Hi there folks, I've recently bought a model 1 NTSC-U HKT-3020 Dreamcast console. It features a VA0 mainboard (with heatpipes) which says: Code: SEGA 1998 171-7782C IC BD KATANA MAIN VA0 MADE IN JAPAN 837-13392-04 The GDROM drive says: 3.3v made by Samsung I modded it with a piggy-back bios, pico-psu kit and an internal CF-card reader making it look like this: It's working very well and I've been playing lots of Solder of Fortune without loading times However, DreamShell seems to be a bit buggy (using the latest pre-compiled release from github) when flashing the bios chip. According to the dev, the VA0 boards haven't been fully tested. So I recently decided to swap the mainboard with a VA1 one instead. The bios chip would be fully compatible with DreamShell and it produces less heat as well. I got the new board from eBay which says: Code: SEGA 1999 171-7965E -X3 IC BD KATANA MAIN VA1 MADE IN JAPAN 83713778 I piggybacked it with a flashable bios chip: 29LV160TMC-70. And I've transfered the heatpipes from my VA0: However it doesn't work up when I power it :/ The led on the controller-board doesn't even turn on. I've tried to enable both the original bios and flashable chip on top of it.. but no dice. And according to eBay the boards were tested and working. So my question, could it be that the controller board from the VA0 being not compatible with my VA1 replacement board? Something about a fan-check? This is the controller board: I know the GDROM drive should be compatible as it's a 3.3V one. But even without the drive connected the led still does not turn on.. Any ideas?
I'm looking at my VA1 controller board and it appears yours only has 19 pins while mine has 20 pins, so that may have something to do with it.
Yeah it's the fan. The DC doesn't even try to not of it doesn't detect the fan spinning. You could make a small circuit/pic to emulate the fan signal, but you need to ensure that the fan really is blowing to avoid overheating. I think there's a new way to do it requiring only to solder a wire. Check the picoPSU or replacement fan threads for more info, people are doing that in there IIRC.
Just to make it clear - even if you defeat the fan failure detection the VA0 controller PCB still won't work on a VA1 because the pinout is different.
You don't need to "emulate" the fan signal, some solder is enough to "fake" it: https://bitbuilt.net/forums/index.php?attachments/fake_pwm_hack-png.4562/ (It will even work without cutting the trace, especially if you just want to quickly test it) Sadly I never had a VA0 in my hands.. all i can say from the schematics is that the LED is connected to 3.3V via Pin 18 of the FFC. So it should turn on right away (assuming you have a VA0 sub-board) - otherwise there might be something wrong with your powersupply.)
Thanks a lot guys, I really appreciate all the information. Now I just hope the 19pin VA0 ribbon cable didn't short the 5V and ground to a point where the mainboard or pico psu are fried.. So to sum it up, I'll get my hands on a VA1 controller board and ribbon cable and fake the PWN signal by cutting the trace and connecting both sides of R627 to 3.3v. Then I'll just hook up the fan wires to the controller board pcb fan connector (leaving the signal wire disconnected.
you don't even need to disconnect the signal wire from the fan/DC. Basically the IC next to R627 translates the analog/PWM signal from the fan to 0v/3.3v. So after this workaround, it just doesn't matter that the IC's input is, as it's "output" always is a logical high/true.
If you can't find a 20pin controller board, I probably have one laying around that I could sell for cheap+shipping.
