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 - SF-7000 replica

Reply to topic
Author Message
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
SF-7000 replica
Post Posted: Mon Aug 15, 2022 11:17 am
I resumed a project found on a french site and made a SF-7000 replica, but i get on boot only some weird screens and multiple but randomic beeps.
Since i noticed the original SF-7000 doesn't boot without the original 3" drive attached, i wonder if i missed some setup in the D8255AC ports that make fails the initial ROM check... there is some diag ROM or similar for further test?
sf-7017_1.jpg (129.64 KB)
SF-7000 replica pic 1
sf-7017_1.jpg
sf-7017_2.jpg (144.13 KB)
SF-7000 replica pic 2
sf-7017_2.jpg

  View user's profile Send private message
  • Site Admin
  • Joined: 19 Oct 1999
  • Posts: 14225
  • Location: London
Reply with quote
Post Posted: Mon Aug 15, 2022 12:30 pm
Did you refer to https://github.com/maxim-zhao/smsbioses/tree/master/SF-7000%20IPL for the IPL disassembly? Maybe that would help? Unfortunately I guess debugging into it will be difficult.
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Mon Aug 15, 2022 1:58 pm
Thanks, i red the disassembly but it not seems useful for my case, the diag check ROM, RAM and VRAM.
I probably have an issue with /ROMSEL or the FDC...
Anyone has changed the 3" drive with a 3,5" drive in the SF7000, and can confirm that it doesn't boot if the 3" cable is not connected, as i stated with mine when using the Gotek?
  View user's profile Send private message
  • Joined: 07 Jun 2010
  • Posts: 183
  • Location: Greece
Reply with quote
Post Posted: Mon Aug 15, 2022 10:59 pm
aotta wrote
Anyone has changed the 3" drive with a 3,5" drive in the SF7000, and can confirm that it doesn't boot if the 3" cable is not connected, as i stated with mine when using the Gotek?

Here are some instructions for replacing the drive with a 3,5" one:
http://members.optusnet.com.au/spacetaxi64/MAIN/SEGA-SF7000.htm

Although I haven't tried it.
  View user's profile Send private message
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Tue Aug 16, 2022 5:17 am
In that guide (the only I found in the net too) is stated it's possible to completely remove the original floppy, but it's not true from my test (with gotek, not yet tried with a pc floppy drive).
I guess some of the PPI signals (/motor, /inuse, etc.) must be pulled low for initializing the FDD
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sat Oct 01, 2022 7:10 am
aotta wrote
In that guide (the only I found in the net too) is stated it's possible to completely remove the original floppy, but it's not true from my test (with gotek, not yet tried with a pc floppy drive).
I guess some of the PPI signals (/motor, /inuse, etc.) must be pulled low for initializing the FDD


If you need support, ask me.
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sat Oct 01, 2022 7:50 am
siriokds wrote
aotta wrote
In that guide (the only I found in the net too) is stated it's possible to completely remove the original floppy, but it's not true from my test (with gotek, not yet tried with a pc floppy drive).
I guess some of the PPI signals (/motor, /inuse, etc.) must be pulled low for initializing the FDD


If you need support, ask me.


Thank you, but i've already made great progress, thank to the help of @kamillebidan and the full schematics that @Bock published!
I have to fix the rs232 part, and some stability issue, but i'm near the goal!
SF7017_3.jpg (426.97 KB)
SF7017_3.jpg
SF7017_6.jpg (286.5 KB)
SF7017_6.jpg

  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 02, 2022 4:00 am
aotta wrote
siriokds wrote
aotta wrote
In that guide (the only I found in the net too) is stated it's possible to completely remove the original floppy, but it's not true from my test (with gotek, not yet tried with a pc floppy drive).
I guess some of the PPI signals (/motor, /inuse, etc.) must be pulled low for initializing the FDD


If you need support, ask me.


Thank you, but i've already made great progress, thank to the help of @kamillebidan and the full schematics that @Bock published!
I have to fix the rs232 part, and some stability issue, but i'm near the goal!


Nice to read it!
  View user's profile Send private message Visit poster's website
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 02, 2022 11:57 am
aotta wrote
siriokds wrote
aotta wrote
In that guide (the only I found in the net too) is stated it's possible to completely remove the original floppy, but it's not true from my test (with gotek, not yet tried with a pc floppy drive).
I guess some of the PPI signals (/motor, /inuse, etc.) must be pulled low for initializing the FDD


If you need support, ask me.


Thank you, but i've already made great progress, thank to the help of @kamillebidan and the full schematics that @Bock published!
I have to fix the rs232 part, and some stability issue, but i'm near the goal!


Hi, how do you fix the missing floppy issue? For ROM/RAM selection, have you used the 74LS32?
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sun Oct 02, 2022 12:49 pm
siriokds wrote

...

Hi, how do you fix the missing floppy issue? For ROM/RAM selection, have you used the 74LS32?


Yes, the floppy issue was fixed... the SF-7000 Ground is connected via the external wire to the case, the case is connected to the floppy via a screw, and the floppy drive give the gnd to the sf-7000 motherboard via the 34 pin cable.
So, if floppy cable is disconnected, the SF-7000 doesn't boot.
I've added a wire from case to motherboad, and it works!

For ROM/RAM selection, i use several logic IC, included a 74ls32
Here's the last schematic, it's not the final version but i hope i'm not far from it
SF-7017_FD_ok.pdf (383.89 KB)

  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 02, 2022 1:54 pm
aotta wrote
siriokds wrote

...

Hi, how do you fix the missing floppy issue? For ROM/RAM selection, have you used the 74LS32?


Yes, the floppy issue was fixed... the SF-7000 Ground is connected via the external wire to the case, the case is connected to the floppy via a screw, and the floppy drive give the gnd to the sf-7000 motherboard via the 34 pin cable.
So, if floppy cable is disconnected, the SF-7000 doesn't boot.
I've added a wire from case to motherboad, and it works!

For ROM/RAM selection, i use several logic IC, included a 74ls32
Here's the last schematic, it's not the final version but i hope i'm not far from it



Nice, great to share your schematics. I did a cutoff version of it. Replaced memory selection and partial I/O logic with a PLD to reduce the number of components. I just removed serial and parallel IC/connectors inserting a CH376 USB interface which I will support instead parallel. Everything is still WIP.

  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sun Oct 02, 2022 4:47 pm
siriokds wrote

...

Nice, great to share your schematics. I did a cutoff version of it. Replaced memory selection and partial I/O logic with a PLD to reduce the number of components. I just removed serial and parallel IC/connectors inserting a CH376 USB interface which I will support instead parallel. Everything is still WIP.


Great idea the pld to replace the original logic... and timing the RAM and ROM is really a pain running on the original schematics.
Does it works fine, did you already made some tests?
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 02, 2022 7:34 pm
aotta wrote
siriokds wrote

...

Nice, great to share your schematics. I did a cutoff version of it. Replaced memory selection and partial I/O logic with a PLD to reduce the number of components. I just removed serial and parallel IC/connectors inserting a CH376 USB interface which I will support instead parallel. Everything is still WIP.


Great idea the pld to replace the original logic... and timing the RAM and ROM is really a pain running on the original schematics.
Does it works fine, did you already made some tests?


Yes, I did some test but I'd like to upgrade it to a real CPLD. As far as I know SC-3000 provides /MEMRD, /MEMWR, /IORD and /IOWR already merged with /MREQ, you can check it with a logic analyzer so, if you want to use it onnly on SC-3000, it not required to obtain /MEMR1 and /MEMW1. I don't know why it's present on SF-7000 original schematics.
Second point: new EEPROM and SRAM (the IC I used on it at least) can accept /OE before /CE so probably you can avoid the latency added with 74LS04.
My goal is to reduce as far as I can. When I will have a stable version I will share it. Using a 5V CPLD (like my EPM7064 can host at least other 3 IC). 74LS14 and 74LS06 can be necessary but I'm not sure. I handled a floppy drive with a simple Arduino as well as other microcontroller without any other chips. If you use a bigger EPM7128 probably you can put the 8255 logic used by SF-7000 into it. I will do other tests with HCT version of that IC instead the LS ones (taking in account that they can be faster).
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sun Oct 02, 2022 7:47 pm
@kamillebidan (who is giving me a big help in adjusting my board) already made a "sf-7000 on a cart" using plcd and new hw, did you miss this post?
https://www.smspower.org/forums/18607-SF7000OnACartridge
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Mon Oct 03, 2022 5:50 am
aotta wrote
@kamillebidan (who is giving me a big help in adjusting my board) already made a "sf-7000 on a cart" using plcd and new hw, did you miss this post?
https://www.smspower.org/forums/18607-SF7000OnACartridge


Usually I do projects by myself for learning purpose. I can put everything in FPGA as well, or on an ARM like MCU, but it not my goal. Using oldskool 74LSxx IC is quite fun to understand how the Sega engineers did their job.

Have you tested the serial port btw?
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Mon Oct 03, 2022 5:16 pm
siriokds wrote

Using oldskool 74LSxx IC is quite fun to understand how the Sega engineers did their job.

That what i think too! ;)

siriokds wrote

Have you tested the serial port btw?


It doesn't work, and no tx nor rd leds light. But i've made no further test, i'm waiting for a new pcb with less "floating wires" and (i hope) more stable system.
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Tue Oct 04, 2022 4:53 pm
aotta wrote
siriokds wrote

Using oldskool 74LSxx IC is quite fun to understand how the Sega engineers did their job.

That what i think too! ;)

siriokds wrote

Have you tested the serial port btw?


It doesn't work, and no tx nor rd leds light. But i've made no further test, i'm waiting for a new pcb with less "floating wires" and (i hope) more stable system.


Do you have an oscilloscope to test what's happening on the serial?
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Thu Oct 06, 2022 12:48 pm
siriokds wrote

Do you have an oscilloscope to test what's happening on the serial?


Yes, without a scope and a logic analyzer I would not have made it this so far!
  View user's profile Send private message
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sat Oct 15, 2022 6:45 pm
Found some bugs and now i have a stable station, and serial works fine too, even at the "new" 19200 baud speed.
Waiting for new pcb batch for testing the new schematic that i hope will be the final version
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 16, 2022 6:57 am
aotta wrote
Found some bugs and now i have a stable station, and serial works fine too, even at the "new" 19200 baud speed.
Waiting for new pcb batch for testing the new schematic that i hope will be the final version


19200 is very good! ☺️
  View user's profile Send private message Visit poster's website
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Fri Oct 21, 2022 4:14 am
aotta wrote
Found some bugs and now i have a stable station, and serial works fine too, even at the "new" 19200 baud speed.
Waiting for new pcb batch for testing the new schematic that i hope will be the final version


Hi, have you checked if the board layout you build, fit into the original case? I'm trying to make precise measurements to rebuild one.

I'm doing servicing due to "beep" errors. Strangely I have 2 different errors on 2 different I/O carts: 1 beep on the original one (probably cable issue since I checked the UV EPROM), 2 beeps on my home made cart (dram isse?). Checking with oscilloscope I saw 4 of 8 dram chips (4164) with a a suspect waveform...
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Fri Oct 21, 2022 7:32 am
siriokds wrote


Hi, have you checked if the board layout you build, fit into the original case? I'm trying to make precise measurements to rebuild one.

Hi, the pcb i designed has a similar form factor but it's smaller than the original one, and the rear connector and screw's holes are different.
I've only one "filled" original SF-7000 case, so i'll place the new board in a black metal case 28cmx20cm.
To make a perfect replacement board, it should be resized the pcb, repositioned the holes, and changed the power supply (now i use a 5v barrel input and a switch on the board, instead of the connector to get the 5v from the original PSU).

siriokds wrote

I'm doing servicing due to "beep" errors. Strangely I have 2 different errors on 2 different I/O carts: 1 beep on the original one (probably cable issue since I checked the UV EPROM), 2 beeps on my home made cart (dram isse?). Checking with oscilloscope I saw 4 of 8 dram chips (4164) with a a suspect waveform...


I have no original cart but only 2 home made ones, that i use with no issue with both the SF-7000 and its clone.
In all my tests, i found that the one or two beeps not always was related to a bad RAM or ROM: cables too long, missing or poor common ground connection and (in my case the more frequent) wrong timing or error in the RAM and ROM CE and OE signals, gave the one or two beeps (somethimes the 3 beeps error too!).
I can suggest you to make some test using the ROM that Fabio made for me:
https://github.com/fabiodl/sftest
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Fri Oct 21, 2022 8:29 am
aotta wrote
siriokds wrote


Hi, have you checked if the board layout you build, fit into the original case? I'm trying to make precise measurements to rebuild one.

Hi, the pcb i designed has a similar form factor but it's smaller than the original one, and the rear connector and screw's holes are different.
I've only one "filled" original SF-7000 case, so i'll place the new board in a black metal case 28cmx20cm.
To make a perfect replacement board, it should be resized the pcb, repositioned the holes, and changed the power supply (now i use a 5v barrel input and a switch on the board, instead of the connector to get the 5v from the original PSU).

siriokds wrote

I'm doing servicing due to "beep" errors. Strangely I have 2 different errors on 2 different I/O carts: 1 beep on the original one (probably cable issue since I checked the UV EPROM), 2 beeps on my home made cart (dram isse?). Checking with oscilloscope I saw 4 of 8 dram chips (4164) with a a suspect waveform...


I have no original cart but only 2 home made ones, that i use with no issue with both the SF-7000 and its clone.
In all my tests, i found that the one or two beeps not always was related to a bad RAM or ROM: cables too long, missing or poor common ground connection and (in my case the more frequent) wrong timing or error in the RAM and ROM CE and OE signals, gave the one or two beeps (somethimes the 3 beeps error too!).
I can suggest you to make some test using the ROM that Fabio made for me:
https://github.com/fabiodl/sftest


I will try to replace the board of my SF as well as make a very tiny board with only 3,5 floppy drive in it.

With my homemade I/O cart, with a very short cable, I get 1 beep, but the I checked the EPROM and it's ok. With a cart built by myself I get 2 beep error.

What do you think about envelope the cable with the GND?

I will try that tool.

Thanks.
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Fri Oct 21, 2022 11:16 am
My carts works with cables long 30cm, with no shielding, so i don't think it will resolve your issues.
But it's not clear if your SF-7000 stopped working or never boot up...
i started using an 8k game as ROM for testing the ROM enabling / logic was ok.. give a try with Tennis or Guzzler, it worked for me (and was funny use the whole 7000 hw only to run a rom game;) !
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sat Oct 22, 2022 6:36 am
aotta wrote
My carts works with cables long 30cm, with no shielding, so i don't think it will resolve your issues.
But it's not clear if your SF-7000 stopped working or never boot up...
i started using an 8k game as ROM for testing the ROM enabling / logic was ok.. give a try with Tennis or Guzzler, it worked for me (and was funny use the whole 7000 hw only to run a rom game;) !


I found 4 out of 8 dram chips with a suspicious waveform on the "DOUT" pin. I bet my cable is short enough 😅

  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sat Oct 22, 2022 10:24 am
siriokds wrote

I found 4 out of 8 dram chips with a suspicious waveform on the "DOUT" pin. I bet my cable is short enough 😅

Without the floppy, you have no ground in the mainhoard since it's not connected to the case shield

Edited: i saw only now you grounded the floppy to the pc with the white wire.. sorry
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Sun Oct 23, 2022 8:35 am
aotta wrote
siriokds wrote

I found 4 out of 8 dram chips with a suspicious waveform on the "DOUT" pin. I bet my cable is short enough 😅

Without the floppy, you have no ground in the mainhoard since it's not connected to the case shield

Edited: i saw only now you grounded the floppy to the pc with the white wire.. sorry


Yes, I remember the post when you wrote that your system won't boot up if drive is disconnected.
I checked the PCB for GND connection and the floppy is the only part which close the GND connection, so I built a terminal connector.

I found a strange behavior (bottom image) on DOUT signal of 4 of 8 DRAM chips (on top the almost clean signal for - probably - good chips).

  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sun Oct 23, 2022 3:50 pm
siriokds wrote

I found a strange behavior (bottom image) on DOUT signal of 4 of 8 DRAM chips (on top the almost clean signal for - probably - good chips).

I'm not an expert, but in the past i get no DOUT waving signal at all in bad DRAM chips.
If you can burn an eprom, try the attached ram test replacing the IPL rom with it.
It show if ROM and RAM are OK or KO.
SF7000-testram.zip (1.58 KB)

  View user's profile Send private message
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sun Oct 23, 2022 5:08 pm
With a metal project box i bought from china and some 3d printed panels, i made a new case for my SF-7017:
SF7017_case_rear.jpg (183.34 KB)
SF7017_case_rear.jpg
SF7017_case.jpg (186.24 KB)
SF7017_case.jpg

  View user's profile Send private message
  • Joined: 25 Feb 2013
  • Posts: 353
  • Location: Osaka
Reply with quote
Post Posted: Mon Oct 24, 2022 11:46 pm
aotta wrote
With a metal project box i bought from china and some 3d printed panels, i made a new case for my SF-7017:

This is gorgeous!
  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Wed Oct 26, 2022 1:30 pm
aotta wrote
With a metal project box i bought from china and some 3d printed panels, i made a new case for my SF-7017:


Awesome!
  View user's profile Send private message Visit poster's website
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Wed Oct 26, 2022 4:11 pm
aotta wrote
siriokds wrote

I found a strange behavior (bottom image) on DOUT signal of 4 of 8 DRAM chips (on top the almost clean signal for - probably - good chips).

I'm not an expert, but in the past i get no DOUT waving signal at all in bad DRAM chips.
If you can burn an eprom, try the attached ram test replacing the IPL rom with it.
It show if ROM and RAM are OK or KO.


How do I interpret it?

  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Wed Oct 26, 2022 4:51 pm
siriokds wrote


How do I interpret it?


It seems you get some write error at showed locations.
The right screen you should get is the attached (i got it with emulator):
Immagine.png (19.18 KB)
Immagine.png

  View user's profile Send private message
  • Joined: 10 Dec 2009
  • Posts: 39
Reply with quote
Post Posted: Thu Oct 27, 2022 2:38 am
aotta wrote
siriokds wrote


How do I interpret it?


It seems you get some write error at showed locations.
The right screen you should get is the attached (i got it with emulator):


I hope I fixed it. The problem was broken LS chips on the I/O cart. I replaced all of them with the 74HCT version. Your test worked as well as you indicated.

  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Tue Nov 01, 2022 9:52 am
For anyone's interested, the Kicad projects of both SF-7017 replica board and SC-7326 I/O Cassette cart are published on github:
https://github.com/fabiodl/sf7000
  View user's profile Send private message
  • Joined: 19 Apr 2019
  • Posts: 12
  • Location: Canada
Reply with quote
Post Posted: Fri Nov 18, 2022 10:40 pm
aotta wrote
For anyone's interested, the Kicad projects of both SF-7017 replica board and SC-7326 I/O Cassette cart are published on github:
https://github.com/fabiodl/sf7000


Thanks to both of you for doing the hard work of publishing this!

I will be sure to build one, since Fabio asked me to make sure that my SG-1000 clone had the proper signals on the cart slot to decode the SF-7000's IO ports.

(If you want me to send you a bare PCB of the clone please let me know, also! It would be super impressive to have an all-new system from SG to SF, although I still have to build an SK-1100 keyboard clone.)
  View user's profile Send private message Visit poster's website
  • Joined: 25 Jul 2022
  • Posts: 23
Reply with quote
Post Posted: Sat Nov 19, 2022 9:31 am
leadedsolder wrote

(If you want me to send you a bare PCB of the clone please let me know, also! It would be super impressive to have an all-new system from SG to SF, although I still have to build an SK-1100 keyboard clone.)

I have a couple of bare pcb left, but internetional shipping from Italy are quite expensive, i suggest you to orderd them directly from china (i spent less than 20€ for five pcbs, + 13€ S.S.). Feel free to PM me if you need more info in ordering.
  View user's profile Send private message
Reply to topic



Back to the top of this page

Back to SMS Power!