Vhd interface for victor hc90/95

Page 11/25
4 | 5 | 6 | 7 | 8 | 9 | 10 | | 12 | 13 | 14 | 15 | 16

By popolon_

Champion (420)

popolon_'s picture

07-10-2022, 18:30

First attempt with replicated pcb, and commands are not working. There comes I/O timeouts for _VP commands.
P_20221007_192539.jpg
P_20221007_192531.jpg

By leonlas

Master (145)

leonlas's picture

19-10-2022, 14:09

You are doing a good job, I am looking forward to the final effect of the interface. And I am ready for it, if you have a PCB for sale, it is a reservation for me.
I have HC95 and managed to get Storm (but no casette)

By popolon_

Champion (420)

popolon_'s picture

04-12-2022, 09:37

Damn, I missed that I had put Intel 8255A( not rated 5mhz), when in original PCB there is Nec 8225-2(rated 5Mhz max). So the speed is not suitable. But my only nec8255-2 ics from AliExpress seems to be faulty or fake. So have to wait for other source now from eBay(nemoclock seller).

By popolon_

Champion (420)

popolon_'s picture

08-12-2022, 20:48

At the mean time i received one NEC D71055C, but when I put that into pcb, HC-95 does not boot (hangs at boot, black screen, but tv gets sync).
I measured the voltage from 8255-pin17(PC3) from working (original pcb) and non-working, working=0v, non-working=4,6v. And the INT01 line voltage was: working=4,6v, non-working=0.
So then i just disconnected the 10k resistor going from 8255-pin17 to Q1 (D637) base pin (collector pin goes to INT01 line), and then the HC-95 will boot with this D71055C. But of course commands are getting I/O timeout. I rechecked all logic ICs with retro chip tester, and those are ok.
The strange thing is that with slower Intel 8255A the HC-95 boots fine even when this resistor is connected, but commands are getting I/O timeouts. So this would mean that the Q1 is not faulty? what could be the problem? Signal routing in pcb?

By popolon_

Champion (420)

popolon_'s picture

09-12-2022, 11:05

Damn it again. I have missed the A6/RESET line going to 80C49 _RESET pin (pin-4). Now the HC-95 boots up and at least _VP("g0000") now does not hang but goes through quickly with OK. Next will test with real VHD player. I have not updated the github yet, as have to also reroute the pcb.
By the way, the 96-pin signal description is not accurate in our wiki page https://www.msx.org/wiki/Victor_96-PIN_Slot_Connector , because it does not say that RESET is actually active low. I mean in 50 pin cartridge page https://www.msx.org/wiki/MSX_Cartridge_slot we have active low signals shown with that macron. But in 96 pin case all are just upper case letters, no macron for the signals that would be active low.

By gdx

Enlighted (6438)

gdx's picture

09-12-2022, 11:20

I fixed some signals but there are probably other signals that are active low.

By popolon_

Champion (420)

popolon_'s picture

09-12-2022, 11:52

Quick test with real VHD player it started playback with command _VP("P") and gets correct frame number with
_VP("G",A): PRINT A
And _VP("E") stopped play and ejected disk.
So the schematics for C9D card seems to be ok now.

By popolon_

Champion (420)

popolon_'s picture

09-12-2022, 12:01

that INT01 line shuold be active low also, as I measured in working case 4,6v when idle. So I edited wiki page to have those INT01/INT02 as active low also.

By gdx

Enlighted (6438)

gdx's picture

09-12-2022, 12:11

This datasheet gives almost all signals:
https://www.datasheetarchive.com/pdf/download.php?id=379e42a...

By popolon_

Champion (420)

popolon_'s picture

09-12-2022, 13:34

So only RAMSEL, BUSCON, JVCREQ and JVCACK are the ones that the active state signal is not known. Or is it also active low? In this VHD card case JVCREQ was used to detect if it is just I/O address (JVCREQ=1), or normal address (JVCREQ=0).

anyway next i will try to fit this card to Victor HC-7 module cartridge port if that would work also there somehow.

Page 11/25
4 | 5 | 6 | 7 | 8 | 9 | 10 | | 12 | 13 | 14 | 15 | 16