Another black screen topic

Page 3/3
1 | 2 |

By RetroTechie

Paragon (1414)

RetroTechie's picture

10-12-2016, 05:33

Philip wrote:

Anyway here are the results from the measurements (almost more exciting as hamlet's quiz...)
9 reset high, becomes low when reset is pressed
30 csw high, blinks once shortly after reset.
31 csr high
62 ras blinks
61 cas0 blinks
60 cas1 blinks
57 vram r/w select high

Continuous /RAS activity would be expected, I think (VRAM refresh). /CSR (mostly) inactive is also normal, not much VDP -> CPU data transfers under normal circumstances.
Both /CAS0 and /CAS1 remaining active (= VRAM data access) may suggest that at least some screen mode was initialized, with VDP reading pixel data from both sets of VRAM chips. But I'm not sure what normal behaviour of a V9938 is in this regard. Would have to compare with a known working but ROM-less MSX2.
Like /CSW, pin 57 (vram r/w select) should at least show some activity when it comes time to display the boot logo. Some data must be written to VRAM then, right? Try resetting while monitoring this pin?

Anyway, between VDP and VRAM, bad VRAM chips are more likely than a dead VDP. Although reliability of (single-supply) 4464's isn't too bad compared with some older type DRAMs like the 4116's used with some MSX1 VDP's. And broken V9938's do occur, yes (but the few cases I've seen were result from +12/+5V short circuits or other mishaps like that Wink ).

If anyone knows of an easy/good way to test VRAM or VDP individually (other than desoldering & test in other system): suggestions welcome. This is certainly a situation where some kind of diagnostics ROM would come in handy...

By Philip

Master (204)

Philip's picture

10-12-2016, 12:03

Quote:

Try to check also if the /CS pin of the RP5C0A is blinking.

This one does blink.

RetroTechie wrote:

Like /CSW, pin 57 (vram r/w select) should at least show some activity when it comes time to display the boot logo. Some data must be written to VRAM then, right? Try resetting while monitoring this pin?

vdp pin 57 just stays high, no activity during or after reset (also tried with the mem function of the probe).

RetroTechie wrote:

Anyway, between VDP and VRAM, bad VRAM chips are more likely than a dead VDP. Although reliability of (single-supply) 4464's isn't too bad compared with some older type DRAMs like the 4116's used with some MSX1 VDP's. And broken V9938's do occur, yes (but the few cases I've seen were result from +12/+5V short circuits or other mishaps like that Wink ).

But if the vram is broken, usually there is some corrupted picture ?

I tested all 4 vram's (IC102-105):
1 oe low
4 write stays high
5 ras blinks
9 Vcc high
16 cas blinks
18 gnd low
i/o 1-4 blink
a0-a7 blink

But I guess this doesn't say much, I mean the data and address would still blink even if it forgot a bit here and there.

So my plan is:
- First replace the v9938 for a v9958 as I want to do that anyway.
- If that doesn't work purchase the eeprom programmer and test the hotbit rom (and this would allow me to make a custom (2+) bios)
- If it still doesn't work test the vram somehow...

By Alexey

Paragon (1229)

Alexey's picture

11-12-2016, 13:59

Don't forget to add 100nF capacitors over the resistors connected to 9958 pins 5 and 6 like in this video (skip to 18:45):

https://www.youtube.com/watch?v=dz-oaFYrFdg

This should stabilize the video. Worked well for our Yamaha MSX2+ upgrades.

By Philip

Master (204)

Philip's picture

11-12-2016, 17:52

Thanks, I read something similar here but the video makes it much clearer.
Now I just have to wait for the v9958 to arrive...

By Taburoto

Champion (294)

Taburoto's picture

28-07-2017, 19:29

Hi.
I also have a 8250 with the same symptoms, it does not start and the screen stays black.
The power supply works properly and there is a 50hz video signal but black.

I have checked the following:

RAM
VRAM
ROMS
CPU
Boot with external mapper
Clock (with logic probe)
Reset (with logic probe)

I do not know what else to check, there is no way ...

By Philip

Master (204)

Philip's picture

28-07-2017, 20:28

Well, I did replace the v9938 but that didn't help.
In the end I managed to fix the 8250 by just buying another... In fact things got slightly out of hand and I have now 4 8250's, a 8245 and a hb-75p.
I did order the tlc866 eprom programmer and I still want to try the hotbit rom as the clock chop might be broken.
For now I just used the eprom programmer to burn new eproms for the working machines.

By [D-Tail]

Ascended (8222)

[D-Tail]'s picture

30-07-2017, 09:00

Note that the 8250 has a grey/white-ish switch in the center of the bottom motherboard, iirc it should be in the OFF position for the computer to function properly. If it's in the other position, then indeed the screen stays black. Try switching that and see ifit works for you.

By Taburoto

Champion (294)

Taburoto's picture

24-08-2017, 20:00

Hello, after getting a socket and testing several VDP the MSX2 still does not work properly ......

He tried a boot by putting an MSX1 BIOS (Hotbit from MSX-Pro) and the computer was started, but in the following way:

1-The computer is recognized and executed in the MSX1 BIOS, but displays a defective or ghost image
2-Does not recognize or run the DISK-ROM
3-Does not recognize or execute any cartridge in SLOTS 1 and 2
4-The keyboard works incompletely (You can not be by keyboard keyboard in this MSX1 Hotbit BIOS)

So now you have to look for the fault for another site, the VDP is discarded.

Any ideas?

By tvalenca

Paladin (674)

tvalenca's picture

14-11-2017, 17:04

@Taburoto If your 8250 boots with Hotbit 1.3 BIOS so everything works besides the original ROM chip and the RP5C01 (RTC) chip.

Page 3/3
1 | 2 |
My MSX profile