCC pins. Since write-enable is active low, when the SIMM is in the Mac it will send +5V to that pin, and writes will be disabled. But when the SIMM is mounted in a custom programmer board, it could pulse write-enable as necessary. Then you wouldn't need any physical changes to the SIMM at all, other than the routing of that one signal.

This does raise a question, though: how is having a second "ROM SIMM programmer board" really any different than just using a commercial EPROM programmer? So maybe the whole idea is a little silly.

[trag]

Miscellaneous thoughts:

1) I have a bunch of AM29F040B-70JC, AT49F002-70JC, and AT49F010-120JC on hand. The latter are good for anyone who just wants a ROM SIMM and doesn't care about experimenting. The middle one provides a moderate amount of extra program space. The first one would be the biggest, but I have fewer of them on hand. Also, the middle one has a RESET_ pin on pin 1, which is NC on the others, so pin 1 needs to be tied high if one uses the AT49F002-70JC. Ah, I see that pin1 is A18 on the 4 Mb chips. That's inconvenient.

Doug, I might be able to lower your cost a bit and possibly simplify your logistics so that you don't feel like you need to stock up on too many chips for your next batch, since I already have several hundred (111, 200, 55 respectively) on hand.

I also have about 3200 AT28C64-25TI on hand, but I don't know what anyone would do with those....

2) The X100 (NuBus PowerMac) ROM module appears to be the same as the X500 ROM module, which is the same as the 7200, ANS, PEx, and Beige G3 ROM module. There is a few pins variation in the latter to change the supply voltage from 5V to 3.3V. I've never traced the whole X100 module out, but the form factor is identical, and I traced a few pins and those that I checked all matched up.

I have designed and built about 200 of this ROM module. I have Gerber files available, but it is a four layer board, so it is expensive to have made. It operates a lot faster than the old Mac II modules and I recommend using at least 90ns Flash on board.

The X100 module is a DIMM and is 64 bits wide. So you either need eight PLCC32 chips on board, or to use wider 16 bit Flash. My design uses four PSOP44 chips which is fine for a fixed module, but not so useful for an experimenting module. I tried installing PSOP44 sockets on the thing in support of the PEx ROM project, but I can't get it to work with the PSOP44 sockets.

The pinout which is in the 68K Wiki is the one I used to build the module, so that pinout is tested, unless I mistyped something.

3) I have no idea if SM PLCC sockets will tolerate oven soldering, but they'd be pretty useless if they wouldn't. In any case, I'd get the manufacturer's datasheet and check that. That information should be readily available from the manufacturer.

4) I should have mentioned the Flash I have on hand before you got to the ordering point, but was off the forums for a while. In case other projects come up, here's an incomplete list of chips I have on hand that might be useful and are almost certainly cheaper from me than from a distributor:

PLCC32 flash/EEPROM
111 AM29F040B-70JC
200 AT49F002-70JC
55 AT49F010-120JC
3200 AT28C64-25TI

TSOP20 flash/EEPROM
300 SST39VF016-90-4C-EI

PSOP44 flash/EEPROM
250 HY29F800BG-90
450 HY29F800BG-70

DRAM/SRAM/SGRAM
2000 KM4132G512Q-10 SGRAM
310 Hitachi PSRAM HM658512ALFP (512K X (Portable memory?)
457 MCM6206BA 32K X 8 SRAM
249 M5M417800CT 2M X 8 DRAM
324 Oki MSM514400D 1M X 4 DRAM
800 ISSI 61C632A-7TQT 32K X 32 SRAM
48 Galvantech GVT71128D32T-5I 128K X 32 Burst SRAM
53 Samsung K7N801845M-QC15 512K X 18 NtNRAM

Misc.
1000 ICS9158-03CW24T-CT Clock Buffer/multiplier chips
660 NCR53C96 SCSI controllers
285 GAL16VD-15LJ (need to check part number, D = 8?)
184 DP83223V (ethernet something or other)
52 DP83840AVCE (see above)
24 Xilinx XC3090-100 PQ160C (good luck finding compatible development software)
75 AM85C30-10JC (serial port controllers)
40 Citizen W1D Notebook floppy drives

5) The Mac II family ROM looks great! I'm tempted to buy one, but I'm so backed up on projects here, it'd just be another thing laying around. Sigh. I finally started a to-do list the other day and typed out 40 items without even thinking hard. If I could get rid of the fish, the boy, the job, and stop coaching LL baseball....I would be very unhappy, but then I could catch up on the projects.

[dougg3]

Oh duh, why didn't I think to look at the datasheet on those sockets? Thanks techknight and trag! The more I think about it, it makes perfect sense--the sockets are designed to be reflowed -- otherwise they wouldn't make the pins so hard to access. The sockets I ordered can take 260 celsius for a maximum of 30 seconds. The durability is "25 cycles" -- I'm assuming that means removals/insertions. That seems kind of low...hopefully I'm not making cheap garbage that will wear out quickly.......

I did think more about a reprogrammable version, bigmessowires. I see what you're saying -- use one of the VCC pins on the SIMM for VCC, and a different VCC pin on the SIMM for the write enable line, so they would actually be broken out separate if I plugged it into an external programmer. That would work...

As for in-circuit programming, I'm guessing I would need to use a weak pull-up resistor to force WE high at all times unless when pulled low by the microcontroller...(I guess the microcontroller could just be in control of keeping the pin high when not programming, but I wonder if the microcontroller would start up fast enough to pull it up before the IIci started accessing it...)

I also thought about making a separate programmer board with a SIMM socket, but I came to the same conclusion as you -- it doesn't really get us anywhere since we can already get a EEPROM programmer board. I guess it would help because I could make custom software to use with it on Mac/Linux/Windows and we wouldn't have to worry about sockets, but the USB is just cooler. Also, it's hard to find 64 pin SIMM sockets -- none of the usual suppliers carry them to my knowledge. Looks like Tyco used to make them, and they're discontinued now?

As far as USB goes, first of all, how hard will it be to find a 5V microcontroller? It seems like everything is 3.3V nowadays. I have plenty of room on the back of the board physically for the connector, a small microcontroller, and shift registers and whatever else I need, but I don't think I'd be able to route the traces without adding more board layers. Maybe I'm just being pessimistic like I was before when routing the first SIMM, but I'm thinking it's going to be tough...most of the front side of the board is already used up, and it's going to be hard to reach everything solely from the back side. Especially the data pins, which are all near the bottom of the board. I guess I could let an autorouter try to take care of it...anybody have any thoughts on this? I really think this is where the project needs to be headed (toward being USB programmable).

I also thought about making the SIMM taller, but there's barely any clearance for anything taller in the SE/30, and I'd like to make it physically fit in all the models it supports.

Whew trag, thanks for all the info! Looks like the AM29F040B-70JC is exactly the same pinout as the 4 megabit chip I'm using, and the AT49F010 is the same pinout as the 1 megabit chip I'm using. Makes me glad for JEDEC standards Those would both work fine as far as I can tell. The middle one would probably cause too many problems since like you said it has a /RESET pin.

I would definitely be interested in anything you could do to help with the cost/logistics of the next version of the SIMM, trag...I ordered enough parts to build all of the first batch, so once I move on to the second "pirate edition" (with LEDs), I will be needing parts again.

Ah, that's great about the 6100 pinout! Hopefully it's exactly the same in which case your module would already work. That would be awesome. The four layer part of it does kind of make it difficult financially to make more I think I'm using 70 ns flash already, even on the IIci SIMM. Is the thickness still about 50 mils on the Power Mac ROM DIMM PCB?

I didn't even know there was such thing as a PSOP44 socket. Interesting...yeah, we'd need some way of making it work with sockets to be useful for ROM hacking. Thanks for all the info about the Power Mac ROM modules!

I know what you mean about being busy with all your projects. To be honest, just this SIMM project alone has kept me busy enough; I couldn't imagine juggling everything else you're doing on top of that!

[bigmessowires]

Dougg3's ROM project was featured today on Hack-a-Day! http://hackaday.com/2011/09/29/booting- ... k+a+Day%29