Sega Master System / Mark III / Game Gear|
SG-1000 / SC-3000 / SF-7000 / OMV
Home - Forums - Games - Scans - Maps - Cheats
Music - Videos - Development - Translations - Homebrew
SMS Power! 17th Anniversary Competitions are now in progress! Listen, play, vote!
SF-7000 Schematics needed please / Memory Test ROMPosted: Sat Feb 11, 2012 11:04 am
Does anyone have schematics for the SF-7000 super control station? There is an old SMS Power thread about schematics here: http://www.smspower.org/forums/viewtopic.php?t=9813
So if anyone knows MadMikeAU I'd be very appreciative if I can find a way to get in touch and get a copy of those partial schematics. Please please please :)
I have been helping a tech over in Aussie with his SF-7000 Super Control Station repair. He's much more experienced at fixing vintage tech than me and has done it by the numbers without success so far.
He's removed and socketed all the the IC's with quality machined pin sockets. He has a Hi-Lo System IC Programmer / Tester and has also tried alternate chips for everything. He's checked the power supply voltage and the voltage near the ICs. He's even built an ESR meter to try to locate bad caps which showed up a couple that needed replacing. And we *think* his I/O Cassette is working correctly and the cable is probably ok.
So we're at the stage where some schematics would really come in handy so we don't have to spend several hours drawing up our own.
Note - as part of the troubleshooting I wrote a diagnostic plug-in replacement for the IPL that loads up the text mode tiles and runs an onscreen memory test without using any stack space - ie. it all runs in registers. That was entertaining to write, but you can do it by in-lining a lot of code, only calling one or two functions deep, and using the shadow registers for storage. I used Maxim's IPL disassembly as a start point.
That is useful because if you have a simple DRAM failure, then it will pinpoint exactly which DRAMs need replacing pretty quickly so you don't have to swap them all. And for more complex faults it tells you what part of the memory system isn't working and gives you a baseline reading to judge improvements against.
Once I get clear of the SC-3000 Multicart project I'll post the code for the Diagnostic ROM and write a bit more about it and some of the things we've done whilst troubleshooting this SF-7000 and one of my ones too.
But take a look at the attached photo and you will get the idea. This is actually from one of my SF-7000s, not the Aussie one. Mine has a similar fault.
The test starts at $4000 and alternately writes $55 and $AA to RAM then reads it back. Only failures are displayed. When you get a full screen of errors, it asks you to press the space bar to continue. And when you get to $FFFF it says 'finished - press space to start again'. Based on which bits are wrong you can figure out which ICs aren't working.
Here's how the information is useful. Fire up the Calculator on your computer and convert $55 (hex) to binary. You will see that
$55 (hex) = 01010101 binary
$AA = 10101010 binary
That is the initial test pattern used by the SF-7000 IPL stack test, so it seemed like a reasonable starting point. And it is good because together the patterns exercise all the bits.
Take a look at the screenshot. You can see that the first failure is at $4003. I wrote $AA to memory and it read back as $EA, or 11101010. So bit 6 (the 7th most significant bit) is sticking in the on position.
If you follow the traces from the 34-pin SF-7000 I/O cassette connector on the motherboard to the DRAM data inputs, you will see that IC12 handles bit 7 for every byte (the most significant bit) and IC5 handles bit 0 (the least significant bit).
By looking at the other failures, hopefully you can see a consistent pattern which will pinpoint which memory chips to try to replace first. From looking at multiple pages of results I can see that I have problems with both bits 6 and 7 being wrong - both on when they should be off and vice versa.
So in my case, I can focus my hunt on the parts of the memory system relating to those two bits. It isn't the DRAMs themselves as I've tried swapping them. I assume the LS157s on the motherboard are used to break up the RAS / CAS inputs, so one of those could be stuffed. But I haven't had the time to look at the circuit traces in detail to see how it is wired up. Once again, schematics would be great to have here :)
But you can see the value in the memory test for narrowing down the search area.
||Posted: Sat Feb 11, 2012 12:03 pm|
I'm totally irrelevant to help you, but this page may interest you:
Please, forgive me, if this has no connection with what you're looking for...
||Posted: Sat Feb 11, 2012 4:18 pm|
Thanks. No, not quite what I'm looking for, sorry. That page is an excellent resource, but it is really a collection of excerpts from the SF-7000 User's Manual. The User's Manual is a 220 page beast and by today's standards it has a vast amount of information in it. But no schematics, sadly.
The schematics show how all the individual capacitors, resistors, transistors, and integrated circuits of the system are connected together. You can see most of that by looking at the traces on the PCB, but it is a lot easier if you have a high level diagram to look at :)
SF-7000 Fixed - it was the I/O Cassette cablePosted: Mon Feb 13, 2012 5:28 pm
We think there has been some success overnight. He removed the ribbon cable from the I/O Cassette, installed a PC right angle 34 pin plug, and attached a PC floppy cable in place instead. Now he can consistently start the system and even seems to have kicked the disk drive into life. Fingers crossed, but it looks like that was the problem. Success!
The I/O cassette is the critical link between the SC-3000 and the SF-7000. So if the main unit powers up, but you either get just the standard SC-3000 black screen plus 'medium tone' or you get the IPL test beeps then it is a good candidate to check first. All the address lines, data lines, /IOW, /IOR etc. are passed out through it, so a break in any of the wires, or a fault in any of the ICs on the cassette will prevent the system from working. But since the system was kind of responding, we didn't want to touch that cable until we had to.
I'll write up a bit more on the whole troubleshooting process just for reference in case it can help someone else out in future when I get a bit of spare time.
I'd still love to get the main PCB schematics for future reference though :) And many thanks to gallos_11 for the PM with the I/O cassette schematic he found at the bottom of Enri's SC-3000 page. That was helpful.