Forums

Sega Master System / Mark III / Game Gear
SG-1000 / SC-3000 / SF-7000 / OMV
Home - Forums - Games - Scans - Maps - Cheats - Credits
Music - Videos - Development - Hacks - Translations - Homebrew

View topic - New Master system 2 mainboard

Reply to topic Goto page Previous  1, 2
Author Message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Mon Sep 21, 2020 7:10 am
Last edited by L10N37 on Sat Oct 03, 2020 5:01 am; edited 4 times in total
Found a couple mistakes before these even arrived, so won't be mucking around with traces.

Made a few adjustments. Changed slide switch footprint. Found slide switches on ebay that had the feets pitch noted and appear to be the correct height.

Used a larger 0402 footprint because these would be next to impossible to solder to.

shoved the power pack barrel across a couple mm
getting there. I knew the first boards would have mistakes :)

Moved the Power LED down further because I'd rather a light on the front of the shell than the top. Would reach either



I also found most revisions use 330ohm resistors on the controller inputs, not 3K3 as found on the Brazilian schematic. Not sure if that schematic is correct.

** you would have to remove the little piece of shielding under the input capacitor to prevent S/C

**Due to different DB9 connectors mounting holes being a little closer to the edge of the PCB you'd probably be best off taping this section off to prevent controller lines shorting to the shielding. (tape over shielding)

**The RF hole isn't as large as the A/V output used, you'd have to dremel this out a little larger

**Takes a different size centre positive barrel for power, I have a multi voltage one with different tips to swap out. Unsure yet on its exact size, but this was a mistake. Doesn't make a difference to me anyway as I plan on running it on this pack at 6v


Z80 pictured is an I/O controller not a drop in replacement CPU. Didn't check the markings properly and googled Sharp Z80 CPU thinking it was just a drop in replacement capable of higher speeds

  View user's profile Send private message
  • Joined: 22 Sep 2019
  • Posts: 98
Reply with quote
Post Posted: Mon Sep 21, 2020 7:15 am
This is cool, but if you were ever planning on releasing the design I'd remove the Sega logo for sure.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Mon Sep 21, 2020 7:55 am
There are bypass boards with Sega logo on them? I doubt I'd get in any trouble... it just looks more original as that's where it is on the original front silk

**Actually, maybe they removed it before release too
photos exist though
  View user's profile Send private message
  • Joined: 25 Jul 2007
  • Posts: 716
  • Location: Melbourne, Australia
Reply with quote
Post Posted: Mon Sep 21, 2020 10:22 am
That's a nice looking proto board you got there, looking forward to future updates.
  View user's profile Send private message
  • Joined: 02 Mar 2011
  • Posts: 165
  • Location: Valencia,Spain.
Reply with quote
Post Posted: Mon Sep 21, 2020 4:20 pm
I have been following your topic, and I have a question L10N37

Would it be possible to buy one in the next future?

Thank you.
  View user's profile Send private message
  • Joined: 21 Oct 2015
  • Posts: 303
Reply with quote
Post Posted: Tue Sep 22, 2020 8:35 pm
Really nice this board.
In my mind lives an idea about MS mini, is it possible to make it smaller? and maybe a guy who has knowledge about 3D modeling...
:D
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Wed Sep 23, 2020 2:47 am
Last edited by L10N37 on Wed Sep 23, 2020 2:48 am; edited 1 time in total
Aranya wrote
I have been following your topic, and I have a question L10N37

Would it be possible to buy one in the next future?

Thank you.


I didn’t make this for money, but it would be nice to break even for what I’ve spent prototyping. I will change the sega logo to “Enthusiast” as I decided to call the board this.

I will happily ship out bare PCB’s, you don’t need to necessarily use the components I have.

The EMI filters would be ugly had I chosen the footprints to suit the available eBay ones at the correct 270pf I will be using 220pf Murata ones and this should be ok. Aesthetics were important.

Anyway, parts are taking a long time to arrive from overseas (Sega ones)

I don’t want to pull ICs off my complete segas

When the Enthusiast V1 is ready you’ll know :) they won’t be expensive.

I will still have to do extensive testing, and leave the console powered on a full 72 hours or so and monitor
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Wed Sep 23, 2020 2:48 am
law81 wrote
Really nice this board.
In my mind lives an idea about MS mini, is it possible to make it smaller? and maybe a guy who has knowledge about 3D modeling...
:D


Yes, you can get smaller equivalent IC”s of some of the parts and squeeze this onto a 4 layer board. It could be quite tiny
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Wed Sep 23, 2020 7:06 am
So I assumed the data sheet was looking at a top down view for PCB placement.

The pinouts on the AV out are backwards.

i.e

should be

1 2 3
4 5
6 8 7


not

3 2 1
5 4
7 8 6


Which i just confirmed with a meter on continuity while playing with a video circuit/ scart cables

NOT HAPPY
so at the moment that section is not right.
Something so simple I should have confirmed - i just copied the numbering straight from the data sheet as is, onto my custom footprint (which fits the component perfectly)

I can still test the boards out properly and just use a panel mount DIN
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Wed Sep 23, 2020 11:26 am
SO i've now got a video circuit I can adjust to suit the master system board.

I made this for Megadrive version 1.

Even then you still have to cut clock traces to clean up the jail bars

I don't have issues with the model 2 and cheap SCART cables (except muddier sound)
Anyway off topic

But i will be testing it with this

It doesn't need to be AC coupled on the input but it just helped with a tv I was RGB hacking, at least I think it did. IT probably only helped because my external prototype has something wrong with it.

Channels now balanced to .7vpp





I forgot to get 0805 75ohm resistors lol
zm1U1HD.jpg (322.97 KB)
zm1U1HD.jpg
jPOWjCm.jpg (252.85 KB)
jPOWjCm.jpg

  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Mon Sep 28, 2020 3:24 am


2 brand new NOS I/O controllers


waiting on:

new VDP

BIOS flasher (it was an extra 20-30 bucks from Australia, so i wasn't in a rush and saved money by getting it from taiwan)
z4gJkX5.jpg (224.91 KB)
z4gJkX5.jpg

  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Oct 01, 2020 9:26 am
https://oshpark.com/shared_projects/Ntagdloi

considering adding this circuit to the board

https://lh3.googleusercontent.com/-Rkf9zlBxtN0/VCrAFIR8t0I/AAAAAAAAJfQ/HM2uyirYd...
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 5:43 am
Quick update: I ran out of patience and used a board to pull a VDP off to transplant.

I quite possibly fried all my ram by using what I thought was a pin compatible drop in Z80 CPU- This I/O controller *facepalm* actually has it's ground pin where the CPU has +5v. I was checking everything out with a meter which looked good when i noticed the *I/O controller* was heating up.

I switched this out for a zilog from the donour board

I get scrolling graphics (but its a garbled mess) with the original alex kidd bios in a socket

I get different garbled graphics with an MC27C256B programmed with the EURO sonic BIOS - I am using my graphics board for the megadrive as is which seems to take .5vpp in and then .5vpp back out by the time it goes through its dividers into the set. (this is only .2vpp under RGB standards)

I may have fried my RAM as this put 5v across data lines to the RAM/ VDP etc when i plugged in the wrong Z80

I used the 50/60hz switch which works - it speeds up and stretches to fill the screen @60hz

I am looking at RAM issues and hoping my changes to the original VRAM/ VDP interface aren't the issue - unfortunately I used my only sockets on another board leaving me ONE left for the BIOS and I should have removed these so i could easily swap but the new ones are hardwired into my other proto board. (reason being jumping the gun - its impossible to solder the 22nf 0402 components with sockets in the way)

* I did find one missing trace for CONT as the official schematic labels them differently on each. It's just one trace from I/O controller to cart slot so a non issue for testing.

I also lifted pin 1 of each RAM/VRAM to leave floating - as these are NC on SRAM and +5v on original
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 5:55 am
If you’re interested in the garbled mess- it seems to at least be running Alex Kidd intro in background with no sound

https://imgur.com/gallery/KfY4aUv

The VDP transplant was a bit messy but pretty sure it’s all connected properly

It’s also responsive to the controller and changes to vertical scroll as it would when you run off the edge of level 1
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 6:01 am
https://imgur.com/gallery/X0QEZb4

And sonic bios which seems to lock on this screen or something similar and not respond to controls.

I would happily heat gun the old ram out and swap it all from donour board to confirm but my SMD components will come off too- 0402 was an absolute pain to solder

Shall see ...
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 6:58 am
Audio fixed, i went off a version 1 schematic with a different VDP
this has a 1.5k pull up as well as the 2.2k pull down.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 7:03 am
https://imgur.com/gallery/Krhgn8a

Audio
Un8vUnr.mp4 (5.91 MB)


  View user's profile Send private message
  • Joined: 05 Nov 2014
  • Posts: 435
  • Location: Auckland - NZ
Reply with quote
Post Posted: Sat Oct 03, 2020 7:04 am
Any of the bioses later than the JP 2.1 one use a mapper, so the crashing on the sonic bios would be as a result of missing mapper and rom, as its also bigger than 256kbit.

Does it boot a game if you put one in? Try something cheap or easy to replace rather than an everdrive incase there are still problems and it kills the game.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 7:47 am
wasup wrote
Any of the bioses later than the JP 2.1 one use a mapper, so the crashing on the sonic bios would be as a result of missing mapper and rom, as its also bigger than 256kbit.

Does it boot a game if you put one in? Try something cheap or easy to replace rather than an everdrive incase there are still problems and it kills the game.


no luck on loading a cartridge either!
It doesn't continue to alex kidd either - it just stays blank.

I actually ditched my initial part of the schematic for one that was already somewhere on this site and just relabelled the slight anomalies. (i.e reset as oppose to rst and so forth) for the cart connectors.

theres something intermittent happening because I could get in and "play" alex kidd before, then it started freezing and unfreezing when I used directional left, sort of looping me in the game from my initial position back to the start - but was fine to jump and punch on the spot. Now it doesn't seem to want to let me move at all.

I am a bit suspect on the I/O controller
But at the same time good chance I've mixed some pins or something - I was hoping for a more functional prototype than this

I will have to sit down and go through every leg on every IC when I get time and just make sure it's all OK
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 7:48 am
Also the BIOS chip i used is 256 kilobytes the same size as the BIOS/ ROM for sonic. **wooops its kilobits my bad. Stupid thing let me flash it
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 8:18 am
OK i flashed the Japanese 2.1 as you mentioned.
it loads straight up and play some funky music with a full screen of corruption.

So the boards not that far off being polished up and fully working


Ok guys so that's where it's at

i'm not in the position to keep prototyping at this stage as they laid off 50 techs last monday... I work in telecommunications. Bloody covid.

Once i sort some things out I will get back into the project in my spare time. I may check over the schematics in my spare time but I won't be ordering any boards.

* oh its space harrier, how could i forget !
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 03, 2020 10:49 am
I have decided to share this project - all I ask is that you sign up to these forums if you find this from outside of here, document your findings in this thread/ list the mistakes you find and keep to the original side loader design to have sd card loading out the side.

This is a drop in replacement for SMSII shell, and fits straight in - only changes necessary are minor ones to the RF shield to stop power shorting and maybe some tape under controller inputs.

Sound is working but I am missing a 1.5k pullup, I linked one across from the 5v pad to the 2k2 audio pulldown on my prototype- the small cap is just for frequency filtering. I can't really tell the difference with the master system being unfiltered. Audio levels are a little low, but always were. It's a clean output though. The positive side of the cap should be marked on the silk screen. (R/H side of footprint)

As documented the AV out is back to front, I needed to use flipboard view to copy it across from the data sheet, and didn't.

I made the footprint to fit this part

DELTRON COMPONENTS 671-0801 DIN Audio / Video Connector, 8 Contacts, Jack, PCB Mount, Through Hole Right Angle

I've spent a couple hundred bucks on this project and a lot of time but I've reached my limit for now.

Most of the vias are to flood fill with ground every single little space I could. When I downloaded this file from my other PC the flood was missing but the outlines of the flood are there ?? I had this same problem downloading other peoples boards.

Finally, a lot of final stuff was done without using a schematic - the power section - the switch - the AV port - this was all drawn in manually with no rats nest. The schematics a bit of a mess but I was just learning. It will give you a lot of DRC errors but they really won't affect functionality.

Also couldn't find a suitable switch the was the same as the original for the power

AND there's some weird mirror duplicate site with this thread on it - this wasn't me. Just found it on google. IT claims the ADMIN/ site director of that site made this thread and it's stolen the SMS power logo etc... it's arduino.melbourne slash something or rather -


https://drive.google.com/file/d/1qgO7E5FnfbNRgPZTCzKDhODYyhTv-RnW/view?usp=shari...

Download entire file and decompress into a folder.
Made with Kicad
  View user's profile Send private message
  • Joined: 23 Apr 2014
  • Posts: 131
  • Location: NYC
Reply with quote
Post Posted: Mon Oct 05, 2020 1:46 pm
Hey, awesome project! If you don't mind, I have a few questions:

- Would you consider this open source, or are you just sharing privately?

- The gdrive link is private, so I couldn't take a look.

- Based on the picture, this seems like it's designed for only an Everdrive. Is there a version that also has a top cartridge slot? I think an fun idea might be both: Have the Everdrive on the inside, but if a real cartridge is plugged in, it boots to that instead of the Everdrive.

Also, that clone SMS Power site is so weird!! I've seen a lot of stolen work lately (even more than usual), but that's the first time I've seen a cloned forum with the same name on it.
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2007
  • Posts: 716
  • Location: Melbourne, Australia
Reply with quote
Post Posted: Tue Oct 06, 2020 9:25 am
L10N37 wrote
AND there's some weird mirror duplicate site with this thread on it - this wasn't me. Just found it on google. IT claims the ADMIN/ site director of that site made this thread and it's stolen the SMS power logo etc... it's arduino.melbourne slash something or rather -


No conspiracies or little green men to be found here. What you found was a mockup that was intended to demonstrate porting the site to another platform. Rather than using Lorem Ipsum I copied some of the top posts at that time, one of which was yours.

I'm puzzled as to why Google indexed the site.
  View user's profile Send private message
  • Joined: 29 Oct 2018
  • Posts: 16
  • Location: New Zealand
Reply with quote
Post Posted: Tue Oct 06, 2020 9:53 am
djbass wrote

No conspiracies or little green men to be found here. What you found was a mockup that was intended to demonstrate porting the site to another platform. Rather than using Lorem Ipsum I copied some of the top posts at that time, one of which was yours.

I'm puzzled as to why Google indexed the site.

You don't have robots.txt or a robots meta tag that says not to index it so Google will index it
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2007
  • Posts: 716
  • Location: Melbourne, Australia
Reply with quote
Post Posted: Tue Oct 06, 2020 10:00 am
elephantflea wrote

You don't have robots.txt or a robots meta tag that says not to index it so Google will index it


Whilst that is true, Google only indexes sites that have been linked from other sites, that is the pinacle of how it's page rank algorithm works, and I've never linked it anywhere.
  View user's profile Send private message
  • Joined: 14 Apr 2013
  • Posts: 623
Reply with quote
Post Posted: Tue Oct 06, 2020 1:23 pm
djbass wrote
elephantflea wrote

You don't have robots.txt or a robots meta tag that says not to index it so Google will index it


Whilst that is true, Google only indexes sites that have been linked from other sites, that is the pinacle of how it's page rank algorithm works, and I've never linked it anywhere.

Doesn't it also index what gets entered into Chrome?
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2007
  • Posts: 716
  • Location: Melbourne, Australia
Reply with quote
Post Posted: Tue Oct 06, 2020 1:53 pm
No idea, anyway sorry for derailing the thread, lets get back on topic.

Hope L10N37 can back on his feet and continue this awesome project.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Oct 08, 2020 1:37 am
Guys, the issue with prototyping is the stupid Quad Inline package VDP. With no sockets available, it makes this super tedious

The ones on ali express aren't exactly cheap, and i don't like destroying perfectly working master system boards to steal organs from! (as in, its not exactly easy keep every leg unbent, you won't really be able to transplant these multiple times)

My new one hasn't arrived but I'm tempted to use that to repair the old donor board - I'm not sure yet. This is why im sharing the project. I have done a quick look over the cartridge issue and I can't seem to find a reason for them not to load when the BIOS does.

We aren't dealing with DDR RAM or anything either so i really can't see why my method of re-assigning address and I/O data lines for a tidier BUS could be the issue either.

The last master system board I saw on ebay was over $50 alone and in this day and age with the retro collectors and retro scene in general just blowing right up its not really a far fetched price either.

Bob, thanks for the heads up! I will fix that now.
Don't be too judgemental - I could go back and make a super neat schematic but it won't make a scrap of difference. Once the entire boards flood filled its very neat.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Oct 08, 2020 1:44 am
Last edited by L10N37 on Tue Oct 13, 2020 3:13 pm; edited 1 time in total
*removed*
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Oct 08, 2020 1:50 am
I’m not so stressed out that I can’t relax anymore.

I decided I’d sit down and map out the cartridge connections soon.

I can use an existing genuine mainboard for any confusion.

Should be easy to fix with schematics and continuity on a meter.

Will be super careful this time and just not seat the new VDP all the way down when it arrives.

As for VDP/ VRAM- suppose I will re-do schematics and just see if I can squeeze something neatly in there.

Really wasn’t a bad effort for how long this took.
Don’t expect updates as you know the stage it’s at! It’s simply not reading carts and displaying garbled graphics.

Not too much to rectify really
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Fri Oct 16, 2020 4:24 am
OK,
I mentioned not updating
But the cartridges not loading is an absolute facepalm of a mistake

I was just working on another project where i labelled cart pins naturally as

1 to 30 - and on the flipside 31 to 60

When I thought hang on - the master system counted pins differently
So i double checked
It actually doesn't

Its 1 to 25 on one side
25 to 50 on the other

for some reason counted it as

1357 rear row
2468 front row

Etc.

as oppose to

1234:25
5678:50

**********************
jumped the gun on this conclusion - seems everyone labels them differently

1234
8765

1234
5678

1357
2468

Guess numbers aren't important, but labels are / where they connect.
******************


I did jump on quickly the other day and tidy some things up, fix the AV out and the sound circuit....

Will be changing the 0402 to 0603's. The 0402's are solder-able but it's overkill - found some 5% tolerance 0603's. Same goes for the XTAL caps.

So really the only issue is VRAM to VDP now.

Next update will be a working video of the board but it could be some months until I jump in , fix it up and do a tidy.

**also, **IF** there's any interest in the board, I'd consider remaking it with the cart slot centered as per the standard board as I understand not everyone wants an Everdrive hanging out the side with a cutout for SD cards

**unlike, say ....a NES... where you literally can't stand the pain of leaving your finger on the regulator heatsink for more than a couple seconds because it gets so hot you wonder how it even lasts so damn long - these are fine. The last proto was a 1.5a ST branded 7805 variant as my MIC2940 LDO's hadn't arrived. It doesn't get uncomfortably hot at all, left Alex demoing in the background behind the garble for quite a while- the input/output caps dont cop any heat either.
  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 740
  • Location: Adelaide, Australia
Reply with quote
New Master system 2 mainboard
Post Posted: Fri Oct 16, 2020 5:02 pm
I don't think the SRAM you are using is compatible with the VDP.

The VDP uses multiplexed address and data lines for the VRAM bus. The 4168 XRAM supports multiplexed address and data lines. The SRAM doesn't appear to though.

The only times I have seen SRAM used for VRAM is where there's been an additional chip (the 315-5330) between the VDP and VRAM.

I'm sorry I didn't see this earlier. I missed the posts where you were discussing sourcing the VRAM until just now.

For the BIOS not booting when a cart is inserted, that could be bus contention. I'd start by looking at the BIOS and Cart Slot /CE lines and the BIOS /OE as well.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Oct 29, 2020 4:33 am
Good information
Suspected the type of RAM was the issue as it’s always the XRAM used on every board I’ve seen for VRAM but never for WRAM.

I think this may explain what the 74LS373 is doing between the PPU and SRAM (VRAM) on a NES mainboard. Something similar to this situation anyway.

Way over my head to re-engineer this to suit SRAM. I was fairly confident the way it’s currently wired should be fine thus haven’t made any changes to the BUS.

What’s frustrating is the idea behind this is to make one from scratch, not demolish old boards for donor parts.

The xRam is still out there. It’s very expensive!

There doesn’t appear to be any alternatives

specially made SRAM with multiplexing was apparently made for a short while and then taken off the market due to costs.

There is this one obscure part I’ve come across advertised as SRAM with multiplexed I/O but it’s also advertised as standard SRAM.

srm2064c15

Apparently military grade, cheap and easy to obtain but I can’t for the life of me find a bloody data sheet
  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 740
  • Location: Adelaide, Australia
Reply with quote
New Master system 2 mainboard
Post Posted: Fri Oct 30, 2020 2:17 am
I found the datasheet. Send me your email in the PM and I’ll send it to you. It doesn’t look like it supports multiplexed address and data lines from the timing diagrams though. :(

There is still a small slice of MUXed SRAMs out there still. They’re sometimes referred to as “low pin count SRAM”. ISSI for example does MUXed SRAM targeted for phones and automotive as small as 4Mbit of storage where low pin count is desired. Element14 stock some of these chips so they’re still available.

But they’re nowhere near as available as common parallel access SRAM chip. So a circuit between the VDP and VRAM to demultiplex the address and data lines is to me a good solution.

The way I would do this is to use two 74HC573’s (they’re the same as the 373’s but the pinout is friendlier). You would connect AD0 to AD12 from the VDP to the inputs of the 573’s; connect the outputs of the 573’s to the A0-A12 of the SRAM; connect /CE from the VDP to the LE (Latch Enable) of the 573’s; tie the /OE (Output Enable) of the 573’s low; and leave the data lines and /WE between the SRAM and VDP as is. This circuit will latch the address on the address lines to the SRAM when the VDP drives the /CE line low so that the address will stay stable on the address lines of the SRAM for the duration of the access cycle.

For Proof of Design, rather than refab the entire board, you could save some money and use your existing board, by removing the 2 VRAM chips, solder in some DIP28 sockets, and fab up a small daughter board with the 573’s and SRAM chips on it that will plug directly into the DIP28 sockets. Just a suggestion.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat Oct 31, 2020 9:43 am
That's actually a really good idea and solution!

I'll check the sheet out!
I'd say your assumption would probably be better than mine though if you've already gone over it.

I will definitely make an SRAM adapter some time in the future using this suggestion. For the next next prototype though it won't hurt to remove the original XRAM from the donor board.

The VDP from ali express still isn't here
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Mon Nov 09, 2020 5:33 am
https://i.imgur.com/UjB7RLJ.jpg
UjB7RLJ.jpg (167.91 KB)
UjB7RLJ.jpg

  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 740
  • Location: Adelaide, Australia
Reply with quote
New Master system 2 mainboard
Post Posted: Wed Nov 18, 2020 1:57 am
Any updates on this of late?
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Thu Nov 19, 2020 6:40 am
I’ve gone from early finishes due to the pandemic in my previous role (workload dropped) to emptying massive shipping containers most days of the week. When I’m not doing this it’s other labour hire work.

This is where the spare time came from for this project. My early finishes.

Pretty sure most of the tweaks are already done on the board, bar cartridges not loading. I haven’t quite figured that one out. The final controller input parts arrived just a week or two back and the Main Sega ICs are good to go ... could probably solder them in seated up really high so it’s easy to remove and keep legs intact. The 64pin quad inline sockets are not something you can get anymore.

The next batch of boards would hopefully be fine for everything bar cartridge loading. When I find time I’ll sit down and try and figure it out but ordering boards at this stage would be $ down the drain.
  View user's profile Send private message
  • Joined: 25 Jul 2007
  • Posts: 716
  • Location: Melbourne, Australia
Reply with quote
Post Posted: Thu Nov 19, 2020 8:54 am
L10N37 wrote
https://i.imgur.com/UjB7RLJ.jpg


I've also ordered a couple of VDPs from Aliexpress, would be interested to find out if they are indeed genuine.
  View user's profile Send private message
  • Joined: 14 Aug 2000
  • Posts: 740
  • Location: Adelaide, Australia
Reply with quote
New Master system 2 mainboard
Post Posted: Thu Nov 19, 2020 11:37 pm
It's been a while since I measured it with a ruler, but from memory, I think you can use four 16-pin SIL connectors as your socket for IC5, because the pitch between each pin in the row of 16 pins is the standard 0.1", and the pitch between the 2 rows of 16 pins is also 0.1".
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Wed Nov 25, 2020 6:22 am
Hopefully they work. I was considering the strips you can get that you break off to the pin count but that is even better and available from element 14.

Will find out soon hopefully!

Everything on the cartridge connector seems to be connected as per the schematic. Still haven't gotten around to opening it up and going through it super thoroughly.
  View user's profile Send private message
  • Joined: 17 Dec 2020
  • Posts: 1
Reply with quote
Post Posted: Thu Dec 17, 2020 7:31 am
L10N37 wrote
Hopefully they work. I was considering the strips you can get that you break off to the pin count but that is even better and available from element 14.

Will find out soon hopefully!

Everything on the cartridge connector seems to be connected as per the schematic. Still haven't gotten around to opening it up and going through it super thoroughly.


This is looking awesome, keep up the good work!
  View user's profile Send private message
  • Joined: 24 Aug 2012
  • Posts: 15
  • Location: Italy
Reply with quote
Post Posted: Thu Dec 31, 2020 8:16 pm
Hi,good job this is open source?
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sat May 08, 2021 7:42 am
I may eventually open source it. Hopefully it will be finished by the end of the year, just had to back burn the project. No time to sit in front of my computer in front of PCB software at the moment and really not much left to do with this. Go through an original board with a meter and compare each cartridge pins destinations/ order on the BUS and then transplant the correct RAM off an original board for the VRAM and make that adapter to use regular SRAM in it's place. I may make the adapter first and trial it on my test bench master system which has IC sockets in place of direct soldered WRAM/ VRAM/ CPU/ BIOS.
  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Sun Sep 26, 2021 11:40 am
I can't see myself getting back into this project, as I'm busy with other things constantly.

I shared it a while back, then removed it... but I'd like it be open sourced. It's a bit of a mess, but it should be tidied up and released. It was a fair effort.

Issues:

1: Won't load cartridges

2: Used the wrong VRAM so had garbled graphics - by memory, asynchronous had a good suggestion for being able to add multiplexing to normal SRAM on an adapter board. This would be good.

It's currently designed for an external RGB amp circuit to be wired on the rear, but I have some THS7316's here, and I was going to make a video circuit with them so that there was a version of the board ready to go with RGBs output.

S-video would be nice too!

Because the graphics was garbled, I couldn't really tell what was going on, but I know i could jump on the spot in game - I think pressing one of the direction buttons was resetting the console / restarting alex kidd. I couldn't find any issues on the controller traces

I'm not a fan of the pause mod anyway, I don't really like that IC being integrated onto the board as I always figured you may as well just run an extension button from the reset button up to the controller if you wanted a pause button,.

You'd have to drill a hole, and zip-tie a cable neatly to the controllers cable or something. Maybe just install an 'extension pause' port so the controller splits off into 2 cables on the console end. We have arduinos these days anyway which are simple to program for simple logic stuff like this too.

Anyway, if someone wants to take over/ help fault find and tidy things up so we can get an open source SMS board out there, it would be awesome!
Top.png (182.4 KB)
Current state, top
Top.png
Bottom.png (142.66 KB)
Current state, bottom
Bottom.png
OpenSourceSMS.zip (3.72 MB)
Zipped Kicad Files

  View user's profile Send private message
  • Joined: 27 Jun 2020
  • Posts: 212
Reply with quote
Post Posted: Mon Sep 27, 2021 3:04 am
hmm A14 was bypassing the IO chip. So there's one mistake
It looks like this board may only need some final touch ups, the capacitors were a stupid choice hence need switching out to something easy to solder.

It may load cartridges after all, if only the board was around so I could run a dodge wire. Pretty sure I removed the components... facepalm moments

Also: with the use of Vias you can basically flood fill every inch of the board with ground, including almost every single gap.
A14.png (118.26 KB)
A14
A14.png

  View user's profile Send private message
  • Joined: 14 Jan 2024
  • Posts: 1
Reply with quote
Post Posted: Sun Jan 14, 2024 6:37 am
@L10N37 Are you still floating around these parts?

Sorry for the thread dredge but I have started a project of my own recently and did not know this project existed until now.

With my own project I am wanting to design my own PCB but add my own mods integrated into the PCB and re-using using the Sega's chips.

I have a few questions for you or if others are able to respond:

1. Did you design the footprints for the custom chips yourself or find them online? I downloaded your project but it seems your custom footprint libraries are missing.

2. Did you use the SegaMasterSystemIIServiceManual.pdf document from SMSPower as schematic reference? This is what I am using as a guide but I have found many many inconsistancies between this document and other found online...

Kind Regards,
Jeremy
  View user's profile Send private message
  • Joined: 30 Sep 2023
  • Posts: 16
Reply with quote
Post Posted: Tue Jan 30, 2024 11:43 pm
L10N37 wrote
I can't see myself getting back into this project, as I'm busy with other things constantly.

I shared it a while back, then removed it... but I'd like it be open sourced. It's a bit of a mess, but it should be tidied up and released. It was a fair effort.

Issues:

1: Won't load cartridges

2: Used the wrong VRAM so had garbled graphics - by memory, asynchronous had a good suggestion for being able to add multiplexing to normal SRAM on an adapter board. This would be good.

It's currently designed for an external RGB amp circuit to be wired on the rear, but I have some THS7316's here, and I was going to make a video circuit with them so that there was a version of the board ready to go with RGBs output.

S-video would be nice too!

Because the graphics was garbled, I couldn't really tell what was going on, but I know i could jump on the spot in game - I think pressing one of the direction buttons was resetting the console / restarting alex kidd. I couldn't find any issues on the controller traces

I'm not a fan of the pause mod anyway, I don't really like that IC being integrated onto the board as I always figured you may as well just run an extension button from the reset button up to the controller if you wanted a pause button,.

You'd have to drill a hole, and zip-tie a cable neatly to the controllers cable or something. Maybe just install an 'extension pause' port so the controller splits off into 2 cables on the console end. We have arduinos these days anyway which are simple to program for simple logic stuff like this too.

Anyway, if someone wants to take over/ help fault find and tidy things up so we can get an open source SMS board out there, it would be awesome!


Thank you for all of your work on this. It seems you got so close I hope you can continue and make it work.

I thank you for uploading the zip again also. Perhaps at some point I can learn kicad and contribute to this. I have access to most mastery system versions, including a bunch of the very last ones from Tectoy that might be of interest to this project since they also fit into mostly standard sized sms2 shells as well. some of them use a mix of different through hole and surface mount components and they reduced chip counts removed cart connectors, added built in games, etc over the different revisions.

As to the ultimate system, there are a few things I would recommend if you or anyone else develops this further.

1. The arcadetv rgb bypass is good but most newer mods use the ths7374 amp now. But an option to use original cxa1145 or mb3514 (I like the Fujitsu one) and switch between that and the bypass would be nice to have.

2.You mentioned you don't like the start button mod, but I would recommend adding it with an option to turn it on and off maybe. It is such a lifestyle improvement to do that or the real button. And you can integrate it with the region mods to cycle through console regions. And if you add a second similar circuit where one does up+down and the other does left+right then maybe finally save states will become useful on Everdrives. As it is now, it's too hard to use only 1 dpad and both buttons without accidentally triggering the Everdrive commands.

3. An option for top cart slot would be great. I like the internal Everdrive though. I believe with bocks boot bios, you get a device selection screen. Then you could choose between bios, Everdrive, or cart on boot. What about a small edge connector with option for normal cart, normal expansion pinout, one with mk3 pinout, and one with game gear pinout.

4. FM sound module. Since your internal expansion connector uses a cart slot pinout, you need a way to connect FM sound module integrated with the circuit board. Otherwise, the internal edge connector should match an sms1 edge connector, this way existing FM sound modules can be added in for people who want it. The integrated approach might be nicer though. I have a Japanese SMS but haven't looked inside that to see how the FM works.

5. 3d glasses support. Sms power already has schematics for this and on circuitboard de there is a post with lots of info on it. Japanese sms also has this built in.

6. Switch for rapid fire unit. Option to enable/disable. This was another built in feature for the sms in Japan.

7. Integrated BlueRetro HW2. This can be enabled 2 ways. 1: turn system on and Bluetooth comes on and plugging hardwire controller in disables the BT and removing controller enables it. 2: blue retro always running in low power mode and you can remote power on and off the console with Bluetooth controller.

8. Usb-c option for power. Console is 5v anyway.

9. Option for HDMI output. VGA. Component. Svideo. Rgbs.

10. Allow for neogeo 15pin controller input. Why? Combine it #2 above and it would let you use a standard neogeo controller to use select as left+right, start as up+down (and send to pause button to send pause button), and you can set the rapid fire module for 2 buttons to use as turbo for the other 2 buttons. That lets you use 6 face buttons on a console designed for only 2. And would be compatible with the pause button mod on Bluetooth controller using blueretro or any arcade pad without modifying a controller physically. Essentially the db15 sits in front of the 9pin input to the IOchip and in the middle you put the pause mods and turbo button mods. You could then mount either de9 or db15 controller ports to the console
  View user's profile Send private message
Reply to topic Goto page Previous  1, 2



Back to the top of this page

Back to SMS Power!